The Handbook of Behavior Change (Cambridge Handbooks in Psychology) 1108496393, 9781108496391

Table of contents :
Contents
Figures
Tables
Sidebars
Contributors
1 Changing Behavior: A Theory- and Evidence-Based Approach
Part I Theory and Behavior Change
2 Changing Behavior Using the Theory of Planned Behavior
3 Changing Behavior Using Social Cognitive Theory
4 Changing Behavior Using the Health Belief Model and Protection Motivation Theory
5 Changing Behavior Using the Common-Sense Model of Self-Regulation
6 Changing Behavior Using the Model of Action Phases
7 Changing Behavior Using the Health Action Process Approach
8 Changing Behavior Using Self- Determination Theory
9 Changing Behavior Using Control Theory
10 Changing Behavior Using the Transtheoretical Model
11 Changing Behavior Using Integrative Self-Control Theory
12 Changing Behavior Using the Reflective-Impulsive Model
13 Changing Behavior Using Habit Theory
14 Changing Behavior by Changing Environments
15 Changing Behavior Using Integrated Theories
16 Changing Behavior Using Social Identity Processes
17 Changing Behavior Using Ecological Models
18 Changing Behavior Using Theories at the Interpersonal, Organizational, Community, and Societal Levels
Part II Methods and Processes of Behavior Change: Intervention Development, Application, and Translation
19 Design, Implementation, and Evaluation of Behavior Change Interventions: A Ten-Task Guide
20 Moving from Theoretical Principles to Intervention Strategies: Applying the Experimental Medicine Approach
21 Developing Behavior Change Interventions
22 Evaluation of Behavior Change Interventions
23 Implementation Science and Translation in Behavior Change
24 Engagement of Stakeholders in the Design, Evaluation, and Implementation of Complex Interventions
25 Maximizing User Engagement with Behavior Change Interventions
26 Cost-Effectiveness Evaluations of Behavior Change Interventions
27 Addressing Underserved Populations and Disparities in Behavior Change
28 Behavior Change in Community Contexts
29 Changing Behavior in the Digital Age
30 Critical and Qualitative Approaches to Behavior Change
Part III Behavior Change Interventions: Practical Guides to Behavior Change
31 Attitudes and Persuasive Communication Interventions
32 Self-Efficacy Interventions
33 Imagery, Visualization, and Mental Simulation Interventions
34 Affect-Based Interventions
35 Autonomy-Supportive Interventions
36 Incentive-Based Interventions
37 Monitoring Interventions
38 Goal Setting Interventions
39 Planning and Implementation Intention Interventions
40 Self-Control Interventions
41 Habit Interventions
42 Economic and Behavioral Economic Approaches to Behavior Change
43 Dyadic Behavior Change Interventions
44 Social Identity Interventions
45 Motivational Interviewing Interventions
46The Science of Behavior Change: The Road Ahead
Index

Citation preview

The Handbook of Behavior Change Social problems in many domains, including health, education, social relationships, and the workplace, have their origins in human behavior. The documented links between behavior and social problems have compelled governments and organizations to prioritize and mobilize efforts to develop effective, evidence-based means to promote adaptive behavior change. In recognition of this impetus, the Handbook of Behavior Change provides comprehensive coverage of contemporary theory, research, and practice on behavior change. It summarizes current evidence-based approaches to behavior change in chapters authored by leading theorists, researchers, and practitioners from multiple disciplines, including psychology, sociology, behavioral science, economics, philosophy, and implementation science. It is the go-to resource for researchers, students, practitioners, and policy makers looking for current knowledge on behavior change and guidance on how to develop effective interventions to change behavior. m ar t i n s. h ag g e r is Professor of Health Psychology at the University of California, Merced and Finland Distinguished Professor in the Faculty of Sport and Health Sciences at the University of Jyväskylä. lin d a d . c a me r on is Professor of Health Psychology and Department Chair of Psychological Sciences at the University of California, Merced. ky r a h a m i lt o n is Associate Professor in the School of Applied Psychology at Griffith University. ne l l i h a n k on e n is Associate Professor of Social Psychology in the Faculty of Social Sciences at the University of Helsinki. tar u lin t u ne n is Professor of Sport and Exercise Psychology in the Faculty of Sport and Health Sciences at the University of Jyväskylä.

The Handbook of Behavior Change

Edited by

Martin S. Hagger University of California, Merced

Linda D. Cameron University of California, Merced

Kyra Hamilton Griffith University

Nelli Hankonen University of Helsinki

Taru Lintunen University of Jyväskylä

University Printing House, Cambridge CB2 8BS, United Kingdom One Liberty Plaza, 20th Floor, New York, NY 10006, USA 477 Williamstown Road, Port Melbourne, VIC 3207, Australia 314–321, 3rd Floor, Plot 3, Splendor Forum, Jasola District Centre, New Delhi – 110025, India 79 Anson Road, #06–04/06, Singapore 079906 Cambridge University Press is part of the University of Cambridge. It furthers the University’s mission by disseminating knowledge in the pursuit of education, learning, and research at the highest international levels of excellence. www.cambridge.org Information on this title: www.cambridge.org/9781108496391 DOI: 10.1017/9781108677318 © Cambridge University Press 2020 This publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2020 Printed in the United Kingdom by TJ International Ltd, Padstow Cornwall A catalogue record for this publication is available from the British Library. Library of Congress Cataloging-in-Publication Data Names: Hagger, Martin S., author. | Cameron, Linda D. (Linda Diane), 1960– author. | Hamilton, Kyra, author. | Hankonen, Nelli, author. | Lintunen, Taru, author. Title: The handbook of behavior change / edited by Martin S. Hagger, Linda D. Cameron, Kyra Hamilton, Nelli Hankonen, Taru Lintunen. Other titles: Cambridge handbooks in psychology. Description: Cambridge, United Kingdom ; New York, NY : Cambridge University Press, 2020. | Series: Cambridge handbooks in psychology | Includes bibliographical references and index. Identifiers: LCCN 2020002958 (print) | LCCN 2020002959 (ebook) | ISBN 9781108496391 (hardback) | ISBN 9781108733670 (paperback) | ISBN 9781108677318 (ebook) Subjects: MESH: Behavior Therapy | Behavior | Motivation | Biobehavioral Sciences Classification: LCC RC489.B4 (print) | LCC RC489.B4 (ebook) | NLM WM 425 | DDC 616.89/142–dc23 LC record available at https://lccn.loc.gov/2020002958 LC ebook record available at https://lccn.loc.gov/2020002959 ISBN 978-1-108-49639-1 Hardback ISBN 978-1-108-73367-0 Paperback Additional resources for this publication at www.cambridge.org/hagger Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

For Nikos This work is dedicated to all those we lost during its creation

Contents

List of Figures List of Tables List of Sidebars List of Contributors 1 Changing Behavior: A Theory- and Evidence-Based Approach m a r t i n s . h a g ge r , l i n da d . c a m e r o n, k yr a ha m i l t o n , n el li h a n ko n e n, a nd t ar u li n t u ne n

Part I Theory and Behavior Change

page x xii xiv xvii 1

15

2 Changing Behavior Using the Theory of Planned Behavior i c ek a j z en a nd p e t er sc h m i dt

17

3 Changing Behavior Using Social Cognitive Theory a le k s a n dr a lu s z c z yn s ka a n d r a lf s ch w a r z er

32

4 Changing Behavior Using the Health Belief Model and Protection Motivation Theory s h e i n a o r be ll , h i n a za h i d , a n d ca r ol i n e j . h en d e rs on

46

5 Changing Behavior Using the Common-Sense Model of Self-Regulation 60 l i nd a d. c am e r o n , s a r a fl e s z a r - p a vl o v i c´ , an d t en i e k h a ch i k i a n 6 Changing Behavior Using the Model of Action Phases lu c a s k el le r , p et er m . g o ll w i t z er , a n d p a s c h al s h e er a n

77

7 Changing Behavior Using the Health Action Process Approach ralf schwarzer and kyra hamilton

89

8 Changing Behavior Using Self-Determination Theory m a r t i n s . h a g ge r , n e l l i ha n k o ne n , n i k o s l . d . ch at zis a ran ti s, a n d ric h ar d m . r yan

104

9 Changing Behavior Using Control Theory w a rr e n m a n s e ll

120

10 Changing Behavior Using the Transtheoretical Model c a r l o c . d i c l e m e n t e a n d m e a g an m . g r ay d o n

136

Contents

vii

11 Changing Behavior Using Integrative Self-Control Theory w i lh e lm h o f m a n n , s i mo n e d o h le , a n d ka t h i di e l

150

12 Changing Behavior Using the Reflective-Impulsive Model ro l an d d eu t s c h a n d f ri tz st ra c k

164

13 Changing Behavior Using Habit Theory s h ei na o rb e ll an d ba s v e rp l an k en

178

14 Changing Behavior by Changing Environments th e re s a m . m a r te a u, pa u l c . f le t ch e r, ga r et h j . h o l la n d s , an d m a r c u s r. m u na f o`

193

15 Changing Behavior Using Integrated Theories m ar t i n s . h a g ge r a n d ky r a h a m i l t o n

208

16 Changing Behavior Using Social Identity Processes ka t h er i n e j. re y n ol d s , ny l a r . br an s co m b e, em i n a s u b asˇ i c´ , an d lo r en wil li s

225

17 Changing Behavior Using Ecological Models j o s a l m o n , k y l i e d . h e s k e t h , l a ur e n a r un d e l l , k a t h e r i ne l . do w n i n g , a n d s t u a r t j. h . b i d dl e

237

18 Changing Behavior Using Theories at the Interpersonal, Organizational, Community, and Societal Levels rob er t a. c. ru it er , r ik cr u tze n, eve lyn e d e le e uw, an d g er jo ko k

Part II Methods and Processes of Behavior Change: Intervention Development, Application, and Translation

251

267

19 Design, Implementation, and Evaluation of Behavior Change Interventions: A Ten-Task Guide ch a rl e s a br a ha m a n d s ar a h d e n f o rd

269

20 Moving from Theoretical Principles to Intervention Strategies: Applying the Experimental Medicine Approach al e xa n d e r j. ro t hm a n , w i l l i a m m . p . kl e i n, a n d p as c h a l s h ee r an

285

21 Developing Behavior Change Interventions ne l li h a n ko n en a n d w e n d y h a rd e m an

300

22 Evaluation of Behavior Change Interventions l y n s a y m at t he w s a n d s h a r o n a. s i m p s o n

318

viii

Contents

23 Implementation Science and Translation in Behavior Change a le k s a n dr a lu s z c z yn s ka , ka r ol i n a lo b cz o w s k a , a nd k a r o l i na h o r o d y s k a 24 Engagement of Stakeholders in the Design, Evaluation, and Implementation of Complex Interventions joanna l. hudson, zoe moon, lyndsay d. hughes, and rona moss-morris

333

349

25 Maximizing User Engagement with Behavior Change Interventions lu c y y a r dl e y, le a n ne m o rr i s o n , i n g r i d m u l le r, and kath e rin e br adb u ry

361

26 Cost-Effectiveness Evaluations of Behavior Change Interventions ti an j i a o w a n g, m ar t i n d o w n es , j os h ua b yr n es , a nd pa u l s c uf f h a m

372

27 Addressing Underserved Populations and Disparities in Behavior Change be n jamin sc hu¨ z a nd mo ni ca we bb ho o pe r

385

28 Behavior Change in Community Contexts ed i s on t ri c k et t an d su s a n p a t er s o n

401

29 Changing Behavior in the Digital Age d av i d j. ka v a na g h

416

30 Critical and Qualitative Approaches to Behavior Change k e r r y c ha m b e r l ai n a n d a n t o n i a l y o n s

430

Part III Behavior Change Interventions: Practical Guides to Behavior Change

443

31 Attitudes and Persuasive Communication Interventions k yr a ha m i l t o n a nd b l a i r t . j o hn s on

445

32 Self-Efficacy Interventions l i s a m . w a r n e r a n d da v i d p. f r e n c h

461

33 Imagery, Visualization, and Mental Simulation Interventions martin s. hagger and dominic conroy

479

34 Affect-Based Interventions m a r k c o n n e r , d a v i d m . w i l l i a m s , a n d r y an e . r ho d e s

495

35 Autonomy-Supportive Interventions j o h nm a r s h a ll re e ve a nd s u n g hy e o n c h eo n

510

36 Incentive-Based Interventions u ri g n ee z y, a gn e k aj ac k ai te , a n d s te p h a n m ei e r

523

Contents

ix

37 Monitoring Interventions th omas l. we bb an d m arij n d e b r ui n

537

38 Goal Setting Interventions t r a c y ep t on a nd c h r i s t o p h e r j . a r m i t a g e

554

39 Planning and Implementation Intention Interventions ry a n e . r h o de s , st i n a g r a nt , a n d ge r t- j a n d e b r ui j n

572

40 Self-Control Interventions de n i s e d e r i d de r , m a r l e e n g i l l e b a a r t , a n d m a l t e f r i e s e

586

41 Habit Interventions b e n j a m i n g a r dn e r , am a n d a l . r e ba r , a n d p h i ll i p p a l a ll y

599

42 Economic and Behavioral Economic Approaches to Behavior Change p a u l m . br ow n, li nd a d . c am e r on , ma r ti n w i l k i n s o n , an d d en is e t ay l o r

617

43 Dyadic Behavior Change Interventions ur t e s ch o lz, cor ina b er li , jan in a l u¨ s c h er , a n d n i n a kn o l l

632

44 Social Identity Interventions m ar k ta r ra n t, ca t h er i n e h a s l am , m a ry ca r te r , r a f f c a l i t r i , a n d s. al e xa n d e r ha s l a m

649

45 Motivational Interviewing Interventions an n e h . b e r m a n , m a r i a b e c k m a n , a n d h e l e n a l i n d qv i s t

661

46 The Science of Behavior Change: The Road Ahead m ar t i n s . h a g ge r , l i nd a d . c am e r on , ky r a h a m i l to n , ne l li h a n ko n en , an d ta r u li n t un e n

677

Index

700

Figures

2.1 3.1 4.1 4.2 5.1 6.1 6.2 7.1 8.1 8.2

9.1 9.2 10.1 11.1 12.1 13.1 13.2 14.1 15.1 15.2 15.3 15.4 17.1 17.2 18.1

Theory of planned behavior page 19 An illustration of social cognitive theory (Bandura, 2000a) 33 Schematic representation of the health belief model 47 Schematic representation of protection motivation theory 49 The common-sense model of self-regulation 61 The model of action phases 79 Meta-analysis of meta-analyses of implementation intention effects on goal achievement 84 The health action process approach (HAPA; Schwarzer 1992, 2008) 95 Diagram summarizing three key mini-theories of self-determination theory: Cognitive evaluation theory, organismic integration theory, and basic needs theory 106 Processes in self-determination theory (a) self-determination theory process model; (b) path model showing effects of a teacher-delivered autonomy-support intervention on school children’s leisure-time physical activity mediated by autonomous motivation and intention 109 The feedback control unit 122 An example of a control hierarchy for smoking behavior 124 Understanding treatment as promoting and assisting self-change 141 A diagram of integrative self-control theory (after Kotabe & Hofmann, 2015) 159 Overview of the structure of the reflective-impulsive model 169 The development of habit 180 Strategies to inhibit habit 185 Schematic illustration of a reinforcement learning model of behavior indicating possible points at which interventions may exert their effects 198 Schematic representation of the major theorists’ model 212 Schematic representation of the integrated behavioral model (Montaño & Kasprzyk, 2015) 214 Schematic representation of the prototype willingness model (Gibbons, Houlihan, & Gerrard, 2009) 214 Schematic representation of the integrated behavior change model (Hagger & Chatzisarantis, 2014) 217 Behavioral epidemiology framework 240 Common correlates associated with health behaviors 241 Socioecological approach indicating different levels of environmental influence on behavior 252

List of Figures

18.2

19.1 20.1 23.1 26.1 27.1 29.1 33.1 35.1 35.2 35.3 36.1 36.2 37.1 37.2 41.1 41.2 41.3 41.4 43.1 44.1 45.1 46.1

Logic model for relationships among behavior change methods, behavioral determinants, behaviors, environmental conditions, and desired outcomes such as health, safety, and quality of life (Bartholomew Eldredge et al., 2016) The information–motivation–behavioral skills model (Fisher & Fisher, 1992) Mapping the paths within the experimental medicine approach Three overarching aims of using implementation science theories, models, and frameworks Cost-effectiveness plane Contribution of health behaviors to inequities in all-cause mortality by SES indicator in Petrovic et al. (2018) Screenshot of the Fitz app, a digital behavior change intervention tool Diagram illustrating the potential mechanisms of imagery interventions in changing behavior Dimensions of motivating style and the specific instructional behaviors within the autonomy-supportive and controlling styles Mechanisms of action within self-determination theory’s dual-process model Procedural timeline for an autonomy-support intervention program (ASIP) Impediments and facilitators of behavior change Average gym visits with and without incentives A framework showing how monitoring may lead to changes in behavior A chronology plot depicting a patient’s medication intake from a Medication Event Monitoring System A model of the habit formation process Four forms of “habit disruption” A decision tree to aid selection of habit change strategies (Gardner, 2019) The habit growth curve Continuum of dyadic behavior change techniques The social identity model of behavior change Theoretical conceptualization of how motivational interviewing (MI) works based on Miller and Rose (2009) A basic model of a behavior change mechanism of action

xi

259 274 288 337 378 390 418 483 511 513 518 524 526 539 548 601 601 603 610 633 652 665 689

Tables

2.1 3.1 4.1 5.1 7.1 10.1 10.2 10.3 10.4 11.1 17.1 17.2 17.3 17.4 18.1 19.1 21.1 22.1 22.2 23.1 24.1

24.2

Main effects of theory of planned behavior interventions (from Steinmetz et al., 2016) page 27 Social cognitive theory behavior change techniques 41 Summary of prospective relations between constructs specified by the health belief model and protection motivation theory and behavior 52 Translation of common-sense model constructs into techniques and intervention components for promoting physical activity and a healthy diet to lose weight 65 Health action process approach (HAPA) constructs according to phases of behavior change 90 Stage of change: example client and provider tasks for behavior change intervention 138 Defining and recognizing client processes of change 139 Therapist activities to promote experiential and behavioral processes of change 143 Therapist inventory of intervention strategies 143 Overview of some intervention approaches mapped onto components of integrative self-control theory 155 Examples of ecological models that have been developed and used in predicting and changing behavior 239 Pairings of ecological models and theories 242 Strategies and constructs targeted by the H2GO! intervention adapted from Wang et al. (2016) 243 Example intervention targets and strategies across the domains of the ecological model 245 Summary of major environmental-level theories including key readings and examples and methods for change 254 Ten tasks involved in the design, development, implementation, and evaluation of behavior change interventions 270 Intervention development tasks mapped onto phases and steps of two influential frameworks and two reviews of frameworks 302 Examples of experimental and quasi-experimental evaluation designs 321 Issues to consider when evaluating a behavior change intervention 326 Implementation strategies, involved actors, and implementation characteristics targeted by the strategies (Leeman et al., 2017) 339 Stakeholder engagement and key learning points while developing myHT, an app-based intervention to promote adherence to hormonal therapy in breast cancer survivors 354 Patient and health care professional stakeholder views of screening for depression in acute care settings and suggested actions to increase its implementation 357

List of Tables

27.1 30.1 31.1 32.1 32.2 32.3 32.4 33.1 34.1 34.2 35.1 37.1 37.2 38.1 38.2 38.3 39.1 41.1 42.1 43.1 44.1 44.2 44.3 45.1 45.2

45.3 46.1

Steps to developing an evidence-based tobacco cessation intervention for African American adults Key qualitative approaches for research Example techniques for changing attitudes from the behavior change technique taxonomy v1 (Michie et al., 2013) and related descriptions Techniques to prompt mastery experiences matched with techniques from behavior change taxonomies Techniques to prompt vicarious experiences matched with techniques from behavior change taxonomies Techniques to prompt verbal persuasion matched with techniques from behavior change taxonomies Techniques to prompt somatic and affective states matched with techniques from behavior change taxonomies Mental imagery techniques matched with techniques from behavior change taxonomies with related techniques and descriptions Affective intervention techniques matched with techniques from behavior change taxonomies with related techniques and descriptions Examples of simple messages used to change affective attitudes Autonomy supportive intervention program (ASIP)-enabled changes in students’ adaptive and maladaptive behavior How behavior change taxonomies incorporate self-monitoring techniques, along with related techniques and descriptions Steps and considerations for using monitoring to promote changes in behavior Use of goal setting in everyday life Behavior change techniques related to goal setting from existing taxonomies The dos, don’ts, and personal preferences for goal setting interventions Planning concepts and their relationship with intervention taxonomies Techniques for changing habits from the behavior change technique taxonomy v1 (BCTTv1; Michie et al., 2013) with related techniques and descriptions. Mapping nudge approaches to cognitive biases Examples of dyadic behavior change techniques along the continuum of dyadic behavior change techniques displayed in Figure 43.1 Six hypotheses from the social identity approach to health (Haslam et al., 2018) that can make group-based interventions a powerful resource for behavior change Techniques to promote shared social identity Techniques from Michie et al.’s (2013) behavior change technique taxonomy version 1 interpreted through a group-based behavior change lens The four interrelated components of motivational interviewing (MI) “spirit” (after Miller & Rollnick, 2013) Examples of motivational interviewing (MI) techniques matched with techniques from behavior change taxonomies with related techniques and descriptions and MI coding schemes Categories of in-session client change talk Themes emerging from the handbook with source chapters and further reading, key concepts, and a summary of the significance of each theme for behavior change

xiii

395 432 449 464 465 466 467 485 498 500 515 541 544 556 557 567 574 605 623 634 651 655 657 662

668 671 678

Sidebars

2.1 2.2 3.1 4.1 4.2 5.1 6.1 6.2 6.3 7.1 7.2 8.1 8.2 9.1 9.2 10.1 11.1 11.2 11.3 12.1 13.1 13.2 13.3 13.4 15.1 16.1 16.2 16.3 17.1 17.2 17.3 18.1

Testing the measurement model of the theory of planned behavior Testing the structural model of the theory of planned behavior Specific versions of social cognitive theory Protection motivation in an occupational setting Protection motivation in advertising to promote purchasing behavior Need help with asthma control? There’s a common-sense model app for that! How to induce deliberative and implemental mindsets Experimenting with implementation intentions Strong tests of the behavioral impact of implementation intentions A health action process approach intervention in oral health Using the health action process approach (HAPA) and technology to change behavior Williams et al.’s (1998) test of a process model of autonomy support, autonomous motivation, and medication adherence Chatzisarantis and Hagger’s (2009) process evaluation of an autonomy-support intervention A history lesson: What happened to Powers’s control theory? Control theory: The behavioral illusion and testing the controlled variable Using a readiness ruler to assess motivation and stage of change status Origins of self-control research Desires and goals in integrative self-control theory Enactment constraints in integrative self-control theory Key terms and concepts of the reflective-impulsive model Habits cause “behavioral slips” The measurement of habit strength A field experiment testing the habit discontinuity hypothesis Using the self-report habit index to measure cue-contingent habit strength What is an integrated theory? Brief definitions of theoretical terms used in social identity approaches Self-definition varies with shifts from personal to social identity The importance of social norms and social identification Examples of research on the correlates of behavior using an ecological framework Example of how the ecological model is paired with social cognitive theory Examples of interventions that have used an ecological framework Case study: Planned change at the organizational level – promoting a fuel-saving driving style among van drivers delivering mail

page 24 25 37 50 50 69 81 82 84 97 99 109 114 121 129 138 151 152 153 165 182 183 186 188 209 227 228 231 242 243 245 258

List of Sidebars

18.2 18.3 20.1 20.2 21.1 22.1 22.2

22.3 23.1 23.2 23.3 24.1 25.1 25.2 25.3 26.1 27.1 27.2 27.3 28.1 28.2 28.3 28.4 29.1 29.2 29.3 29.4 29.5 30.1 32.1 32.2 32.3 33.1

Case study: Planned change at the societal level – Greenpeace activists recruit powerful allies Case study: Planned change at the organizational level – influenza vaccination Connecting principles and practice: the case of vaccination behavior Resources for making informed decisions about constructs and measures Recording and reporting the intervention development process Case study: Personalizing behavioral interventions through single-patient (N-of-1) trials (Davidson et al., 2014) How do individuals’ health behaviors respond to an increase in the supply of health care? Evidence from a natural experiment (Fichera, Gray, & Sutton, 2016) Summary of key points for choosing evaluation outcomes Key constructs used in implementation science The monitoring and evaluation of implementation of behavior change interventions according to taxonomies of best practices An example of the questionnaire to measure behavior change intervention adoption Using stakeholders to guide the development of the myHT app to support adherence to hormone therapy in breast cancer survivors Developing guiding principles for an intervention for cancer survivors Illustration of using nonjudgmental language Using think-aloud methods to optimize an activity planner The increase of health care expenditure and the application of economic evaluation PROGESS-Plus: a framework addressing heterogeneity in socioeconomic status measures Health behaviors as causal links between inequity and health Developing culturally specific behavior change interventions New settings affect communities: community gardens as an example Effects of client choice in food pantries Spotlight on policy: the COMMIT study Spotlight on policy: How can community members be engaged in the policy formation process? Types of digital resources and tools for behavior change Example of a standalone digital behavior change tool: The Fitz app Participatory codesign and development Sample digital behavior change intervention study Challenges in assessing the efficacy of digital interventions Case study of behavior change in the workplace using a social practice perspective (Hargreaves, 2011) The “This Girl Can” campaign Can reading fluency and self-efficacy of reading fluency be enhanced with an intervention targeting the sources of self-efficacy? (Aro et al., 2018) Intervention on sources of self-efficacy for breastfeeding (Nichols et al., 2009) Behavior change techniques relating to mental imagery from existing taxonomies

xv

259 260 286 293 308 323

325 327 334 344 345 354 363 365 367 376 388 390 395 406 407 409 410 417 418 421 422 423 437 465 471 471 485

xvi

33.2 34.1 34.2 34.3 35.1 35.2 36.1 36.2 37.1 37.2 38.1 39.1 40.1 40.2 41.1 41.2 41.3 41.4 42.1 42.2 43.1 44.1 45.1 45.2

List of Sidebars

Defining and measuring imagery “ability” Behavior change technique(s) from existing taxonomies Evaluative conditioning Gamification and exergames A behavior change technique taxonomy for teachers’ behavior change Refining the autonomy-support intervention program (ASIP) Incentives to exercise Switching costs and market power Behavior change technique(s) related to monitoring from existing taxonomies Self-monitoring in the Adherence Improving self-Management Strategy (AIMS) Goal setting behavior change technique(s) from existing taxonomies Approaches to planning Meta-analytic evidence on self-control training Example of a self-control intervention protocol Integrating habit-based techniques into the behavior change technique taxonomy version 1 (BCTTv1) Discontinuing habitual car use How are habits formed? Developing healthy child-feeding habits Heuristics or characteristics that cause decisions to deviate from expected utility theory Example of a nudge Dyadic behavior change techniques and relations with existing taxonomies Building social identity interventions The motivational interviewing (MI) “spirit” Empirical evidence for the theoretical conceptualization of motivational interviewing (MI)

488 498 499 503 514 519 526 532 541 548 557 575 591 592 604 609 610 612 624 626 634 655 662 664

Contributors

ch a rl e s a br a ha m , University of Melbourne ice k aj ze n , University of Massachusetts, Amherst ch r i s t o p h er j. ar m i t ag e , University of Manchester la u re n ar u n de l l, Deakin, University m ar i a be c km a n , Karolinska Institute co r i na b er l i, University of Zurich an n e h . be rm a n , Karolinska Institute s t ua r t j . h . b i d dl e , University of Southern Queensland kat h er in e b radbu r y , University of Southampton ny l a r . b r an s co m b e, University of Kansas p a u l m . b r o w n, University of California, Merced and Auckland University of Technology ge r t- j a n d e b r ui j n , University of Amsterdam m ar i j n d e b ru i n , Radboud University Medical Center and University of Aberdeen j o s h u a b y rn e s, Griffith University raff cal i tr i , University of Exeter l i n d a d. ca m e r o n , University of California, Merced m ar y ca r te r, University of Exeter ke r ry ch a mb er lain, Massey University ni ko s l . d . ch a tz is ar a n tis , Curtin University s u n g h y eo n ch e o n, Korea University m ar k co n n er , University of Leeds do m i n i c co n ro y , University of East London rik c ru t ze n , Maastricht University sarah d en for d , University of Exeter

xviii

List of Contributors

rol a nd deu t sc h , University of Würzburg car lo c . d ic le men t e, University of Maryland Baltimore County k at h i di e l , Ruhr University Bochum s i m o n e d o h l e , Ruhr University Bochum m a r t i n d o w n e s , Griffith University k at h e r i n e l . d o w n i n g, Deakin University tr acy e pt on , University of Manchester s a r a f l e s z a r - p a v l o v i c´ , University of California, Merced p a u l c . f le t ch e r, University of Cambridge d av i d p. fr e n c h , University of Manchester malt e f rie se , Saarland University be n j a m i n ga r dn e r, King’s College London m a rl ee n gi l l eb a ar t , Utrecht University u ri g n ee z y, University of California, San Diego p e te r m. g ol l w i t ze r , New York University and University of Konstanz s t i n a g r a n t , University of Victoria m e a g a n m . g r a y do n , University of Maryland Baltimore County m a r t i n s . h a g ge r , University of California, Merced and University of Jyväskylä k yr a ha m i l to n , Griffith University n el li h a n ko n e n, University of Helsinki w e n d y h a r d e m a n , University of East Anglia cat h er in e h asl a m, University of Queensland s . a l e x an d e r h a s l am , University of Queensland car o lin e j. h e nde rs o n, Open University k yl i e d. h es ke t h, Deakin University w i l he l m h o f m a n n , Ruhr University Bochum g ar e t h j. ho l l a n d s, University of Cambridge m o ni c a w eb b h o o p e r, Case Comprehensive Cancer Center k ar o l i n a h or o d ys k a, SWPS, University of Social Sciences and Humanities j o a nn a l . h ud s on , King’s College London

List of Contributors

xix

ly n d s a y d . hu g h es , King’s College London b l a i r t . j oh n s o n , University of Connecticut ag n e k a j a c ka i t e, Berlin Social Science Center da v i d j. k a v a na g h , Queensland University of Technology lu c as ke l le r, University of Konstanz te n i e k h a ch i k i a n, University of California, Merced w ill ia m m . p . k le in , National Cancer Institute ni na k no l l, Freie Universität Berlin ge r j o ko k , Maastricht University ph il lip pa lal ly , University College London ev e ly n e d e l e eu w , Maastricht, University he l en a lin d q vi s t , Karolinska Institute ta r u l i n tu n e n, University of Jyväskylä k a r o l i n a l o b cz o w s k a , SWPS, University of Social Sciences and Humanities ja ni na l u¨ s ch e r, University of Zurich al e ks a nd r a l u s z c zy n s k a , SWPS, University of Social Sciences and Humanities and University of Colorado at Colorado Springs an t o nia l yo n s, Victoria University of Wellington warr en man se ll , University of Manchester th e re s a m . m a r te a u, University of Cambridge ly n s a y m at t he w s , University of Glasgow st ep han m ei er , Columbia University zo e m oo n , King’s College London le a nn e m or r i s o n , University of Southampton r o n a m o s s- m o r r i s , King’s College London i n g r i d m u l l e r , University of Southampton m ar c us r. m un a f o` , University of Bristol s h ei na o rb e ll , University of Essex s u s a n pa t e r s o n , University of Miami am a n d a l . r e b a r , Central Queensland University john mar sh a ll re eve , Australian Catholic University

xx

List of Contributors

kat h er in e j . r eyn o lds, The Australian National University r y a n e . r h o d e s , University of Victoria d e n i s e d e r i d de r , Utrecht University a l e x an d e r j. ro t hm a n , University of Minnesota rob er t a. c. ru it er , Maastricht University ri ch ard m . r yan , Australian Catholic University j o s a l m o n , Deakin University p e t e r sc h m i d t , University of Gießen u rt e s ch o lz , University of Zurich be n jamin sc hu¨ z , University of Bremen r a l f s c h w a r z e r, Freie Universität Berlin p a u l s cu f f h a m, Griffith University p a s c h a l s he e ra n , University of North Carolina at Chapel Hill s h a r o n a . s i m p s o n , University of Glasgow f r i t z s t r a c k , University of Würzburg emin a su b asˇ i c´ , University of Newcastle m a r k t a r ra n t , University of Exeter d e n i s e t ay l o r, Auckland University of Technology e d i s o n t r i c k e t t , University of Miami bas v e rp l an ken , University of Bath ti an jiao wan g, Griffith University li sa m. warn e r, Medical School Berlin t h o m a s l . w e b b , University of Sheffield m a r t i n w i l k i n s o n , University of Auckland d av i d m . w i l l i a m s, Brown University lor en w il li s, Australian National University lu c y y a r dl e y, University of Bristol and University of Southampton h i na z a hi d, University of Essex

1

Changing Behavior: A Theoryand Evidence-Based Approach Martin S. Hagger, Linda D. Cameron, Kyra Hamilton, Nelli Hankonen, and Taru Lintunen

1.1 Introduction Many problems observed in today’s society can be linked, directly or indirectly, to human behavior. Problems with roots in, or links with, behavior include debilitating illnesses and chronic conditions (e.g., cardiovascular disease, cancers, obesity, sexually transmitted infections), global pandemics of communicable diseases (e.g., SARS, H1N1, COVID-19), mental health problems (e.g., depression, anxiety), addictions (e.g., substance abuse), social and interpersonal problems (e.g., bullying, abuse and violence in relationships), financial difficulties (e.g., personal debt, problem gambling), criminal behavior (e.g., social disorder, vandalism), educational challenges (e.g., truancy, attentional difficulties), and environmental concerns (e.g., overuse of nonrenewable resources, failures to recycle or save energy). Analogously, regular participation in relevant behaviors is associated with adaptive outcomes such as better health and wellbeing, positive mental health, better functioning in the workplace, in interpersonal relationships, and at school, and more environmentally conscious choices and consumer behavior. Vast databases of archival statistics demonstrating how behavior is linked to social problems are at the disposal of organizations responsible for developing policy to tackle them. Such data signal the need for behavioral solutions and have catalyzed fervent interest in the determinants of behavior and in methods and strategies to change behavior. Governments, organizations (private and public corporations, schools, community organizations), and professionals (government officials, health care workers, managers,

teachers) recognize the value of developing strategies to change the behavior of targeted population groups in order to promote adaptive outcomes. To date, legislation (e.g., seat belt use) and regulation (e.g., banning smoking in public places) stand as some of the most successful means to change population behavior. However, in many cases, such initiatives are not possible, feasible, or acceptable. As a consequence, alternative approaches to behavior change are needed. Scientific inquiry into behavior change has entered into the mainstream. Recognition of the importance of behavior change to solving social problems has led governments to engage scientists from various disciplines within the social and behavioral sciences to inform policy and develop effective behavior change strategies targeting highpriority, behavior-related problems. For example, governments and organizations have invested in funding initiatives to develop research evidence (e.g., National Cancer Institute, 2019; National Institutes of Health, 2019; Nielsen et al., 2018; OBSSR, 2016), commissioned reports and evidence syntheses (e.g., Behavioral Insights Team, 2019b; Cabinet Office, 2011; NICE, 2007, 2012, 2014), and set up working groups, expert panels, and conferences with an advisory purview on behavior change (e.g., Behavioral Insights Team, 2019a; Brandt & Proulx, 2015; House of Lords, 2011; Ogilvie Consulting, 2019; Spring et al., 2013). Martin S. Hagger’s contribution was supported by a Finnish Distinguished Professor (FiDiPro) award from Business Finland (1801/31/2015). https://doi.org/10.1017/9781108677318.001

2

MARTIN S. HAGGER, LINDA D. CAMERON, KYRA HAMILTON, NELLI HANKONEN, AND TARU LINTUNEN

Researchers in the fields of psychology, sociology, behavioral economics, philosophy, implementation science, education, communication science, and political science have been at the forefront of research on behavior change (e.g., Little & Akin-Little, 2019; Nielsen et al., 2018; Sheeran, Klein, & Rothman, 2017; Young et al., 2015). Scientists in these disciplines have been primarily responsible for creating and disseminating evidence on behavior change at all levels on the “continuum of evidence,” from basic theoretical research on determinants and mechanisms to translational research on the application of strategies to change behavior in specific contexts. The proliferation of behavior change research is predicated on the recognized importance of evidence-based practice that began in fields like medicine (Guyatt et al., 1992) and allied health (NICE, 2019) and has since been adopted in other domains such as education (EEF, 2019) and crime reduction and policing (College of Policing, 2019). Such evidence is critical to the application of scientific principles to inform the development of effective behavioral solutions to social problems – a science of behavior change (Michie, Rothman, & Sheeran, 2007; Nielsen et al., 2018).

1.2 A Theory- and Evidence-Based Approach to Behavior Change 1.2.1 Charting Progress in TheoryBased Behavior Change The development of a science of behavior change owes a great deal to formative research applying behavioral theories to predict and understand and change behavior. For example, research beginning in the 1950s in the field of social psychology, particularly social cognition research on persuasion, motivation, and decision-making, focused on identifying the determinants of behavior in social contexts (e.g., Bandura, 1971; Bem, 1965; Festinger, 1964). Such research employed laboratory and field

experiments to provide controlled tests of the basic theory-derived mechanisms (Klein et al., 2015; Sheeran et al., 2017). This research built the foundations of many contemporary theories of behavior and the basis for many of the methods used to change behavior (Michie, 2008). Parallel to this theory-focused experimental research, many behavioral interventions have tended to focus on the design features of interventions (e.g., recruitment, randomization, measurement evaluation, etc.) and on change in behavioral and associated outcomes, with less focus on theory, mechanisms, and intervention content responsible for behavior change (Prestwich et al., 2014; Prestwich, Webb, & Conner, 2015). While such research is informative on the effects of interventions in particular contexts, it provides little information on how the intervention worked and the processes involved. Such intervention research defines efficacy and effectiveness in terms of behavioral outcomes alone, without evaluation of the processes that led to the changes. These two parallel disciplines of research have resulted in a rich but disparate literature that includes a combination of rigorous experimental research focusing on testing specific theories and particular mechanisms, that is, research that attempts to unpack the “black box” of how change works, and behavioral intervention trials with a broader focus on changing behavior and related outcomes. It is only relatively recently that researchers have engaged in coordinated efforts to develop formal theories and systems that reconcile these bodies of research and broaden understanding of how to develop, evaluate, and implement behavior change interventions (Bartholomew Eldredge et al., 2016; Michie, van Straalen, & West, 2011).

1.2.2 The Value of Theory and the Emergence of a Science of Behavior Change Behavioral theories provide important information on the aspects of interventions responsible

Changing Behavior: A Theory- and Evidence-Based Approach

for, and likely to facilitate, behavior change and the individual, social, contextual, and environmental conditions that may magnify or diminish intervention effects (Glanz & Bishop, 2010; Kwasnicka et al., 2016; Michie et al., 2008). However, while behavioral scientists recognize the value of a theoretical basis in guiding interventions and typically claim that their interventions are based on theory, syntheses of research testing the efficacy of behavioral interventions have revealed that the reported detail of their basis in theory tends to be limited. In fact, reviews of behavioral interventions purported to be theory-based suggest that relatively few describe how the theory has been used, and those that do seldom test how elements of the theory change alongside changes in behavior and outcomes (Goodwin et al., 2016; McDermott et al., 2016; Prestwich et al., 2014). Further, while some research suggests that theory-based interventions have greater efficacy and reliability in changing behavior than those that do not, or, at least, those based on theory lead to more reliable, less variable outcomes (Bishop et al., 2015; McEwan et al., 2018; Webb et al., 2010), others suggest that a theoretical basis does not confer greater efficacy (Dalgetty, Miller, & Dombrowski, 2019; Prestwich et al., 2014). Such research is, however, held back by limitations in the extent and precision of reporting of intervention content and use of theory (e.g., how the theory was used in developing the intervention content, the appropriateness of the theory for the target problem and population) and, particularly, by insufficient or unclear descriptions of intervention content. This presents challenges to researchers aiming to identify links between theory and intervention content (Connell Bohlen et al., 2019). In addition, behavioral interventions with no reported basis in theory tap into similar mechanisms to those that report using theory, making comparisons relating to theory effectiveness difficult to interpret.

3

Recent developments in the science of behavior change have sought to resolve some of these issues. One of the most important advances has been the development of formal systems to efficiently and effectively describe behavioral theories and interventions. Pioneering work derived from content analyses of behavioral interventions has sought to identify the methods or techniques used to change behavior (Abraham & Michie, 2008; Kok et al., 2016; Michie et al., 2013; Michie et al., 2015). The goals of this research are to identify the unique, separable techniques that represent the essential “building blocks” of behavioral interventions, arrive at a common set of terms to describe behavioral interventions, and develop a formal means to classify them. Conceptual work and reviews of behavioral intervention research internationally have led to the development of taxonomies of behavior change techniques (Kok et al., 2016; Michie, Ashford et al., 2011; Michie et al., 2013). The taxonomies are classification systems of isolated behavior change techniques. Recently, this work has been extended to link the intervention techniques described in behavior change technique taxonomies with constructs from theories that represent “mechanisms of change,” that is, how the techniques purportedly change behavior (Carey et al., 2018; Connell et al., 2018; Michie et al., 2008; Michie et al., 2017; Michie, Webb, & Sniehotta, 2010). Further research has also sought to describe the key processes required for the specification, development, testing, and reporting of behavioral interventions (Abraham, 2012; Bartholomew Eldredge et al., 2016; French et al., 2012; Michie, van Straalen, & West, 2011; Michie et al., 2015; Sheeran et al., 2017). These efforts have been directed toward developing an evidence base that is optimally informative of the intervention methods that are effective in changing behavior, how such interventions work, and how they can be converted and implemented into workable, feasible solutions to behavioral problems.

4

MARTIN S. HAGGER, LINDA D. CAMERON, KYRA HAMILTON, NELLI HANKONEN, AND TARU LINTUNEN

1.2.3 Emerging Approaches to Behavior Change Intervention Development Identifying behavior change techniques, and describing links between the techniques and theory-based constructs, forms part of broader approaches that seek to describe essential processes in the development of behavior change interventions. These approaches are based on the premise that developing knowledge on the methods that are optimally effective and reliable in changing behavior, and the factors that determine their effectiveness, is essential if the science of behavior change is to offer meaningful solutions to those tasked with tackling problems linked to behavior. Numerous examples of these approaches exist, and many have adopted a theory-to-practice approach that focuses on identifying not only “what works” when it comes to methods of changing behavior but how those methods work and how they can be developed into practical and acceptable interventions for delivery with high fidelity to a target population (for a review, see O’Cathain et al., 2019). Prominent approaches to intervention development with a strong focus on theoretical basis include the intervention mapping approach (Bartholomew Eldredge et al., 2016); the behavior change wheel (Michie, van Straalen, & West, 2011); application of the theoretical domains framework (French et al., 2012); the experimental medicine approach (Sheeran et al., 2017); and the mapping change mechanisms approach (Abraham, 2012; see Appendix 1.1 in the supplemental materials for details). Key steps common to these approaches are (1) identifying the problem that warrants change; (2) identifying the behavior or behavior-related outcome of interest; (3) identifying the theory- and evidence-based mechanisms on how a particular change technique or approach is likely to “work” in changing behavior and working them into a “logic model”; (4) embedding the change technique or approach into an intervention and planning and designing a method or “trial” to

test the proposed model; (5) planning means to evaluate efficacy/effectiveness as well as process; and (6) planning for implementation of the intervention. Some of the approaches focus mainly on describing the first four steps (steps 1 to 4) in the process (French et al., 2012; Sheeran et al., 2017), while others follow all steps from problem specification to implementation. These approaches mark important progress on behavior change intervention development, and they have provided researchers and practitioners with a clear blueprint of the required procedures to develop theory- and evidence-based behavior change interventions and, for some of the approaches, the necessary procedures to evaluate their efficacy, proposed mechanism of change, and implementation effectiveness (see Appendix 1.1, supplemental materials).

1.3 The Handbook of Behavior Change The Handbook of Behavior Change was developed to provide comprehensive coverage of research and practice in behavior change, from basic research based on theory to the application of behavior change interventions that are optimally effective in solving social problems. The handbook brings together current evidence in research and practice into a single resource that outlines the fundamental principles and latest advances in theory on behavior change; details evidence on key considerations required to develop, implement, evaluate, and translate behavior change interventions; and provides a series of clear-language, step-by-step guidelines for practitioners and interventionists from multiple fields. It pools knowledge from leading experts at the cutting edge of behavior change theory, research, and practice and provides in-depth, evidence-based works that summarize current knowledge in this emerging science. The handbook reflects the multidisciplinary nature of behavior change, encompassing perspectives

Changing Behavior: A Theory- and Evidence-Based Approach

from diverse disciplines in the social sciences, both established (e.g., psychology, sociology, economics, research methods) and emerging (e.g., intervention design, behavioral economics, implementation science, translational medicine). Central to the handbook is a basis on theory and evidence from these disciplines, comprehensive coverage, balance in views and perspectives, and emphasis on the translation of behavior change research into practices that lead to meaningful changes and solutions to problems with a behavioral cause. Chapter authors have been selected because they are at the forefront of generating evidence in behavior change through their own theory, research, and practice and are therefore eminent authorities on their selected topic. The handbook is organized into three parts: Part I: Theory and Behavior Change; Part II: Methods and Processes of Behavior Change: Intervention Development, Application, and Translation; and Part III: Behavior Change Interventions: Practical Guides to Behavior Change. These parts reflect themes from the generalized approaches to developing theorybased behavior change interventions outlined in the previous section, beginning with the application of theory, through to the development, implementation, and evaluation of interventions, and the important considerations involved in translating interventions into practice (Abraham, 2012; Bartholomew Eldredge et al., 2016; French et al., 2012; Michie, van Straalen, & West, 2011; Sheeran et al., 2017). Part I focuses on the use of psychological, behavioral, social, and environmental theories to inform behavior change and is targeted at all those interested in how theory is used to inform interventions and how applying those theories postulate the mechanisms that engender behavior change. Part II focuses on the processes and methods needed to design, develop, implement, evaluate, and translate behavior change interventions. Part III provides sets of practical guidelines on how to develop behavior change interventions using

5

particular behavior change techniques or methods. The next sections provide an overview of the chapters in each section.

1.4 Part I: Theory and Behavior Change Part I addresses the application of theory to behavior change. The chapters cover key approaches that have been applied to identify behavioral determinants and predict behavior and to inform the development of behavior change interventions. Each chapter provides an outline of the key tenets of the theory, including its basic assumptions, constructs, and predictions, followed by a review of relevant empirical evidence. Next, the ways in which the theory has been used and operationalized in changing behavior, particularly the behavior change methods or techniques implied by the theory, and how these have been embedded in interventions to test their effects on behavior change, is reviewed. The chapters then provide a review of experimental and intervention research that has applied the identified methods or techniques in changing behavior and the relative strength, value, and quality of the findings for research and practice. Finally, the chapters outline possible avenues for further development of research and practice, particularly gaps in knowledge and how they may be addressed. Many of the theories covered in the chapters stem from the field of applied social psychology, a discipline that has contributed much to the prediction of behavior and the means to change it. A key perspective is the social cognition approach, which focuses on individual attitudes and beliefs as key determinants of behavior change. Chapters from this perspective include the theories of reasoned action and planned behavior (Ajzen, 1991; Chapter 2, this volume), social cognitive theory (Bandura, 1986; Chapter 3, this volume), the health belief model and protection motivation theory (Rogers, 1975; Rosenstock,

6

MARTIN S. HAGGER, LINDA D. CAMERON, KYRA HAMILTON, NELLI HANKONEN, AND TARU LINTUNEN

1974; Chapter 4, this volume), and the commonsense model of self-regulation (Leventhal, Meyer, & Nerenz, 1980; Chapter 5, this volume). Applications of these theories have been highly influential in identifying the social determinants of behavior change. However, noted boundary conditions and limitations of social cognition theories (e.g., Head & Noar, 2014; Noar & Zimmerman, 2005; Trafimow, 2012), particularly the observed “shortfall” in the relationship between individuals’ intentions and their behavior (Orbell & Sheeran, 1998; Rhodes & de Bruijn, 2013), have inspired approaches that incorporate other decision-making constructs and processes. Notable among these are “dual-phase” theories of action that distinguish between a motivational phase, in which intentions are formed, and a volitional phase, in which intentions are augmented with implemental strategies like planning to facilitate enactment. The model of action phases (Heckhausen & Gollwitzer, 1987; Chapter 6, this volume) and the health action process approach (Schwarzer, 2008; Chapter 7, this volume) are dual-phase theories in which constructs such as planning determine the strength of the intentionbehavior relationship. Part I also covers theories that adopt alternative perspectives on behavior change. These perspectives share common features in that they view behavior change as a function of internal motivational and regulatory processes. For example, selfdetermination theory focuses on the quality rather than quantity of motivation as a determinant of behavior (Deci & Ryan, 1985; Ryan & Deci, 2017; Chapter 8, this volume). Another approach, control theory (Carver & Scheier, 1982; Powers, 1973; Chapter 9, this volume), adopts a systems perspective from physics and engineering to provide an analysis of behavior based on the regulation of perceptual inputs and outputs and maintenance of homeostatic equilibrium. A further contrasting approach is offered by the transtheoretical model (Prochaska & DiClemente, 1982; Chapter 10, this volume). Developed from therapeutic work in

clinical contexts, the model adopts a stage approach to understanding behavior change from pre-contemplation to action, with processes of change determining shifts from one stage to the next. A final perspective is offered by integrative self-control theory (Chapter 11, this volume). The theory proposes that capacity to regulate impulses and engage in effortful control over behavior determines whether an individual will be successful in controlling their behavior or succumbing to desires. One of the limitations of social cognition and motivational theories applied to behavior change is that they tend to view behavior change as resulting from reasoned, deliberative processes that are considered effortful and cognitively demanding. However, there has been renewed interest in “dual-process” theories (Bargh, 1994; Fazio, 1990), more recently popularized by Kahneman (2011), which suggest that behavior is a function of two interacting processes or “routes” to behavior: an “impulsive” process, in which action is determined by a rapid, low-effort process that occurs with relatively low conscious awareness; and a “reflective” process, in which action is controlled by a slower, intentional process that requires considerable cognitive effort and high awareness (Strack & Deutsch, 2004; Chapter 12, this volume). An understanding of automatic, nonconscious processes is also central to theories on habit. Developing adaptive habits, as well as breaking maladaptive habits, is important to behavior change (Aarts & Dijksterhuis, 2000; Hagger, 2019; Orbell & Verplanken, 2010; Wood, 2017; Chapter 13, this volume). Recent approaches to behavior change have focused on how subtle changes to individuals’ environment at the point of decision can alter behavioral patterns. These approaches come from a broad perspective known as “nudging” or choice architecture, made popular by Thaler and Sunstein (2008). Marteau et al. (2012; Chapter 14, this volume) outline recent perspectives of how these types of interventions, and other interventions based on environment changes, may influence behavior, with a

Changing Behavior: A Theory- and Evidence-Based Approach

predominant focus on implicit, nonconscious processes. More recently, theorists have developed integrated models that bring together constructs from social cognition and motivational theories and nonconscious and planning processes from dual-process and dual-phase theories, respectively (Hagger, 2009; Chapter 15, this volume). These theories integrate different theoretical approaches to produce more comprehensive descriptions of behavior and behavior change. While many theoretical perspectives on behavior change take an individual-focused approach, it is clear that individuals do not act in a “social vacuum,” and their behavior is often a function of beliefs and perceptions influenced by their group membership. Social identity approaches apply group-related constructs to explain individual behavior (Tajfel & Turner, 1986; Chapter 16, this volume). More broadly, ecological theories suggest that behavior change should be considered in the social and environmental contexts in which people behave (Sallis, Owen, & Fisher, 2015; Chapter 17, this volume). These theories suggest that, beyond beliefs and motives, behavior is a function of determinants operating at multiple levels including the individual (e.g., socioeconomic status, age, gender), environmental (e.g., policies supporting behavior, access to facilities, areas of residence), and social structural (e.g., family and peer group structure and beliefs) levels. Similar perspectives are considered in community theories of behavior change, which provide a multilevel systems approach to identifying factors at the individual, organizational, community, and societal levels that influence behavior change (see Chapter 18, this volume).

1.5 Part II: Methods and Processes of Behavior Change: Intervention Development, Application, and Translation Part II focuses on procedures and processes in developing, testing, evaluating, and implementing behavior change interventions, including key

7

methodological and practical considerations to consider when planning and developing interventions to change behavior. Each chapter provides an overview of the topic, summarizes key research, and outlines implications for subsequent research and practice. Emphasis is placed on the means by which behavior change efforts are delivered, evaluated, refined, and put into practice. Part II begins with a broad overview of the process of developing behavioral interventions (Abraham, 2012; Chapter 19, this volume). This is followed by a summary of a systematic experimental approach to developing behavior change interventions (Sheeran et al., 2017; Chapter 20, this volume). Together these chapters provide two broad approaches to developing, implementing, and evaluating behavioral interventions based on theory and mechanisms of change (see Section 1.2 and Appendix 1.1, supplemental materials). Subsequent chapters focus on the development, evaluation, and implementation of behavior change interventions. Multiple guidelines for designing behavioral interventions (e.g., MRC, 2019), informed by interdisciplinary evidence and expert consultation, have been produced over the past two decades (Chapter 21, this volume). The guidelines have also informed how behavioral interventions should be evaluated in formal research as well as ongoing evaluation in practice (Chapter 22, this volume). Considerable emphasis has also been placed on the importance of translating efficacious behavior change interventions into practice (Chapter 23, this volume). Related to this is the necessity of involving appropriate stakeholders (e.g., leaders of organizations, policymakers, personnel involved in intervention delivery; Chapter 24, this volume) and users (i.e., members of the target population; Chapter 25, this volume) in all these processes. Finally, economic evaluations of behavior change interventions provide essential information on cost-effectiveness to those in charge of budgets (Chapter 26, this volume). It is also important that intervention designers recognize the challenges presented by the physical

8

MARTIN S. HAGGER, LINDA D. CAMERON, KYRA HAMILTON, NELLI HANKONEN, AND TARU LINTUNEN

and social contexts in which behavior change interventions are delivered. Documented disparities observed in economically underserved communities in areas such as health and education suggest that such communities are likely to benefit most from behavior change (Schüz et al., 2017). However, evidence suggests that behavior change interventions are less likely to be effective, and engagement is likely to be much lower, in these populations. Interventionists, therefore, need to modify and adjust interventions in order to address and accommodate disparities (Chapter 27, this volume). More broadly, behavioral interventions need to be sensitive to the communities in which they are delivered. Consistent with stakeholder engagement, community interventions need to be tailored to the specific needs of communities through, for example, cocreation by, and consultation with, community stakeholders (Chapter 28, this volume). Part II also includes chapters on special methodological topics in behavior change. Advances in mobile and handheld technology (e.g., smartphones, activity trackers, mobile cameras, and recording devices) afford interventionists with new opportunities to deliver interventions in innovative ways to improve their reach and effectiveness (Chapter 29, this volume). It is also important to note that much of the research evidence on behavior change adopts a quantitative, hypothetico-deductive approach that has become synonymous with the “scientific method.” However, critical and qualitative research approaches provide important perspectives and evidence on behavior change that can augment or supplant evidence from quantitative approaches (Chapter 30, this volume).

1.6 Part III: Behavior Change Interventions: Practical Guides to Behavior Change An overarching goal of this handbook is to provide the most up-to-date, evidence-based guidance on methods that can be used to effectively

change behavior and how to go about doing so. This guidance is for researchers interested in advancing behavior change interventions and producing new evidence of intervention effectiveness, as well as practitioners and stakeholders seeking effective methods for changing behavior based on current theory and evidence. The chapters in Part III, therefore, offer researchers and practitioners specific evidence-based guidelines on behavior change interventions. Each chapter focuses on a particular set of behavior change techniques or approaches that have gained prominence. The techniques and approaches include those that have been frequently used in behavior change research and practice such as persuasion, planning, and support for self-efficacy, as well as emerging approaches such as the use of imagery and strategies based on behavioral economics, self-control, and habit. Each chapter begins with an overview of the behavior change technique or approach, including how the technique has been identified in behavior change taxonomies (when relevant) and a review of current evidence supporting its application. Where evidence is available, chapter authors have produced “step-by-step” guides as examples that outline means to implement the technique in practice, with consideration of key technical issues, including (1) typical means of delivery; (2) target audience and behaviors; (3) enabling or inhibiting factors; (4) training and skills required; (5) intensiveness or “dose” of the intervention technique or method required; (6) evaluation of intervention fidelity; (7) evaluation of intervention effectiveness; and (8) typical materials needed to implement the intervention. Many of the chapters provide exercises, scripts, forms, worksheets, and measures as supplemental materials that can be adapted by interventionists to develop the content of behavioral interventions. These materials are aimed at providing useful compendium of behavioral intervention contents based on current evidence.

Changing Behavior: A Theory- and Evidence-Based Approach

Part III comprises chapters outlining specific techniques for changing behavior by altering individuals’ beliefs, attitudes, risk perceptions, and other social cognition constructs (Chapters 31 and 32, this volume) and changing individuals’ motivation (Chapters 33, 34, 35, 36, 37, and 38, this volume). Additional chapters detail approaches that promote intention enactment using planning techniques based on dual-phase models of action (Chapter 39, this volume) and approaches that promote behavior change by tapping into implicit or nonconscious processes (Chapters 40, 41, and 42, this volume). Beyond individual-level interventions, techniques and methods to change the behavior of individuals and groups (e.g., romantic partnerships and other dyads, groups defined by shared membership, and ad hoc social groupings) through social influence and group processes are also covered (e.g., Chapters 43 and 44, this volume). While the above groupings reflect the predominant target process or mechanism of change of the approaches covered in Part II (see Appendix 1.1, supplemental materials), it is important to note that many approaches include more than one technique and may, therefore, tap into more than one change mechanism (Connell Bohlen et al., 2019). For example, some affect-based interventions focus on changing behavior by enhancing risk perceptions (e.g., fear-inducing messages), but they can also tap into more nonconscious processes (e.g., reducing positive affect). Incentive-based interventions may change behavior by promoting motivation (e.g., increasing perceptions of the value of a behavioral outcome), but they may also evoke more automatic, spontaneous behavior change (e.g., changing behavior by conditioning through reward). Similarly, approaches such as motivational interviewing comprise multiple techniques that overlap with many other behavior change techniques (Hardcastle et al., 2017), as well as techniques and components (e.g., relational components; Dombrowski, O’Carroll, & Williams, 2016; Hagger & Hardcastle, 2014)

9

unique to the approach, but motivational interviewing is treated and applied as a single “approach” (e.g., Chapter 45, this volume).

1.7 Using the Handbook The different parts of the handbook provide overall guidelines on general chapter themes at the global level based on a theory- and evidencebased approach to behavior change. The chapters in Part I are likely to be of most interest to those interested in learning more about specific theories and mechanisms of action relevant to behavior change. The chapters in Part II are designed for those interested in developing, implementing, and evaluating interventions, with keen attention to method and design. Part III is likely to be of primary interest to those seeking practical guidance on the content of interventions and how to put them into practice and in obtaining adaptable materials currently available to do so. Each chapter is designed to “stand alone”, so that it can be read in isolation of other chapters, but, given overlaps in content and approach, references to other chapters and further reading are provided. There are also thematic and conceptual links between many of the chapters, both within and across the parts of the book. For instance, many theories reviewed in Part I are linked with Part III chapters that focus on particular techniques or approaches that target constructs from those theories, consistent with intervention mapping (Bartholomew Eldredge et al., 2016; Chapter 19, this volume) and experimental medicine (Sheeran et al., 2017; Chapter 20, this volume) approaches. It is recommended that readers consult the relevant companion chapters to supplement the insights gained from the chapter they are reading. A useful guide to thematic links between chapters is presented in Appendix 1.2 (supplemental materials). It is also important to note that some of the chapters that outline general approaches and methods with respect to behavior change will be relevant to many or

10

MARTIN S. HAGGER, LINDA D. CAMERON, KYRA HAMILTON, NELLI HANKONEN, AND TARU LINTUNEN

all of the chapters in the handbook. Finally, an important feature of each chapter is the provision of “practical summaries” to accompany the scientific summary provided in the abstract. The summaries highlight the key messages and recommendations relevant to behavior change research and practice covered in the chapter and increase access to this information for readers without a technical background.

1.8 Summary and Conclusion The proliferation of problems with behavioral origins has catalyzed research on, and development of, strategies to promote behavior change, as well as how research findings may be leveraged by interventionists to effectively change behavior in practice. The Handbook of Behavior Change provides a comprehensive overview of research and practice on behavior change authored by specialists from multiple disciplines in the social sciences and other disciplines. The handbook adopts a theory- and evidence-based approach to changing behavior and provides coverage of the major theoretical and empirical developments in this emerging field. As interest in behavior change to address social problems in diverse domains such as health, education, economics, and the workplace grows, the handbook makes a unique contribution to knowledge by bringing together contemporary perspectives and up-to-date evidence with practical guidance on how to change behavior. Whether seeking to gain knowledge of the multiple perspectives on behavior change, conducting research to test the efficacy and effectiveness of behavior change methods, or developing behavior change interventions in practice, the handbook is designed to be useful to readers involved in each of these endeavors. It also presents new ideas and directions for research and practice toward a better understanding of behavior change and producing effective solutions to many of the problems faced by society.

References Aarts, H., & Dijksterhuis, A. (2000). Habits as knowledge structures: Automaticity in goal-directed behavior. Journal of Personality and Social Psychology, 78, 53–63. https://doi.org/10.1037/0022-3514.78.1.53 Abraham, C. (2012). Mapping change mechanisms onto behaviour change techniques: A systematic approach to promoting behaviour change through text. In C. Abraham & M. Kools (Eds.), Writing Health Communication: An Evidence Based Guide (pp. 99–116). Thousand Oaks, CA: Sage. Abraham, C., & Michie, S. (2008). A taxonomy of behavior change techniques used in interventions. Health Psychology, 27, 379–387. https://doi.org/ 10.1037/0279-6133.27.3.379 Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50, 179–211. https://doi.org/10.1016/ 0749-5978(91)90020-T Bandura, A. (1971). Vicarious and self-reinforcement processes. In R. Glaser (Ed.), The Nature of Reinforcement (pp. 228–278). New York: Academic Press. Bandura, A. (1986). Social Foundations of Thought and Action: A Social-Cognitive Theory. Englewood Cliffs, NJ: Prentice-Hall. Bargh, J. A. (1994). The Four Horsemen of automaticity: Awareness, efficiency, intention, and control in social cognition. In R. S. Wyer & T. K. Srull (Eds.), Handbook of Social Cognition, Vol. 2 (pp. 1–40). Hillsdale, NJ: Erlbaum. Bartholomew Eldredge, L. K., Markham, C. M., Ruiter, R. A. C., Fernández, M. E., Parcel, G. S., & Kok, G. (2016). Planning Health Promotion Programs: An Intervention Mapping Approach (4th ed.). San Francisco: Jossey-Bass. Behavioral Insights Team. (2019a). The Behavioural Insights Team. Website. www.bi.team Behavioral Insights Team. (2019b). The Behavioural Insights Team Annual Update Report 2017–18. www.bi.team/publications/the-behavioural-insightsteam-annual-report-2017-18 Bem, D. J. (1965). An experimental analysis of selfpersuasion. Journal of Experimental Social Psychology, 1, 199–218. https://doi.org/10.1016/ 0022-1031(65)90026–0

Changing Behavior: A Theory- and Evidence-Based Approach

Bishop, F. L., Fenge-Davies, A. L., Kirby, S., & Geraghty, A. W. A. (2015). Context effects and behaviour change techniques in randomised trials: A systematic review using the example of trials to increase adherence to physical activity in musculoskeletal pain. Psychology and Health, 30, 104–121. https://doi.org/10.1080/08870446 .2014.953529 Brandt, M. J., & Proulx, T. (2015). QTIPs: Questionable theoretical and interpretive practices in social psychology. Behavioral and Brain Sciences, 38, 277–289. https://doi.org/10.1017/ S0140525X14001149 Cabinet Office. (2011). Government Response to the Science and Technology Select Committee Report on Behaviour Change. https://assets.publishing.service .gov.uk/government/uploads/system/uploads/attach ment_data/file/60538/Government-ResponseHouse-of-Lords-Inquiry-Behaviour-Change.pdf Carey, R. N., Connell, L. E., Johnston, M. et al. (2018). Behavior change techniques and their mechanisms of action: A synthesis of links described in published intervention literature. Annals of Behavioral Medicine. https://doi.org/10.1093/abm/kay078 Carver, C. S., & Scheier, M. F. (1982). Control-theory: A useful conceptual framework for personalitysocial, clinical and health psychology. Psychological Bulletin, 92, 111–135. https://doi .org/10.1037//0033-2909.92.1.111 College of Policing. (2019). What Works College of Policing. Website. https://whatworks.college .police.uk/Pages/default.aspx Connell, L. E., Carey, R. N., de Bruin, M. et al. (2018). Links between behavior change techniques and mechanisms of action: An expert consensus study. Annals of Behavioral Medicine. https://doi.org/ 10.1093/abm/kay082 Connell Bohlen, L. E., Michie, S., de Bruin, M. et al. (2019). Do combinations of behaviour change techniques that occur frequently in interventions reflect underlying theory? PsyArXiv. https://doi .org/10.31234/osf.io/49djm Dalgetty, R., Miller, C. B., & Dombrowski, S. U. (2019). Examining the theory-effectiveness hypothesis: A systematic review of systematic reviews. British Journal of Health Psychology. https://doi.org/10.1111/bjhp.12356

11

Deci, E. L., & Ryan, R. M. (1985). Intrinsic Motivation and Self-Determination in Human Behavior. New York: Plenum Press. Dombrowski, S. U., O’Carroll, R. E., & Williams, B. (2016). Form of delivery as a key “active ingredient” in behaviour change interventions. British Journal of Health Psychology, 21, 733–740. https://doi.org/10.1111/bjhp.12203 EEF (Education Endowment Foundation). (2019). The Education Endowment Foundation. Website. https://educationendowmentfoundation .org.uk Fazio, R. H. (1990). Multiple processes by which attitudes guide behavior: The MODE model as an integrative framework. In M. P. Zanna (Ed.), Advances in Experimental Social Psychology Vol. 23 (pp. 75–109). San Diego, CA: Academic Press. Festinger, L. (1964). Behavioral support for opinion change. Public Opinion Quarterly, 28, 404–417. https://doi.org/10.1086/267263 French, S. D., Green, S. E., O’Connor, D. A. et al. (2012). Developing theory-informed behaviour change interventions to implement evidence into practice: A systematic approach using the Theoretical Domains Framework. Implementation Science, 7. https://doi .org/10.1186/1748-5908-7-38 Glanz, K., & Bishop, D. B. (2010). The role of behavioral science theory in development and implementation of public health interventions. Annual Review of Public Health, 31, 399–418. https://doi.org/10.1146/annurev.publhealth .012809.103604 Goodwin, L., Ostuzzi, G., Khan, N., Hotopf, M. H., & Moss-Morris, R. (2016). Can we identify the active ingredients of behaviour change interventions for coronary heart disease patients? A systematic review and meta-analysis. PLoS ONE, 11, e0153271. https:// doi.org/10.1371/journal.pone.0153271 Guyatt, G., Cairns, J., Churchill, D. et al. (1992). Evidence-based medicine: A new approach to teaching the practice of medicine. JAMA, 268, 2420–2425. https://doi.org/10.1001/jama.1992 .03490170092032 Hagger, M. S. (2009). Theoretical integration in health psychology: Unifying ideas and complimentary explanations. British Journal of Health Psychology, 14, 189–194. https://doi.org/10.1348/ 135910708X397034

12

MARTIN S. HAGGER, LINDA D. CAMERON, KYRA HAMILTON, NELLI HANKONEN, AND TARU LINTUNEN

Hagger, M. S. (2019). Habit and physical activity: Theoretical advances, practical implications, and agenda for future research. Psychology of Sport and Exercise, 42, 118–129. https://doi.org/10.1016/ j.psychsport.2018.12.007 Hagger, M. S., & Hardcastle, S. J. (2014). Interpersonal style should be included in taxonomies of behaviour change techniques. Frontiers in Psychology, 5, 254. https://doi.org/10.3389/fpsyg.2014.00254 Hardcastle, S. J., Fortier, M. S., Blake, N., & Hagger, M. S. (2017). Identifying content-based and relational techniques to change behavior in motivational interviewing. Health Psychology Review, 11, 1–16. https://doi.org/10.1080/17437199.2016.1190659 Head, K. J., & Noar, S. M. (2014). Facilitating progress in health behaviour theory development and modification: The reasoned action approach as a case study. Health Psychology Review, 8, 34–52. https://doi.org/10.1080/17437199.2013.778165 Heckhausen, H., & Gollwitzer, P. M. (1987). Thought contents and cognitive functioning in motivational and volitional states of mind. Motivation and Emotion, 11, 101–120. https://doi.org/10.1007/ BF00992338 House of Lords. (2011). Science and Technology Committee: Second Report of Session 2010–12: Behaviour Change. www.publications.parliament .uk/pa/ld201012/ldselect/ldsctech/179/179.pdf. Kahneman, D. (2011). Thinking, Fast and Slow. New York: Farrar, Straus & Giroux. Klein, W. M. P., Shepperd, J. A., Suls, J., Rothman, A. J., & Croyle, R. T. (2015). Realizing the promise of social psychology in improving public health. Personality and Social Psychology Review, 19, 77–92. https://doi.org/10.1177/1088868314539852 Kok, G., Gottlieb, N. H., Peters, G.-J. Y. et al. (2016). A taxonomy of behavior change methods: An intervention mapping approach. Health Psychology Review, 10, 297–312. https://doi.org/ 10.1080/17437199.2015.1077155 Kwasnicka, D., Dombrowski, S. U., White, M., & Sniehotta, F. F. (2016). Theoretical explanations for maintenance of behaviour change: A systematic review of behaviour theories. Health Psychology Review, 10, 277–296. https://doi.org/ 10.1080/17437199.2016.1151372 Leventhal, H., Meyer, D., & Nerenz, D. (1980). The common sense model of illness danger. In S.

Rachman (Ed.), Medical Psychology, Vol. 2 (pp. 7–30). New York: Pergamon Press. Little, S. G., & Akin-Little, A. (2019). Behavioral Interventions in Schools: Evidence-Based Positive Strategies. Washington, DC: American Psychological Association. Marteau, T. M., Hollands, G. J., & Fletcher, P. C. (2012). Changing human behavior to prevent disease: The importance of targeting automatic processes. Science, 337, 1492–1495. https://doi .org/10.1126/science.1226918 McDermott, M. S., Oliver, M., Iverson, D., & Sharma, R. (2016). Effective techniques for changing physical activity and healthy eating intentions and behaviour: A systematic review and metaanalysis. British Journal of Health Psychology, 21, 827–841. https://doi.org/10.1111/bjhp.12199 McEwan, D., Beauchamp, M. R., Kouvousis, C., Ray, C. M., Wyrough, A., & Rhodes, R. E. (2018). Examining the active ingredients of physical activity interventions underpinned by theory versus no stated theory: A meta-analysis. Health Psychology Review. https://doi.org/10.1080/ 17437199.2018.1547120 Michie, S. (2008). What works and how? Designing more effective interventions needs answers to both questions. Addiction, 103, 886–887. https:// doi.org/10.1111/j.1360-0443.2007.02112.x Michie, S., Ashford, S., Sniehotta, F. F., Dombrowski, S. U., Bishop, A., & French, D. P. (2011). A refined taxonomy of behaviour change techniques to help people change their physical activity and healthy eating behaviours: The CALO-RE taxonomy. Psychology and Health, 26, 1479– 1498. https://doi.org/10.1080/08870446.2010 .540664 Michie, S., Carey, R. N., Johnston, M. et al. (2017). From theory-inspired to theory-based interventions: A protocol for developing and testing a methodology for linking behaviour change techniques to theoretical mechanisms of action. Annals of Behavioral Medicine, 52, 501– 512. https://doi.org/10.1007/s12160-016-9816-6 Michie, S., Johnston, M., Francis, J., Hardeman, W., & Eccles, M. (2008). From theory to intervention: Mapping theoretically derived behavioural determinants to behaviour change techniques. Applied Psychology: An International Review, 57,

Changing Behavior: A Theory- and Evidence-Based Approach

660–680. https://doi.org/10.1111/j.1464-0597.2008 .00341.x Michie, S., Richardson, M., Johnston, M. et al. (2013). The behavior change technique taxonomy (v1) of 93 hierarchically clustered techniques: Building an international consensus for the reporting of behavior change interventions. Annals of Behavioral Medicine, 46, 81–95. https://doi.org/ 10.1007/s12160-013-9486-6 Michie, S., Rothman, A., & Sheeran, P. (2007). Current issues and new directions in Psychology and Health: Advancing the science of behavior change. Psychology and Health, 22, 249–253. https://doi.org/10.1080/14768320701233582 Michie, S., van Straalen, M., & West, R. (2011). The behaviour change wheel: A new method for characterising and designing behaviour change interventions. Implementation Science, 6, 42. https://doi.org/10.1186/1748-5908-6-42 Michie, S., Webb, T. L., & Sniehotta, F. F. (2010). The importance of making explicit links between theoretical constructs and behaviour change techniques. Addiction, 105, 1897–1898. https:// doi.org/10.1111/j.1360-0443.2010.03161.x Michie, S., & West, R. (2013). Behaviour change theory and evidence: A presentation to Government. Health Psychology Review, 7, 1–22. https://doi.org/10.1080/17437199.2011.649445 Michie, S., Wood, C. E., Johnston, M., Abraham, C., Francis, J., & Hardeman, W. (2015). Behaviour change techniques: The development and evaluation of a taxonomic method for reporting and describing behaviour change interventions (a suite of five studies involving consensus methods, randomised controlled trials and analysis of qualitative data). Health Technology Assessment, 19, 99. https://doi.org/10.3310/hta19990 MRC (Medical Research Council). (2019). Developing and Evaluating Complex Interventions. https:// mrc.ukri.org/documents/pdf/complex-interven tions-guidance National Cancer Institute. (2019). Behavioral research program. National Cancer Institute. Website. https://cancercontrol.cancer.gov/brp National Institutes of Health. (2019). Science of behavior change common fund. National Institutes of Health. Website. https://common fund.nih.gov/behaviorchange

13

NICE (National Institute for Health and Care Excellence). (2007). Behaviour Change: General Approaches. Public Health Guideline [PH6]. www.nice.org.uk/guidance/ph6 NICE (National Institute for Health and Care Excellence). (2012). Individual-Level Behaviour Change External Evidence Review 1: Review of Current NICE Guidance and Recommendations. www.nice.org.uk/guidance/ph49/evidence/evi dence-reviews-pdf-430402861 NICE (National Institute for Health and Care Excellence). (2014). Behaviour Change: Individual Approaches. Public Health Guideline [PH49]. www.nice.org.uk/guidance/ph49 NICE (National Institute for Health and Care Excellence). (2019). National Institute for Health and Care Excellence. Website. www.nice.org.uk Nielsen, L., Riddle, M., King, J. W. et al. (2018). The NIH Science of Behavior Change Program: Transforming the science through a focus on mechanisms of change. Behaviour Research and Therapy, 101, 3–11. https://doi.org/10.1016/j .brat.2017.07.002 Noar, S. M., & Zimmerman, R. S. (2005). Health behavior theory and cumulative knowledge regarding health behaviors: Are we moving in the right direction? Health Education Research, 20, 275–290. https://doi.org/10.1093/her/cyg113 O’Cathain, A., Croot, L., Sworn, K. et al. (2019). Taxonomy of approaches to developing interventions to improve health: A systematic methods overview. Pilot and Feasibility Studies, 5, 41. https://doi.org/10.1186/s40814-019-0425-6 OBSSR (Office of Behavioral and Social Sciences Research). (2016). Strategic Plan 2017–2021: Healthier Lives through Behavioral and Social Sciences. Bethesda, MD: OBSSR and National Institutes of Health. Ogilvie Consulting. (2019). Nudgestock. Website. https://nudgestock.co.uk. Orbell, S., & Sheeran, P. (1998). “Inclined abstainers”: A problem for predicting health related behaviour. British Journal of Social Psychology, 37, 151–165. https://doi.org/10.1111/j.2044-8309.1998.tb01162.x Orbell, S., & Verplanken, B. (2010). The automatic component of habit in health behavior: Habit as cue-contingent automaticity. Health Psychology, 29, 374–383. https://doi.org/10.1037/a0019596

14

MARTIN S. HAGGER, LINDA D. CAMERON, KYRA HAMILTON, NELLI HANKONEN, AND TARU LINTUNEN

Powers, W. T. (1973). Behavior: The Control of Perception. Chicago: Adline Publishing. Powers, W. T. (2003). Behavior: The Control of Perception (Vol. 2, rev. ed.). New Canaan, CT: Benchmark Publications. Prestwich, A., Sniehotta, F. F., Whittington, C., Dombrowski, S. U., Rogers, L., & Michie, S. (2014). Does theory influence the effectiveness of health behavior interventions? Meta-analysis. Health Psychology, 33, 465–474. https://doi.org/ 10.1037/a0032853 Prestwich, A., Webb, T. L., & Conner, M. (2015). Using theory to develop and test interventions to promote changes in health behaviour: Evidence, issues, and recommendations. Current Opinion in Psychology, 5, 1–5. https://doi.org/10.1016/j.copsyc.2015.02.011 Prochaska, J. O., & DiClemente, C. C. (1982). Transtheoretical therapy: Toward a more integrated model of change. Psychotherapy: Theory, Research & Practice, 19, 276–288. https://doi.org/10.1037/h0088437 Rhodes, R. E., & de Bruijn, G. J. (2013). How big is the physical activity intention-behaviour gap? A meta-analysis using the action control framework. British Journal of Health Psychology, 18, 296– 309. https://doi.org/10.1111/bjhp.12032 Rogers, R. W. (1975). A protection motivation theory of fear appeals and attitude change. Journal of Psychology, 91, 93–114. https://doi.org/10.1080/ 00223980.1975.9915803 Rosenstock, I. M. (1974). Historical origins of the health belief model. Health Education Monographs, 2, 328–335. https://doi.org/10.1177/ 109019817400200403 Ryan, R. M., & Deci, E. L. (2017). Self-Determination Theory: Basic Psychological Needs in Motivation, Development and Wellness. New York: Guilford Press. Sallis, J. F., Owen, N., & Fisher, E. B. (2015). Ecological models of health behavior. In K. Glanz, B. K. Rimer, & K. Viswanath (Eds.), Health Behavior and Health Education: Theory, Research, and Practice (5th ed., pp. 43–64). San Francisco: Jossey-Bass. Schüz, B., Li, A. S.-W., Hardinge, A., McEachan, R. R. C., & Conner, M. (2017). Socioeconomic status as a moderator between social cognitions and physical activity: Systematic review and meta-

analysis based on the Theory of Planned Behavior. Psychology of Sport and Exercise, 30, 186–195. https://doi.org/10.1016/j.psychsport.2017.03.004 Schwarzer, R. (2008). Modeling health behaviour change: How to predict and modify the adoption and maintenance of health behaviors. Applied Psychology: An International Review, 57, 1–29. https://doi.org/10.1111/j.1464-0597.2007.00325.x Sheeran, P., Klein, W. M. P., & Rothman, A. J. (2017). Health behavior change: Moving from observation to intervention. Annual Review of Psychology, 68, 573–600. https://doi.org/10.1146/ annurev-psych-010416-044007 Spring, B., Ockene Judith, K., Gidding Samuel, S. et al. (2013). Better population health through behavior change in adults. Circulation, 128, 2169–2176. https://doi.org/10.1161/01.cir.0000435173.25936.e1 Strack, F., & Deutsch, R. (2004). Reflective and impulsive determinants of social behavior. Personality and Social Psychology Review, 8, 220– 247. https://doi.org/10.1207/s15327957pspr0803_1 Tajfel, H., & Turner, J. C. (1986). The social identity theory of intergroup behaviour. In S. G. Worchel & W. Austin (Eds.), Psychology of Intergroup Relations (2nd ed., pp. 7–24). Chicago: Nelson-Hall. Thaler, R., & Sunstein, C. R. (2008). Nudge: Improving Decisions about Health, Wealth, and Happiness. New Haven, CT: Yale University Press. Trafimow, D. (2012). The role of auxiliary assumptions for the validity of manipulations and measures. Theory and Psychology, 22, 486–498. https://doi .org/10.1177/0959354311429996 Webb, T. L., Joseph, J., Yardley, L., & Michie, S. (2010). Using the Internet to promote health behavior change: A systematic review and metaanalysis of the impact of theoretical basis, use of behavior change techniques, and mode of delivery on efficacy. Journal of Medical Internet Research, 12, e4. https://doi.org/10.2196/jmir.1376 Wood, W. (2017). Habit in personality and social psychology. Personality and Social Psychology Review, 21, 389–403. https://doi.org/10.1177/ 1088868317720362 Young, W., Davis, M., McNeill, I. M. et al. (2015). Changing behaviour: Successful environmental programmes in the workplace. Business Strategy and the Environment, 24, 689–703. https://doi .org/10.1002/bse.1836

Part I Theory and Behavior Change

2

Changing Behavior Using the Theory of Planned Behavior Icek Ajzen and Peter Schmidt

Practical Summary This chapter describes the theory of planned behavior (TPB) and its use as a framework for behavior change interventions. According to the TPB, human behavior is guided by three kinds of considerations: beliefs about the likely consequences of my behavior, beliefs about what important others think I should do, and beliefs about my ability to carry out the behavior. These beliefs provide the basis for the formation of an intention to engage in the behavior, but actual performance of the behavior depends on behavioral control. Interventions are effective to the extent that they produce changes in the beliefs that underpin intentions and when they ensure that people have the skills and resources needed to enact their intentions. Research has provided extensive empirical support showing that the TPB can be used not only to predict and explain behavior but also to help design effective behavior change interventions.

2.1 Introduction Human behavior is determined by multiple factors and changing well-established behavioral patterns is notoriously difficult. From a purely theoretical perspective, the problem is easily conceptualized. To adopt a novel behavior, people must be motivated to do so and they must be capable of acting on their motivation. The motivation for change may be self-generated, as when a person decides to go on a weight-loss diet, stop smoking, or avoid procrastinating. In these instances, behavior change interventions must focus on ways and means to help people initiate and maintain the desired behavior. In other situations, however, people are not self-motivated and behavior change interventions must first generate the motivation for change, followed by a focus on implementation. Behavior change is often said to proceed in stages, from precontemplation through contemplation, preparation, action, and maintenance (Bamberg, 2013;

Prochaska & DiClemente, 1983, 1992; see Chapter 10, this volume). In this “transtheoretical model of change,” it is assumed that different psychological processes are involved when people move from one stage to the next and that, therefore, different intervention strategies are required at different stages of the change sequence. Although of possible heuristic value, a moment’s reflection reveals that the same intervention can influence people at different stages of change (Fishbein & Ajzen, 2010). For example, a message to the effect that taking an aspirin pill every day can prevent a heart attack may prompt people at the precontemplation stage to contemplate engaging in this behavior, it may motivate people at the contemplation stage to make preparations to take a daily aspirin pill, it may get people at the preparation stage to actually start taking a daily aspirin pill, and it may encourage people at the action stage to maintain this behavior over time. Consistent with https://doi.org/10.1017/9781108677318.002

18

ICEK AJZEN AND PETER SCHMIDT

this argument, empirical research (Armitage & Arden, 2002) has shown that the consecutive stages in the transtheoretical model correspond to increasing levels of behavioral intentions. The approach to behavior change described in the present chapter is based on the TPB (Ajzen, 1991, 2012a). This theory has been used successfully to explain and predict behavior in a multitude of behavioral domains, from physical activity to drug use, from recycling to choice of travel mode, from safer sex to consumer behavior, to name just a few (for meta-analytic syntheses of this research, see, e.g., Albarracín, Fishbein, & Goldestein de Muchinik, 1997; Armitage & Conner, 1999; Hagger, Chatzisarantis, & Biddle, 2002; McDermott et al., 2015; Riebl et al., 2015; Sheeran & Taylor, 1999; Winkelnkemper, Ajzen, & Schmidt, 2019). The TPB is also increasingly being used as a framework for designing and evaluating the effects of behavior change interventions (for meta-analytic syntheses of intervention studies, see Sheeran et al., 2016; Steinmetz et al., 2016; Tyson, Covey, & Rosenthal, 2014). Instead of describing behavior change as moving through five distinct stages, the TPB focuses on the two basic facets of change mentioned at the beginning of this section: motivating people to engage in a desired behavior and, to the extent that engaging in the behavior poses difficulties, enabling them to realize the behavior. As discussed in greater detail in the next sections, the first facet involves forming an intention to perform the behavior of interest, while the second facet has to do with the relation between stated intentions and actual behavior. Practitioners have long recognized the value of a well-formulated, empirically validated theory for designing and evaluating the effects of behavior change interventions (see Bickman, 1987; Coryn et al., 2011; Michie & Prestwich, 2010; Rossi, Lipsey, & Freeman, 2004). As Kurt Lewin (1952) famously observed, “there is nothing more practical than a good theory” (p. 169). Consistent with this reasoning, a meta-analysis of behavior change interventions delivered online (Webb

et al., 2010) showed that interventions based on a theoretical framework were more effective in changing health-related behavior than nontheorybased interventions. The three most frequently used theoretical frameworks were the previously mentioned transtheoretical model (see Chapter 10, this volume), Bandura’s (1977) social cognitive theory (see Chapter 3, this volume), and the TPB. In a comparison of research with these three theories, interventions based on the TPB had, on average, the strongest impact (effect size r = 0.36) followed by the transtheoretical model (r = 0.20) and social cognitive theory (r = 0.15). This chapter aims to provide an introduction to the TPB as a basis for behavior change interventions. A brief description of the theory and its implications for behavior change is followed by an in-depth discussion of the practical steps involved in designing a promising TPB-based intervention and evaluating its effectiveness.

2.2 Brief Overview of the Theory of Planned Behavior (TPB) As extensive discussions of the TPB and of empirical evidence in support of the theory can be found elsewhere (e.g., Ajzen, 2012a; Fishbein & Ajzen, 2010), the following section provides only a brief overview while taking into account recent developments of the theory. The structural TPB model as currently conceptualized is shown in Figure 2.1.

2.2.1 Definition of the Behavior The TPB starts with a clear definition of the behavior of interest in terms of its target, the action involved, the context in which it occurs, and the time frame. Each of these elements can be defined at varying levels of specificity or generality. However, once the behavior has been defined, all other constructs in the theory must correspond to the behavior in all four elements. This is known as the principle of compatibility (Ajzen, 1988). For

Changing Behavior Using the Theory of Planned Behavior

Behavioral beliefs

19

Attitude toward the behavior Intention

Normative beliefs

Subjective norm

Control beliefs

Perceived behavioral control

Behavior

Actual behavioral control

Figure 2.1 Theory of planned behavior

example, to study entrepreneurship, an investigator may define the behavior of interest at a low level of generality, such as “opening (action) a restaurant (target) in Chicago (context) in the next twelve months (time frame).” Alternatively, the investigator may be interested in entrepreneurship at a more general level and define the behavior as “starting (action) a business (target) in the next twelve months (time).” Note that the target has been expanded to include any type of business, not just a restaurant, and that the context is left unspecified. The particular behavioral definition adopted determines how all constructs in the TPB are to be formulated.

2.2.2 Intentions and Behavioral Control As in other “reasoned action” approaches (e.g., Bandura, 1997; Fisher & Fisher, 1992; Triandis, 1972; see also Fishbein & Ajzen, 2010), the immediate antecedent of behavior in the TPB is the intention to perform the behavior in question; the stronger the intention, the more likely it is that the behavior will follow. To return to the example in Section 2.2.1, the intention to start a business in the next twelve months could be measured and data collected to determine whether participants did or did not implement their intentions. However,

unanticipated events; insufficient time, money, or resources; lack of requisite skills; and a multitude of other factors may prevent people from acting on their intentions. The degree to which people have control over the behavior depends on their ability to overcome barriers of this kind and on the presence of such facilitating factors as past experience and assistance provided by others. In light of these considerations, the TPB postulates that the degree of behavioral control moderates the effect of intention on behavior: The greater the actor’s control over the behavior, the more likely it is that the intention will be carried out. The TPB thus contends that the performance of socially significant behaviors is volitional, the result of an intention, not capricious or performed automatically without conscious awareness. To be sure, when behavior has become routine as a result of repeated performance, intentions may be activated automatically and remain implicit unless effortfully retrieved (Ajzen & Dasgupta, 2015; see also Chapter 13, this volume). Even strong habits are not necessarily unintentional, although habitual behavior may occur spontaneously, without a conscious intention (see Ajzen & Fishbein, 2000 and Chapter 41, this volume). Consider, for example, individuals who have been using their cars to commute to work for many years. As they get ready to leave, they may proceed without

20

ICEK AJZEN AND PETER SCHMIDT

forming a conscious intention to drive their cars to work. However, if asked, they could easily retrieve and report their (implicit) intention.

(the experience). In their aggregate, behavioral beliefs are theorized to produce a positive or negative attitude toward the behavior.

2.2.3.2

2.2.3 Determinants of Intentions Consistent with the notion of reasoned action, people’s behavioral intentions are assumed to be guided by some measure of deliberation, where novel behaviors and important decisions receive more thorough contemplation than relatively less important or routine behaviors (Ajzen & Sexton, 1999). According to the TPB, three kinds of considerations guide the formation of intentions: beliefs about the likely consequences and experiences resulting from performance of the behavior (behavioral beliefs), which, in their aggregate, result in the formation of an attitude toward the behavior; beliefs about the expectations and behaviors of significant social referents (normative beliefs), which produce perceived social pressure to engage or not to engage in the behavior, or subjective norm; and beliefs about factors that may facilitate or impede performance of the behavior (control beliefs), which result in perceived behavioral control or a sense of selfefficacy (Bandura, 1997). It is assumed that the behavioral, normative, and control beliefs that are readily accessible in memory are the prevailing determinants, respectively, of attitude, subjective norm, and perceived behavioral control and that these latter variables influence the behavioral intention (see Figure 2.1). These processes are described in greater detail in Section 2.2.4, which deals with the structural model.

2.2.3.1

Behavioral Beliefs and Attitudes

A behavioral belief is the person’s subjective probability that performing a behavior of interest will lead to a certain outcome (instrumental behavioral beliefs) or involve a certain experience (experiential behavioral beliefs), for example the belief that exercising (the behavior) improves physical fitness (the outcome) or is invigorating

Normative Beliefs and Subjective Norm

Two types of normative belief can be distinguished: injunctive and descriptive (see Cialdini & Trost, 1998; Fishbein & Ajzen, 2010). An injunctive normative belief is the expectation or subjective probability that a given referent individual or group (e.g., friends, family, spouse, coworkers, one’s physician or supervisor) approves or disapproves of performing the behavior under consideration. Descriptive normative beliefs, on the other hand, are beliefs as to whether important others themselves perform the behavior. Both types of beliefs contribute to the overall perceived social pressure to engage in the behavior or subjective norm.

2.2.3.3

Control Beliefs and Perceived Behavioral Control

Control beliefs are concerned with the presence of factors that can facilitate or impede performance of the behavior. Control factors include required skills and abilities; availability or lack of time, money, and other resources; cooperation by other people; and so forth. A control belief is defined as a person’s subjective probability that a given facilitating or inhibiting factor will be present in the situation of interest. In their aggregate, readily accessible control beliefs produce the prevailing perceived behavioral control. Empirical research has identified two subfactors of control: capacity and autonomy (see Fishbein & Ajzen, 2010). Capacity refers to the perceived ability to perform the behavior, as determined by the assumed availability of requisite skills and resources, whereas autonomy is the extent to which people believe that the decision to perform the behavior is entirely up to them.

2.2.4 The Structural Model In most applications of the TPB, the three predictors of intention (attitude, subjective norm, and perceived behavioral control) have been

Changing Behavior Using the Theory of Planned Behavior

treated as additive factors, although, in the original formulation of the theory, Ajzen (1985) discussed the possibility of an interaction between attitude and subjective norm with perceived behavioral control. In its current formulation, favorable attitudes and supportive subjective norms motivate people to perform the behavior, but this motivation leads them to form an intention to engage in the behavior only to the extent that they believe that they are capable of performing the behavior in question. This implies that perceived behavioral control moderates the effects of attitude and of subjective norm on intention (see Figure 2.1). A number of studies provide empirical evidence in support of the proposed interaction effects (see Hukkelberg, Hagtvet, & Kovac, 2014; La Barbera & Ajzen, 2018; Yzer & van den Putte, 2014). It can be seen that behavioral control plays a central role in the theory’s current formulation. First, the degree of actual control moderates the effect of intention on behavior and, second, perceived behavioral control moderates the effects of attitude and of subjective norm on intention. Moreover, because it is usually difficult to know how much control people actually have in a given situation, perceived behavioral control is also used as a proxy for actual control. Of course, perceived behavioral control can usefully serve as a proxy for actual control only to the extent that it is veridical, that is, to the extent that it reflects actual control reasonably well.

2.2.5 Intervention as a Background Factor Many factors not included in the TPB may influence intentions and behavior, including demographic characteristics (age, gender, race, education, income, etc.), personality traits, life values, political ideology, mood and emotions, and so forth. In the TPB, these kinds of variables are considered background factors that have no direct effects on behavior but can influence it indirectly by way of the

21

more proximal antecedents of behavior specified in the theory. A behavior change intervention can be viewed as such a background factor and, like other background factors, its impact is expected to be mediated by the proximal determinants of intentions and behavior. This conception is discussed in greater detail in the next section.

2.3 Using the TPB to Change Behavior As noted, an intervention can be designed to motivate people to engage in a desired behavior or, if they already intend to do so, to help them carry out their intentions successfully. A motivational intervention based on the TPB can achieve its aim in at least three ways. First, it can influence behavioral or normative beliefs leading, respectively, to a more favorable attitude toward the behavior or a more supportive subjective norm, thereby raising motivation to engage in the behavior. Many studies have provided support for this proposition (e.g., Bamberg, 2006; Bamberg & Schmidt, 2001). Second, the intervention can increase perceived behavioral control and thus elevate the effects of attitude and/or subjective norm on intention to perform the behavior. Research has indeed demonstrated that interventions can raise perceived behavioral control and thereby influence intention and behavior (e.g., Kelley & Abraham, 2004; Stecker, McGovern, & Herr, 2012), but the moderating effect of perceived behavioral control has not been studied directly. Third, an intervention can change the relative importance of attitude and subjective norm as determinants of intention. Thus, for example, if the behavior of interest was largely under attitudinal control, an intervention targeted at normative beliefs may raise awareness of social norms and increase their influence on intention. To the best of our knowledge, this proposition has not been tested. Note that changing one or two behavioral, normative, or control beliefs may not be sufficient to bring about a significant change in the overall

22

ICEK AJZEN AND PETER SCHMIDT

attitude, subjective norm, or perceived behavioral control. Moreover, a change in one belief may be accompanied by an unexpected countervailing change in another belief, thereby neutralizing the impact of the intervention. Only if the intervention is effective in changing the total composite of behavioral, normative, or control beliefs can change be expected in attitude, subjective norm, or perceived behavioral control, respectively. An intervention designed to enable people to realize existing favorable intentions can achieve its aim in two ways: (1) it can improve actual control over the behavior, allowing people to act on their intentions, and (2) it can raise perceived behavioral control, which may lead people to persevere in the face of difficulties. For example, individuals who are motivated and intend to engage in a weight-loss diet can be provided with a list of recommended food items and taught how to prepare them. An increase in actual control resulting from the instructions on how to prepare the food items may help them successfully implement the weight-loss diet. In addition, an increase in perceived behavioral control may help them persevere if their initial attempts are less than successful. Greater perceived behavioral control can exert this effect by increasing the effect of attitude or subjective norm on intention, thus strengthening the intention to adhere to the weight-loss diet (see Figure 2.1).

2.4 TPB-Based Interventions: Practical Considerations Some investigators have used the TPB to help design their intervention as well as to evaluate its effects (e.g., Norman, et al., 2018; Stecker, McGovern, & Herr, 2012), while others have implemented an intervention that was not based on the TPB but then used the TPB as a basis for evaluating the intervention by examining its effects on some or all of the theory’s constructs (e.g., Macy et al., 2012). This section outlines the practical steps involved in using the TPB as a framework for designing behavior change interventions, and the

following section outlines means to evaluate its effectiveness. As noted, to be effective, interventions must influence the beliefs that are readily accessible at the time that behavioral performance is contemplated. However, although the TPB can help identify the most likely targets for an intervention, it does not prescribe how the desired changes in beliefs are to be brought about. Investigators have to rely on auxiliary hypotheses or other considerations to decide on the particular strategy that may be most useful for their purposes (see Trafimow, 2012). Michie, van Stralen, and West (2011) suggested a behavior change wheel that delineates possible methods of influence: education (increasing knowledge or understanding), incentivization (creating expectation of reward), coercion (creating expectation of punishment or cost), training (imparting skills), restriction (using rules or laws to regulate behavior), environmental restructuring (changing the physical or social context), and enablement (increasing resources or removing barriers). However, perhaps the most frequently used strategy in the behavior change wheel is persuasive communication, a topic of long-standing interest in social psychology (Ajzen, 2012b; Hovland, Janis, & Kelley, 1953; for in-depth discussions, see Dillard & Shen, 2012). Persuasive messages impart new information in the context of one-on-one encounters, in group discussions and workshops, or in mass media campaigns, such as public service announcements. No matter which strategy is employed, a feature common to all is that they expose people to new information designed to effect a change in behavior.

2.4.1 Preliminary Considerations At the risk of belaboring the obvious, the first step before designing an intervention is to ascertain how many people in the target population already perform the behavior of interest. To take an example, the behavior may be defined as eating a lowcalorie diet, a criterion that specifies the action

Changing Behavior Using the Theory of Planned Behavior

(eating) and target (low-calorie diet) but generalizes across context and time. If such a diet is recommended after a heart attack, it may be found that, in a target population of individuals who have experienced a heart attack, virtually everybody has already adopted a low-calorie diet and no intervention is required. Alternatively, it may be found that a subset of individuals has not yet adopted such a diet and an intervention could be targeted at these individuals. If it is found that a sufficient proportion of individuals in the target population fail to perform the behavior of interest, a second important consideration has to do with the reason for the failure. One possibility is that they are not sufficiently motivated and currently do not intend to perform the behavior. This situation would require an intervention directed at the motivational component. Another possibility is that people are motivated but find it difficult or impossible to carry out their favorable intentions, a situation that requires a different intervention, one designed to increase control over the behavior. The need to examine which of these two possibilities explains failure to perform the desired behavior may appear selfevident but it is not always implemented. For example, Sniehotta (2009) used TPB-based interventions to motivate students to take advantage of underused university sports facilities, with little success. Examination of the findings suggests that most students already held quite favorable attitudes and supportive subjective norms with respect to using the sports facilities, that is, they were generally motivated to engage in this behavior. The reason they did not do so had mostly to do with relatively low perceived behavioral control. Although strengthening motivation further could have had some effect, a more promising approach would have focused on issues of control.

2.4.2 Formative Research All too often, interventions are attempted without sufficient preparatory work. Applying the

23

TPB makes it clear that considerable formative research is required to ensure an effective intervention. This formative research involves several consecutive steps (see Fishbein & Ajzen, 2010, appendix).

2.4.2.1

Belief Elicitation

The first step is the elicitation of readily accessible behavioral, normative, and control beliefs in a representative sample from the target population. This is typically done in a small-group session but can also be done online (see Meitinger & Behr, 2016). Sample participants are given a few minutes to list (1) the outcomes and experiences they associate with the behavior of interest; (2) the individuals or groups who approve or disapprove of performing the behavior and who themselves perform or refuse to perform the behavior; and (3) the factors that may facilitate or interfere with performance of the behavior. Responses are content-analyzed and a list of the most frequently mentioned behavioral, normative, and control beliefs is created. A suitable criterion might be the requirement that a belief selected for inclusion must have been listed by at least 25–30 percent of the participants in the elicitation study or that the set of beliefs selected contains at least 50–60 percent of the total number of beliefs listed by all participants.

2.4.2.2

Administration of a Pilot Questionnaire

Items that can serve as direct measures (reflective indicators) of attitude, subjective norm, perceived behavioral control, and intention are formulated. When behavior is a self-report, behavioral items must also be devised. These items, together with the identified behavioral, normative, and control beliefs, are used to construct a pilot questionnaire that is administered to a new sample from the research population.

2.4.2.3

Finalizing the TPB Instrument

As is discussed in greater detail in Sidebars 2.1 and 2.2, the data from the pilot questionnaire are analyzed to establish the reliability, validity, and

24

ICEK AJZEN AND PETER SCHMIDT

Sidebar 2.1 Testing the measurement model of the theory of planned behavior

Past research with the TPB has typically settled for reporting the internal consistency among items designed to reflect the attitude, subjective norm, perceived behavioral control, and intention constructs directly, usually by means of Cronbach’s α coefficient. However, beyond internal consistency, it is also important for researchers to examine the discriminant validity of the items that were used to assess the TPB’s constructs (see Campbell & Fiske, 1959). The preferred method for evaluating the measurement model of the TPB is confirmatory factor analysis (CFA). This analysis can confirm that each item included in the questionnaire contributes adequately to the TPB construct it is meant to assess (known as convergent validity) and that it does not relate to measures of the theory’s other constructs (known as discriminant validity). In studies adopting a true experimental design, it is not necessary to test the invariance of factor loadings in the experimental and control groups because the random assignment should account for any potential variations in the equivalence of measures. In the case of quasi-experimental designs, researchers should test the model using CFA in both experimental and control groups and formally compare their structures (known as multigroup analysis) in order to confirm that the psychometric properties of the measures of theory constructs are consistent across experimental and control groups. These analyses can also be used to support the validity of the TPB measures across different populations or behaviors or across two or more time points (Brown, 2015).

dimensionality of the reflective indicators; to test the relations between behavioral beliefs and attitudes, between normative beliefs and subjective norms, and between control beliefs and perceived behavioral control; and to test the structural TPB model (Figure 2.1). The results are used (1) to refine the direct measures of the theory’s constructs; (2) to validate the behavioral, normative, and control beliefs as predictors of attitude, subjective norm, and perceived behavioral control, respectively; and (3) to make sure that the theory holds up in terms of predicting intentions and behavior (e.g., de Leeuw et al., 2015). Assuming the results are satisfactory, a final TPB questionnaire is formulated (for a sample questionnaire, see Ajzen, 2019; Fishbein & Ajzen, 2010, appendix).

2.4.2.4

Selecting Target Beliefs

The next preparatory step is to use the pilot data to select the behavioral, normative, and/or control beliefs that are to be targeted in the

intervention. Several general criteria can be suggested. 1. Examination of mean levels of attitude, subjective norm, and perceived behavioral control measures in the population of interest can reveal which of these factors is already favorable and which may be amendable to strengthening. For example, it may emerge that attitudes are highly favorable and that subjective norms are also supportive but that perceived behavioral control is quite low. In this case, the intervention would best be directed at control beliefs in an effort to raise people’s perceived behavioral control (see Elliott & Armitage, 2009). 2. Testing the structural model (see Sidebar 2.2) will provide information about the relative importance of attitude and subjective norm, and of the moderating effect of perceived behavioral control, in the prediction of intentions. All else equal, the intervention should target the more important of these components. Thus, if subjective norms are found to be more important in predicting intentions than attitudes, the

Changing Behavior Using the Theory of Planned Behavior

25

Sidebar 2.2 Testing the structural model of the theory of planned behavior

In past research, the TPB model has often been tested by means of multiple regression analyses in which intentions were regressed on attitude, subjective norm, and perceived behavioral control and behavior was regressed on intention and perceived behavioral control. With the widespread adoption of structural equation modeling, it has become possible to obtain a structural test of the overall fit between the model and the data and to simultaneously test hypotheses derived from the theory, such as mediation and moderation, while taking into account errors in measurement (Rodgers, 2010). One way to test the structural model in interventions or experiments involves including the intervention as a predictor in the model, as a “dummy-coded” variable (e.g., intervention group coded as “1” and the control or comparison group coded “0”). The intervention is then modeled to influence directly the behavioral, normative, and control beliefs and to have no direct effect on attitude, subjective norm, perceived behavioral control, intention, and behavior (for illustrations, see Bamberg, 2006; Stark, Berger, & Hössinger, 2018). A second approach relies on multiple group structural equation modeling to compare theory effects across the experimental and control groups by testing the same model over the two groups and also, for example, the equality of certain cross-group constraints, such as equal regression coefficients. This method also allows testing for moderation effects (Hancock, 2004). A poor fit may be an indication of methodological issues, such as inadequate operationalization of the theory’s constructs or low variance in intentions and/or behavior, but it may also indicate lack of full support for the hypothesized mediation and moderation effects specified in the TPB. At a more general level, it may raise questions about the TPB’s applicability for the population or behavior of interest.

intervention could attempt to make subjective norms more supportive of the behavior. 3. Examination of the individual behavioral, normative, and control beliefs can offer insight into the particular beliefs that need to be changed or new beliefs that could be introduced.

2.4.2.5

Pilot-Testing the Intervention

Finally, pilot work is required to make sure that the intervention to be used is capable of changing the targeted beliefs and that it does not adversely affect beliefs not directly targeted.

2.5 Evaluating an Intervention’s Effectiveness In the preferred method for examining an intervention’s impact, an experimental design is

employed in which participants are randomly assigned into experimental and control groups, the intervention is administered in the experimental group, and its impact is evaluated by comparing this group to a control group that is not exposed to the intervention (see Shadish, Cook, & Campbell, 2002; Chapter 20, this volume). A questionnaire assessing the TPB variables is administered in each group following the intervention and, if long-term effects are to be examined, it can be administered again at one or more later points in time. The impact of the intervention is analyzed by comparing means of the TPB variables in the two groups (see Green & Thompson, 2015; Hancock, 2004). The present section describes how, using an experimental design, intervention effectiveness is

26

ICEK AJZEN AND PETER SCHMIDT

evaluated in the context of the TPB. Before this, however, it should be noted that interventions can also rely on quasi-experimental designs without random assignment into experimental and control groups, such as the regression discontinuity design (Angrist & Pischke, 2015; Shadish et al., 2002). Such designs can also allow strong inferences to be made. For example, using a quasi-experimental design, an intervention to encourage the use of public transportation (Bamberg & Schmidt, 1999) assessed the intervention’s outcome repeatedly, showing that the initially observed increase in the use of public transportation remained stable over a three-year period.

2.5.1 Testing the Impact of a TPB Intervention According to the TPB, an intervention’s impact on behavior is contingent on a sequence of effects linking changes in beliefs to changes in the target behavior (see Figure 2.1). This view has important implications for the amount of change in behavior that can be expected. Even under the best of circumstances, each link in the chain of hypothesized effects is far from perfect. It follows that a given amount of change in behavioral, normative, or control beliefs is likely to produce a smaller change in attitude, subjective norm, or perceived behavioral control. Similarly, any changes in attitude, subjective norm, or perceived behavioral control are likely to produce smaller changes in intention, and any change in intention will have diminishing returns in regard to behavior. This latter phenomenon was confirmed in a meta-analysis of intervention studies (Webb & Sheeran, 2006), which found that successful interventions produced large changes in intentions (mean effect size d = 0.66) but much smaller changes (d = 0.36) in behavior. It follows that, to have a significant impact on behavior, the intervention must be powerful enough to produce substantial changes in the targeted beliefs, such that its influence can reverberate along the chain

of effects to reach the final destination of a change in behavior. Our analysis also indicates that it is not sufficient to demonstrate the effect of an intervention on intentions because many factors can weaken the link between intention and behavior or prevent people from carrying out their intentions (for discussions of the intentionbehavior gap, see Fishbein & Ajzen, 2010; and Chapter 6, this volume).

2.5.2 Empirical Support It is beyond the scope of this chapter to review the sizable body of research that has applied the TPB in efforts to modify behavior. A meta-analytic synthesis of 123 behavior change interventions across a variety of behavioral domains (Steinmetz et al., 2016) confirms that these efforts are generally quite successful. As can be seen in Table 2.1, with the exception of behavioral beliefs, the interventions had, on average, statistically significant effects on the theoretical antecedents of intentions. As a result, they also had a considerable impact on intentions and behavior, with effect sizes of 0.34 and 0.50, respectively. Note, however, that the amount of change varied widely across interventions, as is shown by the significant heterogeneity in effect sizes across studies (see Cochran’s Q coefficients in Table 2.1). This suggests that a significant proportion of the variance in effect sizes is attributable to extraneous factors that moderate the impact of the interventions. Indeed, the meta-analysis established that group-based interventions were more effective than interventions administered in individual settings and that public interventions had a stronger impact than private interventions. Additional research is needed to identify other potential moderators that may influence an intervention’s effectiveness. A study by Norman et al. (2018) will serve to illustrate the successful application of the TPB to design and evaluate a behavior change intervention. The investigators tried to discourage excessive consumption of alcohol (binge drinking) among

Changing Behavior Using the Theory of Planned Behavior

27

Table 2.1 Main effects of theory of planned behavior interventions (from Steinmetz et al., 2016) TPB variables

k (ES)

N

d

Q

Behavioral beliefs Normative beliefs Control beliefs Attitude Subjective norm Perceived behavioral control Intention Behavior

11 (14) 12 (17) 6 (6) 70 (101) 61 (85) 80 (113) 72 (108) 40 (51)

2,902 3,962 1,053 21,374 15,711 24,897 22,030 9,395

0.39 0.54* 0.68* 0.24* 0.14* 0.26* 0.34* 0.50*

120.76* 130.64* 59.43* 397.80* 157.77* 623.12* 650.21* 445.80*

Note. k = number of studies; ES = number of effect sizes; N = sample size; d = weighted mean effect size; Q = test of heterogeneity. *p < 0.05.

incoming university students. Prior research had identified three prominent beliefs related to binge drinking: that it is fun, that it has a negative impact on studies, and that binge drinking by friends encourages this behavior. Based on extensive formative research, a message addressing these three beliefs was formulated and delivered online. The message explained that binge drinking has a negative impact on academic outcomes, that it is possible to have fun in other ways, and that most students do not binge drink on a regular basis. The message also mentioned the financial cost of heavy drinking. Three weeks prior to entering the university, participants reported their alcohol consumption, they were or were not exposed to the persuasive message, and they then completed a TPB questionnaire with regard to binge drinking. Alcohol consumption was assessed again at one week, one month, and six months after starting university and the TPB measures were reassessed at one and six months. Comparison of the intervention condition to the no-message control group showed that the intervention had its intended effects on the TPB variables: it influenced the targeted beliefs in the advocated direction and it produced less favorable experiential and instrumental attitudes toward binge drinking, lower injunctive and descriptive subjective norms, lower perceived capacity (but

not perceived agency), and lower intentions to binge drink. The intervention also reduced the amount of alcohol consumed and, although the effect on behavior diminished over time, it was still significant at the six-month follow-up. Finally, consistent with the TPB, a mediation analysis showed that the effect of the intervention on behavior was fully mediated by the TPB variables.

2.6 Summary and Conclusions It can be difficult to bring about new patterns of behavior, and interventions intended to do so can be costly and time-consuming; their design and implementation should not be left to intuition. Reliance on the TPB can provide information about the factors that motivate a behavior of interest and about the resources and barriers that can influence the adoption of a new course of action. According to the TPB, an intervention must first and foremost be designed to influence behavioral, normative, and/or control beliefs regarding the advocated behavior. To the extent that the intervention manages to produce substantial changes in these beliefs, corresponding changes in attitudes and subjective norms can be expected, changes that raise motivation to perform the behavior. In turn, increased motivation will produce an intention to engage in the behavior to the extent that

28

ICEK AJZEN AND PETER SCHMIDT

perceived behavioral control is sufficiently strong. Finally, the favorable intention will be implemented to the extent that the actor has volitional control over the behavior in question. The TPB provides a general framework for planning a behavior change intervention, but there is plenty of room for the investigator’s imagination and creativity. If the intervention is to increase motivation to perform a recommended behavior, the TPB can help to identify the behavioral, normative, and/or control beliefs that should be targeted. On the other hand, if people are already motivated and intend to change their behavior, the theory can direct the design of an intervention that will enable them to carry out their intentions. In either case, it is up to the investigator to design an effective intervention strategy that can produce the desired changes in the targeted variables. These are the action hypotheses not contained in the theory itself (Chen, 2014). They can be derived from other theories, from the results of past interventions, or by consulting practitioners and stakeholders, as proposed by Lewin (1946) and Campbell (1969). Adequate formative research guided by the TPB that incorporates these action hypotheses can increase the likelihood of a successful behavior change intervention.

References Ajzen, I. (1985). From intentions to actions: A theory of planned behavior. In J. Kuhl & J. Beckman (Eds.), Action-Control: From Cognition to Behavior (pp. 11–39). Berlin: Springer. https://doi.org/10 .1007/978-3-642-69746-3/2 Ajzen, I. (1988). Attitudes, Personality, and Behavior. Chicago: Dorsey Press. Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50, 179–211. https://doi.org/10.1016/ 0749-5978(91)90020-T Ajzen, I. (2012a). The theory of planned behavior. In P. A. M. Lange, A. W. Kruglanski, & E. Tory Higgins (Eds.), Handbook of Theories of Social Psychology, Vol. 1 (pp. 438–459). London: SAGE. https://doi.org/10.4135/9781446249215

Ajzen, I. (2012b). Attitudes and persuasion. In K. Deaux & M. Snyder (Eds.), The Oxford Handbook of Personality and Social Psychology (pp. 367–393). New York: Oxford University Press. https://doi.org /10.1093/oxfordhb/9780195398991.001.0001 Ajzen, I. (2019). Sample TPB Questionnaire. https:// people.umass.edu/aizen/pdf/tpb.questionnaire.pdf Ajzen, I., & Dasgupta, N. (2015). Explicit and implicit beliefs, attitudes, and intentions: The role of conscious and unconscious processes in human behavior. In P. Haggard & B. Eitam (Eds.), The Sense of Agency (pp. 115–144). New York: Oxford University Press. https://doi .org/10.1093/acprof:oso/9780190267278 .001.0001 Ajzen, I., & Fishbein, M. (2000). Attitudes and the attitude-behavior relation: Reasoned and automatic processes. In W. Stroebe & M. Hewstone (Eds.), European Review of Social Psychology, Vol. 11 (pp. 1–33). Chichester: Wiley. https://doi.org/10 .1080/14792779943000116 Ajzen, I., & Sexton, J. (1999). Depth of processing, belief congruence, and attitude-behavior correspondence. In S. Chaiken & Y. Trope (Eds.), Dual-Process Theories in Social Psychology (pp. 117–138). New York: Guilford Press. Albarracín, D., Fishbein, M., & Goldestein de Muchinik, E. (1997). Seeking social support in old age as reasoned action: Structural and volitional determinants in a middle-aged sample of Argentinean women. Journal of Applied Social Psychology, 27, 463–476. https://doi.org/10.1111 /j.1559-1816.1997.tb00642.x Angrist, D. J., & Pischke, J. S. (2015). Mastering Metrics: The Path from Cause to Effect. Princeton: Princeton University Press. Armitage, C. J., & Arden, M. A. (2002). Exploring discontinuity patterns in the transtheoretical model: An application of the theory of planned behaviour. British Journal of Health Psychology, 7, 89–103. https://doi.org/10.1348/135910702169385 Armitage, C. J., & Conner, M. (1999). The theory of planned behaviour: Assessment of predictive validity and “perceived control.” British Journal of Social Psychology, 38, 35–54. https://doi.org/10 .1348/014466699164022 Bamberg, S. (2006). Is a residential relocation a good opportunity to change people’s travel behavior?

Changing Behavior Using the Theory of Planned Behavior

Results from a theory driven intervention study. Environment and Behavior, 38, 820–840. https:// doi.org/10.1177/0013916505285091 Bamberg, S. (2013). Applying the stage model of self-regulated behavioral change in a car use reduction intervention. Journal of Environmental Psychology, 33, 68–75. https://doi.org/10.1016/j .jenvp.2012.10.001 Bamberg, S., & Schmidt, P. (1999). Regulating transport: Behavioral changes in the field. Journal of Consumer Policy, 22, 479–509. https://doi.org /10.1023/a:1006391723214 Bamberg, S., & Schmidt, P. (2001). Theory driven subgroup specific evaluation of an intervention to reduce car use. Journal of Applied Social Psychology, 31, 1300–1329. https://doi.org/10 .1023/a:1006391723214 Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84, 191–215. https://doi.org/10.1037 //0033-295x.84.2.191 Bandura, A. (1997). Self-Efficacy: The Exercise of Control. New York: Freeman. Bickman, B. (1987). Using Program Theory in Evaluation. San Francisco: Jossey-Bass. Brown, T. A. (2015). Confirmatory Factor Analysis for Applied Research (2nd ed.). New York: Guilford Press. Campbell, D.T. (1969). Reforms as experiments. American Psychologist, 24, 409–429. https://doi .org/10.1037/h0027982 Campbell, D. T., & Fiske, D. W. (1959). Convergent and discriminant validation by the multitraitmultimethod matrix. Psychological Bulletin, 56, 81–105. Chen, H. T. (2014). Practical Program Evaluation: Theory Driven Evaluation and the Integrated Evaluation Perspective. London: SAGE. Cialdini, R. B., & Trost, M. R. (1998). Social influence: Social norms, conformity and compliance. In D. T. Gilbert, S. T. Fiske, & G. Lindzey (Eds.), The Handbook of Social Psychology, Vols. 1–2 (4th ed., pp. 151–192). New York: McGraw-Hill. Coryn, C. L. S., Noakes, L. A., Westine, C. D., & Schröter, D. C. (2011). A systematic review of theory-driven evaluation practice from 1990 to 2009. American Journal of Evaluation, 32, 199–226. https://doi.org/10.1177/1098214010389321

29

de Leeuw, A., Valois, P., Ajzen, I., & Schmidt, P. (2015). Using the theory of planned behavior to identify key beliefs underlying pro-environmental behavior in high-school students: Implications for educational interventions. Journal of Environmental Psychology, 42, 128–138. https:// doi.org/10.1016/j.jenvp.2015.03.005 Dillard, J., & Shen, L. (2012). The SAGE Handbook of Persuasion: Developments in Theory and Practice (2nd ed.). Thousand Oaks, CA: SAGE. https://doi.org/10.4135/9781452218410 Elliott, M. A., & Armitage, C. J. (2009). Promoting drivers’ compliance with speed limits: Testing an intervention based on the theory of planned behaviour. British Journal of Psychology, 100, 111–132. https://doi.org/10.1348 /000712608X318626 Fishbein, M., & Ajzen, I. (2010). Predicting and Changing Behavior: The Reasoned Action Approach. New York: Psychology Press. Fisher, J. D., & Fisher, W. A. (1992). Changing AIDS-risk behavior. Psychological Bulletin, 111, 455–474. https://doi.org/10.1037//0033-2909 .111.3.455 Green, S. B., & Thompson M. S. (2015). A flexible structural equation modeling approach for analyzing means. In R. Hoyle (Ed.), Handbook of Structural Equation Modeling (pp. 393–416). New York: Guilford Press. Hagger, M. S., Chatzisarantis, N. L. D., & Biddle, S. J. H. (2002). A meta-analytic review of the theories of reasoned action and planned behavior in physical activity: Predictive validity and the contribution of additional variables. Journal of Sport and Exercise Psychology, 24, 3–32. https://doi.org/10.1123/jsep.24.1.3 Hancock, G. (2004). Experimental, quasiexperimental, and nonexperimental design and analysis with latent variables. In D. Kaplan (Ed.), The SAGE Handbook of Quantitative Methodology for the Social Sciences (pp. 318–335). Thousand Oaks, CA: SAGE. https://doi .org/10.4135/9781412986311.n17 Hovland, C. I., Janis, I. L., & Kelley, H. H. (1953). Communication and Persuasion: Psychological Studies of Opinion Change. New Haven, CT: Yale University Press. https://doi.org/10.2307 /2087772

30

ICEK AJZEN AND PETER SCHMIDT

Hukkelberg, S. S., Hagtvet, K. A., & Kovac, V. B. (2014). Latent interaction effects in the theory of planned behaviour applied to quitting smoking. British Journal of Health Psychology, 19, 83–100. https://doi.org/10.1111/bjhp.12034 Kelley, K., & Abraham, C. (2004). RCT of a theory-based intervention promoting healthy eating and physical activity amongst out-patients older than 65 years. Social Science and Medicine, 59, 787–797. https:// doi.org/10.1016/j.socscimed.2003.11.036 La Barbera, F., & Ajzen, I. (2018). Control interactions in the theory of planned behavior: Re-thinking the role of subjective norm. Unpublished manuscript. Lewin, K. (1946). Action research and minority problems. Journal of Social Issues, 2, 34–46. https:// doi.org/10.1111/j.1540-4560.1946.tb02295.x Lewin, K. (1952). Field Theory in Social Science: Selected Theoretical Papers by Kurt Lewin. London: Tavistock. Macy, J. T., Middlestadt, S. E., Seo, D.-C., Kolbe, L. J., & Jay, S. J. (2012). Applying the theory of planned behavior to explore the relation between smoke-free air laws and quitting intentions. Health Education and Behavior, 39, 27–34. https://doi.org/10.1177/1090198111404702 McDermott, M. S., Oliver, M., Svenson, A. et al. (2015). The theory of planned behaviour and discrete food choices: a systematic review and meta-analysis. International Journal of Behavioral Nutrition and Physical Activity, 12. https://dx .doi.org/10.1186%2Fs12966-015-0324-z Meitinger, K., & Behr, D. (2016). Comparing cognitive interviewing and online probing: Do they find similar results? Field Methods, 28, 363–380. https://doi.org/10.1177/1525822X15625866 Michie, S., & Prestwich A. (2010). Are interventions theory based? Development of a theory coding scheme. Health Psychology, 29, 1–8. https://doi .org/10.1037/a0016939 Michie, S., van Stralen, M. M., & West, R. (2011). The behaviour change wheel: A new method for characterising and designing behaviour change interventions. Implementation Science, 6. https:// doi.org/10.1186/1748-5908-6-42 Norman, P., Cameron, D., Epton, T. et al. (2018). A randomized controlled trial of a brief online intervention to reduce alcohol consumption in

new university students: Combining self-affirmation, theory of planned behaviour messages, and implementation intentions. British Journal of Health Psychology, 23, 108–127. https://doi.org/10.1111/bjhp.12277 Prochaska, J. O., & DiClemente, C. C. (1983). Stages and processes of self-change of smoking: Toward an integrative model of change. Journal of Consulting and Clinical Psychology, 51, 390–395. https://doi.org/10.1037//0022-006x.51.3.390 Prochaska, J. O., & DiClemente, C. C. (1992). Stages of change in the modification of problem behaviors. In M. Hersen, R. M. Eisler, & P. M. Miller (Eds.), Progress in Behavior Modification (pp. 184–218). Sycamore, IL: Sycamore Press. Riebl, S. K., Estabrooks, P. A., Dunsmore, J. C. et al. (2015). A systematic literature review and metaanalysis: The theory of planned behavior’s application to understand and predict nutritionrelated behaviors in youth. Eating Behaviors, 18, 160–178. https://doi.org/10.1016/j.eatbeh.2015 .05.016 Rodgers, J. L. (2010). The epistemology of mathematical and statistical modeling: A quiet methodological revolution. American Psychologist, 65, 1–12. https://doi.org/10.1037 /a0018326 Rossi, P. H., Lipsey, M. W., & Freeman, H. E. (2004). Evaluation: A Systematic Approach (7th ed.). Thousand Oaks, CA: SAGE. Shaddish, W. R., Cook T. D., & Campbell, D. T. (2002). Experimental and Quasi-Experimental Designs for Generalized Causal Inference. Belmont, CA: SAGE. Sheeran, P., Maki, A., Montanaro, E. et al. (2016). The impact of changing attitudes, norms, and self-efficacy on health-related intentions and behavior: A meta-analysis. Health Psychology, 35, 1178–1188. https://doi.org/10.1037 /hea0000387 Sheeran, P., & Taylor, S. (1999). Predicting intentions to use condoms: A meta-analysis and comparison of the theories of reasoned action and planned behavior. Journal of Applied Social Psychology, 29, 1624–1675. https://doi.org/10.1111/j.1559 -1816.1999.tb02045.x

Changing Behavior Using the Theory of Planned Behavior

Sniehotta, F. (2009). An experimental test of the theory of planned behavior. Applied Psychology: Health and Well-Being, 1, 257–270. https://doi.org/10 .1111/j.1758-0854.2009.01013.x Stark, J., Berger, W. J., & Hössinger, R. (2018). The effectiveness of an intervention to promote active travel modes in early adolescence. Transportation Research, Part F, 55, 389–402. https://doi.org/10 .1016/j.trf.2018.03.017 Stecker, T., McGovern, M. P., & Herr, B. (2012). An intervention to increase alcohol treatment engagement: A pilot trial. Journal of Substance Abuse Treatment, 43, 161–167. https://doi.org/10 .1016/j.jsat.2011.10.028 Steinmetz, H., Knappstein, M., Ajzen, I., Schmidt, P., & Kabst, R. (2016). How effective are behavior change interventions based on the theory of planned behavior? A three-level meta-analysis. Zeitschrift Für Psychologie, 224, 216–233. https://doi.org/10.1027/2151-2604/a000255 Trafimow, D. (2012). The role of auxiliary assumptions for the validity of manipulations and measures. Theory and Psychology, 22, 486–498. https://doi .org/10.1177/0959354311429996 Triandis, H. C. (1972). The Analysis of Subjective Culture. New York: Wiley.

31

Tyson, M., Covey J., & Rosenthal, H. E. (2014). Theory of planned behavior interventions for reducing heterosexual risk behaviors: A meta-analysis. Health Psychology, 33, 1454–1467. https://doi.org /10.1037/hea0000047 Webb, T. L., Joseph, J., Yardley, L., & Michie, S. (2010). Using the internet to promote health behavior change: A systematic review and meta-analysis of the impact of theoretical basis, use of behavior change techniques, and mode of delivery on efficacy. Journal of Medical Internet Research, 12, 1–27. https://doi.org/10.2196/jmir .1376 Webb, T. L., & Sheeran, P. (2006). Does changing behavioral intentions engender behavior change? A meta-analysis of the experimental evidence. Psychological Bulletin, 132, 249–268. https://doi .org/10.1037/0033-2909.132.2.249 Winkelnkemper, P., Ajzen, I., & Schmidt, P. (2019). The theory of planned behavior: A meta-analysis using structural equation modeling. Unpublished manuscript. Yzer, M., & van den Putte, B. (2014). Control perceptions moderate attitudinal and normative effects on intention to quit smoking. Psychology of Addictive Behaviors, 28, 1153–1161.

3

Changing Behavior Using Social Cognitive Theory Aleksandra Luszczynska and Ralf Schwarzer

Practical summary The key strength of social cognitive theory for practice lies in targeting two constructs: self-efficacy and outcome expectancies. Facilitating change in these constructs in interventions has been shown to be effective in changing a large array of behavioral outcomes. The theory has been applied extensively and has demonstrated effectiveness. It proposes a process that describes behavioral initiation, effort expenditure, and perseverance. High self-efficacy predicts engagement in a range of social and health behaviors. In addition, intervention studies indicate that increasing self-efficacy leads to improvements in behavior (e.g., by providing opportunities to experience success with the behavior, reflecting on past success, and presenting role models to provide exemplars of success with the behavior). Taken together, these findings indicate that individuals with strong self-efficacy beliefs are more likely to undertake behaviors (e.g., performing regular physical activity) and achieve desired outcomes (e.g., weight loss). Less is known about the roles of barriers and facilitating factors. For the practitioner, this makes theory- and evidence-based practice challenging, as the research evidence mainly supports the effects of single constructs such as self-efficacy, outcome expectancies, and goals in changing behavior without demonstrating the effects of the theory as a whole.

3.1 Introduction Social cognitive theory (Bandura, 1986) has become a fundamental framework in social, clinical, educational, developmental, health, and personality psychology to explain human behavior in multiple domains, including education, mental health, physical health, sport, career, and developmental tasks. The origins of the theory date back to the 1970s when a paradigm shift occurred in the discipline of psychology from a focus on behavior to a focus on cognition and social learning processes. Social cognitive theory evolved out of the earlier social learning theory (Bandura, 1977a) and may be considered

a response to behaviorism (Hull, 1943; Skinner, 1938) and a foundation for other theoretical developments, currently known as the social cognitive approach to understanding behavior. The theory has also been a leading approach to guiding interventions aimed at changing behavior. Social cognitive theory retains two of the fundamental assumptions of social learning theory: people learn by watching others’ behaviors and behaviors are learned in social contexts. In addition, it assumes that the maintenance of behaviors over time requires an environmental reinforcement and individual https://doi.org/10.1017/9781108677318.003

Changing Behavior Using Social Cognitive Theory

self-regulation. The theory also emphasizes the concept of reciprocal determinism in which individual factors (e.g., self-efficacy, behavioral responses) and environmental factors (e.g., facilitating conditions) are proposed to affect each other reciprocally. Two constructs are central to the predictions of social cognitive theory: self-efficacy and outcome expectancies. These constructs operate together with goals, socio-structural impediments, and facilitators in determining behavior. This chapter outlines how social cognitive theory has been utilized to change behavior. It begins by providing an overview of the theory and outlining the key constructs and processes that link together. The role of the theory in understanding behavior in multiple contexts (e.g., organizational, educational, sport, and health) is presented. This is followed by an overview of how the theory has been used to change behavior and the empirical evidence providing support for its application across contexts.

3.2 Overview of Social Cognitive Theory and Empirical Evidence According to social cognitive theory, human motivation and action are extensively regulated

by forethought (Bandura, 1986, 2000a, 2000b, 2001). The theory outlines a number of key constructs that influence behavior (see Figure 3.1). The first construct is self-efficacy, which represents people’s beliefs in their capabilities to perform an action required to attain the desired outcome. The second core construct is outcome expectancies, which represents people’s beliefs about the possible consequences of their actions. Self-efficacy and outcome expectancies are generally regarded as the central constructs of the theory. The theory also includes goals, perceived impediments, and facilitators. Perceived self-efficacy refers to individuals’ beliefs in their capability to exercise control over challenging demands and their own functioning (Bandura, 1977b, 1997). It reflects a subjective estimate of the amount of personal control an individual expects to have in any given situation. As a result, self-efficacy beliefs are often labeled “I can” perceptions regarding future behaviors or tasks. If individuals hold strong beliefs that they can master an upcoming task, they are likely to invest the effort in engaging with it. Similarly, if an individual feels confident that they can overcome

Outcome Expectancies: • Physical • Social • Self-Evaluative

Perceived Self-Efficacy

Goals

33

Behavior

Sociostructural Factors: • Facilitators • Impediments

Figure 3.1 An illustration of social cognitive theory (Bandura, 2000a)

34

ALEKSANDRA LUSZCZYNSKA AND RALF SCHWARZER

an imminent threat or challenge (e.g., an exam), they are more likely to approach rather than to avoid the threat. Self-efficacy, therefore, affects the amount of effort individuals will expend to change a particular behavior to obtain a particular goal and the extent to which they will persist in striving to attain that goal in the face of barriers and setbacks that may undermine motivation. There are a number of interpersonal and environmental sources that give rise to self-efficacy (Bandura, 1997). First, self-efficacy can be enhanced through previous experiences of personal accomplishment or mastery of behaviors or tasks, insofar as success is attributed internally and is perceived to be repeatable. A second source is vicarious experience. When a model similar in ability to the individual is viewed by the individual as successfully mastering a behavior or task in a difficult situation, social comparison processes are likely to enhance self-efficacy. Third, selfefficacy may be enhanced through verbal persuasion by others (e.g., a coach reassures an athlete that she can adopt a new training routine due to her competence). The last source of influence is emotional arousal, that is, if the person experiences no apprehension in a threatening situation, they may feel capable of mastering the situation. The four sources of self-efficacy vary in strength and importance in the order presented here, with personal mastery being the strongest source of self-efficacy (Warner et al., 2014). There is considerable evidence supporting the premises of social cognitive theory, particularly links between strategies targeting the four sources of self-efficacy. In the context of physical activity, meta-analytic syntheses of research adopting the theory have compared the effects of various intervention techniques that tap into the four sources of self-efficacy on self-efficacy and behavioral outcomes (Ashford, Edmunds, & French, 2010; Williams & French, 2011). Effective self-efficacy enhancement was observed in physical activity interventions that used vicarious experience, feedback referring to participants’ past

performances or past performances of others (Ashford et al., 2010). Similarly, a meta-analysis in the context of dietary interventions has identified behavior change techniques that result in improved dietary self-efficacy (Prestwich et al., 2013). In other areas of human functioning, metaanalyses of the relationship between work performance and self-efficacy (Stajkovic & Luthans, 1998) as well as achievement goals and selfefficacy (Huang, 2016) have been conducted. While perceived self-efficacy refers to personal action control or agency, outcome expectancies pertain to the perception of possible consequences of one’s action (Bandura, 1997). Outcome expectancies can be organized along three dimensions: (1) area of consequences, (2) positive or negative consequences, and (3) shortterm or long-term consequences. In regard to the first dimension, there are three areas of consequences for outcome expectancies: physical, social, and self-evaluative. Physical outcome expectations, such as expectations of discomfort, refer to the anticipation of what will be experienced after behavior change takes place. These include both the short- and long-term effects of behavior change. For example, immediately after quitting smoking, an ex-smoker might observe a reduction in coughing (positive consequence) and a higher level of muscle tension (negative consequence). Social outcome expectancies refer to anticipated social responses after behavior change. Smokers might expect disapproval from friends who continue to smoke or, positively, they might expect their family to congratulate them on quitting smoking. Self-evaluative outcome expectations refer to the anticipation of affective experiences, such as being ashamed or being proud of oneself. For example, smokers might anticipate that quitting smoking will lead to a sense of satisfaction and pride after they have done so. Different types of outcome expectancies have been found to be important determinant in research on health behaviors, such as dietary change (Hankonen et al., 2013).

Changing Behavior Using Social Cognitive Theory

Both outcome expectancies and self-efficacy beliefs are regarded as important factors that can directly have an impact on goals and behaviors. They also operate through indirect pathways, affecting goal setting and the perception of sociostructural factors. In adopting a desired behavior, individuals form a goal and then attempt to execute the action. Goals serve as self-incentives and guide decisions/motivation to participate in behaviors (see Chapter 38, this volume). According to social cognitive theory, a distinction can be made between distal and proximal goals. A distal goal might be weight loss, whereas a proximal goal might be an increase in physical activity and fiber consumption. The proximal ones regulate the amount of invested effort in pursuing a goal and guide subsequent action. Intentions, as defined in other social cognitive approaches, are similar to proximal goals (Bandura, 1997). Several major theories agree that goals or intentions should be as specific as possible in order to facilitate subsequent action (e.g., Bandura, 1997; Fishbein & Ajzen, 2010; see Chapters 2 and 4, this volume). Distal goals give purpose and general direction to actions. For example, an individual may aim at becoming a medical doctor, which is a distal goal that can be subdivided into a hierarchy of proximal goals such as striving for a medical exam, studying hard for some years, and signing up for a premed course. According to social cognitive theory, forming a goal is a necessary but not sufficient condition for goal pursuit and behavioral enactment; it is a precondition but does not ensure that an individual will actually pursue the goal (Bandura, 2000a, 2000b). People would not set goals for themselves if they thought that the pursuit of such goals would have more disadvantages than advantages (Bandura, 2000a, 2000b). If learning to play tennis has the advantage of being more active and fit, this consequence might be outperformed by the expectation that participating in the sport is resource-demanding (e.g., it demands time and money) and that one requires a partner

35

who is available when needed. Thus positive, as well as negative, outcome expectancies are seen as important determinants in the formation of goals. Self-efficacy, on the other hand, remains crucial for the formation of an intention to adopt a behavior as well as for the post-intentional phase when the task is to translate the intention into action and to self-regulate the goal pursuit process. Goal setting also depends on perceived sociostructural factors. These factors refer to the impediments (barriers) or opportunities (facilitators) that reside in living conditions, personal circumstances, and political, economic, or environmental systems (Bandura, 1997). Optimistic selfbeliefs about one’s efficacy to change behaviors may also control how people perceive opportunities and impediments. An individual’s level of self-efficacy influences whether they will attend to opportunities or barriers in their life circumstances. Self-efficacious individuals who, for example, intend to exercise might focus on cues in their environment, such as hiking paths and cycling routes. Those who are less confident about their physical competence might focus instead on the lack of a gym in their neighborhood. People with strong self-efficacy recognize that they are able to overcome obstacles and focus on opportunities. They believe that they are able to exercise control, even if the environment provides constraints rather than opportunities (see Bandura, 1997, 2000a, 2000b). For example, a meta-analysis of 257 original studies showed that self-efficacy and positive outcome expectancies (e.g., expectations referring to potential rewards) were the strongest predictors of digital piracy behavior, whereas other variables referring to social learning models (e.g., social influence) or moral disengagement produced smaller effects (Lowry et al., 2017). Similarly, a meta-analysis of 143 studies suggested that social cognitive theory was effective in predicting whether or not people chose mathand science-intensive academic courses and

36

ALEKSANDRA LUSZCZYNSKA AND RALF SCHWARZER

majors (Lent et al., 2018). The career-choice behaviors are directly predicted by self-efficacy, goals, facilitating factors (support), and barriers (Lent et al., 2018). Outcome expectancies by themselves may have no direct effects yet they are strongly related to goals, hence they may operate indirectly. Although significant among ethnic majorities and minorities, the associations may be moderated by ethnicity or gender – for example, outcome expectancies were more strongly related to goals among people from ethnic minorities and women compared to ethnic majority participants and men (Lent et al., 2018). Systematic reviews seeking the strongest predictors and correlates of adherence to combination antiretroviral therapy (cART) for people living with HIV indicated that adherence selfefficacy was the strongest predictor of adherence (Langebeek et al., 2014). The associations between other correlates (e.g., social support) and adherence were weaker. Similarly, a systematic review of social cognitive theory in the physical activity context yielded estimates of average-level effects, based on forty-four original studies (Young et al., 2014). Across included studies, variation in physical activity was directly explained by self-efficacy (60 percent of studies yielded significant associations) and goals/intentions (86 percent of analyzed studies). Direct effects of other constructs from the theory were found in less than half of studies. In particular, the evidence for direct effects of outcome expectancies was found in only 30 percent of studies, the effects of impediments were found in only 24 percent of studies, and effects of social support were observed in 20 percent of studies (Young et al., 2014). The associations were more likely to occur in older samples (yet it has to be highlighted that the review included research in studies with samples of children as young as nine years old). In summary, two constructs are central to the predictions of social cognitive theory: selfefficacy and outcome expectancies – and that they operate together with goals, socio-

structural impediments, and facilitators in codetermining cognitions, emotions, and behavior. In particular, self-efficacy is a meaningful and changeable belief that is important for initiating and maintaining behaviors. When self-efficacy is low, taking steps to increase it can help people to change their behaviors. Interventions are mainly based on four sources of information that give rise to self-efficacy: mastery experience, vicarious experience, verbal persuasion, and physiological feedback. Social cognitive theory has been applied to explain cognition and behaviors across various behavioral domains, including health behavior, work performance, academic achievement, sexual functioning, and coping with stress. Besides the use of the original version of the theory, researchers have also developed several adapted versions of the theory that provide more effective explanations of behavior in specific contexts (see Sidebar 3.1).

3.3 How Has Social Cognitive Theory Been Used to Change Behavior? Social cognitive theory has been used to guide interventions to change behavior in many contexts and populations. Research applying the theory has utilized a number of intervention strategies or behavior change techniques targeting change in the key constructs of the theory (see Chapter 20, this volume). Abraham and Michie (2008) provided a formal classification of seven basic behavior change techniques that may modify social cognitive theory constructs: (1) modeling or demonstration of behavior; (2) provision of instructions on how to perform a behavior; (3), provision of general encouragement; (4) prompting barrier identification; (5) setting graded tasks; (6) prompting intention formation; and (7) provision of information on consequences. The latter two techniques target constructs that are also present in other theories (e.g., the information-motivation skills model and the theory of planned behavior; see Chapter 2, this

Changing Behavior Using Social Cognitive Theory

Sidebar 3.1 Specific versions of social cognitive theory

One development of social cognitive theory has been in the area of human adaptation, such as the explanation of posttraumatic adaptation (Benight & Bandura, 2004). Traumatic stress survivors with high self-efficacy beliefs have been found to better adhere to physical and mental health treatments and engage in less substance abuse–related behaviors (Luszczynska, Benight, & Cieslak, 2009). In line with this approach, self-efficacy is a focal resource enabling individuals to effectively search for the support needed for recovery. It also allows traumatic stress survivors to regain control over one’s own behaviors after traumatic events. Another application is in the form of social cognitive career theory, which aims to explain the processes by which individuals develop their interests, make choices, and attain performance in work-related and academic contexts (Brown et al., 2008; Lent et al., 2003). The model stresses the role of self-efficacy, outcome expectancies, and goals operating jointly in predicting achievements and performance. Furthermore, social cognitive career theory highlights the role of barriers and supports (i.e., impediments and facilitating factors), representing a perception of specific environmental influences (Lent et al., 2003). Employees and students with stronger self-efficacy and outcome expectancies will tend to strive toward more challenging job or academic goals than will those with weaker efficacy beliefs or less positive outcome expectancies. More challenging or difficult goals are, in turn, assumed to motivate employees/students to work harder toward goal fulfillment, leading to more favorable academic or work-related outcomes. Meta-analyses have confirmed that persistence (i.e., continuous efforts to attain an academic degree) is indirectly explained by self-efficacy, with educational goals playing a mediating role (Brown et al., 2008; Unrau et al., 2018). A similar pattern of associations between self-efficacy, goals, and indicators of work performance (e.g., training participation, sales, absenteeism) has been found (Brown et al., 2008). A theoretical application of social cognitive theory was also developed to explain leadership behaviors (McCormick, 2001). According to this approach (McCormick, 2001), goals and self-efficacy affect leadership motivation and task strategy development, which, in turn, explain leadership behaviors in specific performance environments. Sport and athletic performance constitutes yet another context in which theoretical applications of social cognitive theory have been made. Feltz et al. (1999; for a meta-analysis, see Myers et al., 2017) proposed a conceptual model of coaching efficacy consisting of three key elements: sources of coaching efficacy information, dimensions of coaching efficacy, and outcomes of a coach’s efficacy beliefs. The sources included the extent of coaching experience/preparation, prior success, the perceived skill of athletes, and social support. The proposed dimensions of coaching efficacy were character building, game strategy, motivation, and technique, whereas the outcomes of coaching efficacy were coaching behavior, player/team satisfaction, player/team performance, and player/team efficacy.

37

38

ALEKSANDRA LUSZCZYNSKA AND RALF SCHWARZER

volume). It is also important to give people opportunities to try out the behavior and experience early success. However, this is not a unique issue as many techniques may affect change in constructs from multiple theories. Interventions focused on social cognitive theory primarily target change in self-efficacy and outcome expectancies, given the pervasive effects of these constructs on predicting behavioral outcomes in formative research applying the theory across contexts and populations (Prestwich et al., 2014). For example, interventions may include strategies designed to increase participants’ sense of mastery and their ability to handle difficult situations that might arise during the initiation or maintenance of a novel or difficult behavior. Such strategies may lead to increased self-efficacy beliefs when it comes to initiating or persisting with the behavior when barriers arise (Tang et al., 2019). In addition, providing opportunities to experience success with a target behavior of interest might be considered, such as reflecting on past success with the behavior, developing skills such as academic learning strategies or coping techniques, setting proximal goals, monitoring goal progress, providing supportive feedback on progress toward goals, enhancing skills to manage setbacks such as resetting goals or reattribution of failure, and presenting role models to provide vicarious experience of success with the behavior (Tang et al., 2019; see also Chapter 32, this volume). Self-efficacy beliefs can improve through personal accomplishment (Bandura, 1997). Thus, to foster mastery experiences, one can guide clients to perform small steps that are likely to be achieved successfully. One can then provide positive feedback to reinforce this mastery experience and encourage the person to master subsequently more challenging steps. Such graded tasks can be useful in clinical settings such as in physiotherapy (e.g., gradual progression of balance and strength exercises; Ghazi et al., 2018) or cognitive behavior therapy for phobias (Bandura, 1977b). In addition, vicarious experiences provide individuals with

a “role model” that provides a “behavioral template” for the behavior and affects estimates of future confidence to do the behavior and, hence, individuals may form motives or intentions to do the behavior in future. When individuals witness other people (similar to themselves) successfully master a difficult situation, social comparison and imitation of the behavior can strengthen selfefficacy beliefs (Bandura, 1997). Similarly, imagery may enhance self-efficacy by presenting individuals with a “self-model” of the behavior (see also Chapter 33, this volume). Self-efficacy beliefs can also be changed through verbal persuasion. For example, a teacher could reassure students that they can prepare for a demanding exam, due to their effort, competence, and ability to plan. They could be informed that they have what it takes to succeed in anything they put their efforts into. These types of persuasion can strengthen selfefficacy for successfully managing the task at hand (Unrau et al., 2018).

3.4 Evidence Base for the Use of Social Cognitive Theory in Changing Behavior Social cognitive theory offers a sound theoretical basis for informing interventions to change a wide range of behaviors. Research has demonstrated that interventions adopting strategies and techniques to change behavior from the theory has led to significant behavior changes in numerous contexts, including health-related behaviors (French et al., 2014), behaviors in occupational and organizational settings like career exploration and decision-making (Lent et al., 2016, 2017), behaviors in educational contexts like persistence with studying and attention in class (Schunk & Zimmerman, 2007; Unrau et al., 2018) or reading (Aro et al., 2018; Unrau et al., 2018), performance in sports (Te Velde et al., 2018), parenting (Liyana Amin et al., 2018), and breastfeeding (Galipeau et al., 2018). In addition, interventions based on the theory have demonstrated efficacy in changing

Changing Behavior Using Social Cognitive Theory

salient outcomes such as changes in health indicators like weight loss (Hays et al., 2014), productivity and attendance in occupational settings (Cherian & Jacob, 2013), academic attainment (Honicke, & Broadbent, 2016; Huang, 2016; Talsma et al., 2018), and criminal behavior such as digital piracy (acts of copyright infringement of electronic goods, including software, music, books, movies, TV shows, and games; Lowry, Zhang, & Wu, 2017) or cyberbullying perpetration (Chen, Ho, & Lwin, 2017). In the educational context, for example, modeling has been found to be an effective means of building selfregulatory and academic skills and of raising selfefficacy. In reading and writing interventions, modeling was employed to enhance self-efficacy, skills, and self-regulation (Aro et al., 2018; Schunk & Zimmerman, 2007; Unrau et al., 2018). However, it is in health contexts where social cognitive theory approaches have been most frequently applied to change a range of health-related behaviors (Davis et al., 2015). For example, in the domains of treatment and rehabilitation in the context of chronic illness, often new behaviors are required to be adopted (e.g., engaging in rehabilitation exercises, incorporating complex medication adherence regimes into daily routines). Selfefficacy has been shown to form moderate associations with such rehabilitation-related behaviors, with larger effects found for individuals with conditions that have not progressed to end-stage salvage surgery and for younger and more physically active individuals (Ghazi et al., 2018). A further example is the promotion of physical activity and nutrition, two behavioral domains in which social cognitive theory has been applied very frequently, and the large amount of intervention studies based on the theory in these domains presents challenges to synthesis. However, a review of 265 nutrition interventions based on the theory published in the 1980s and 1990s indicated that outcome expectancies or self-efficacy were included in about 90 percent of the studies (Contento, Randell, & Basch, 2002). The authors

39

conclude that changes in preferences or different attitudes seem to be of less importance than changes in self-efficacy or outcome expectancies when it comes to behavior change. Other systematic reviews have focused on the effects of social cognitive theory–based interventions on physical activity and nutrition in specific contexts – for example, cancer-related treatment and recovery processes (see Stacey et al., 2015). These authors’ metaanalysis of social cognitive theory–based interventions in this context, including twelve studies with cancer survivors, found favorable results in terms of the overall effectiveness of such interventions to improve dietary behaviors as well as physical activity. However, they were unable to indicate which of the theory constructs was most likely to change these behaviors and how this construct should be changed. Theory-based interventions were effective in improving physical activity assessed at postintervention with a small-to-medium effect size, and six out of eight trials showed a statistically significant improvement in at least one dietary outcome (Stacey et al., 2015). Self-efficacy was a primary construct targeted in the interventions, yet they were heterogeneous in terms of the number of theory constructs targeted and the behavior change techniques applied. Furthermore, analyses testing if particular social cognitive theory constructs changed or mediated the effects of the intervention on nutrition or physical activity were inconsistent. The majority of research on interventions or experimental studies based on social cognitive theory has focused on comparisons of the predictive role of self-efficacy, alongside effects of other constructs from either social cognitive theory or other theories (e.g., Jekauc et al., 2015; Luszczynska, Horodyska et al., 2016). For example, experimental studies tested whether the most conspicuous construct, self-efficacy – as opposed to the key variable from implementation intention theory, planning – exerts similar or larger effects on dietary changes (Luszczynska, Horodyska et al., 2016) or physical activity changes (Luszczynska, Hagger et al., 2016). For these

40

ALEKSANDRA LUSZCZYNSKA AND RALF SCHWARZER

behaviors, self-efficacy–enhancing interventions yielded similar effects as the interventions targeting planning. Importantly, combining selfefficacy with planning in a single intervention did not result in superior behavior changes than one of these components alone. Although there is a myriad of interventions using the underpinnings of social cognitive theory to change a behavior, it is unclear to what extent all the constructs assumed in the theory should be targeted. One of the key issues in translating theory to practice with respect to interventions based on the theory is which constructs to target and whether or not targeting more theory constructs yields stronger effects. A systematic review of the use of social cognitive theories to promote healthy eating and physical activity indicated that the interventions applying social cognitive theory more extensively did not result in larger effects than those that used theories less extensively (Prestwich et al., 2014). This, however, may be an effect that is not specific for social cognitive theory, as similar conclusions were drawn for the transtheoretical model (see Chapter 10, this volume), the other theory that was used in an adequate number of experimental trials providing data sufficient to conduct a systematic analysis. A possible future direction of research may be the search for moderator effects. For example, the effects of self-efficacy on goals and behavior may be moderated by risk perceptions. Heightened risk perception may trigger an intention to manage risk only when individuals have high selfefficacy to engage in suitable behaviors to minimize the challenges (Kok et al., 2018). Also, the level of goals may be moderated by subsequent self-regulatory factors such as planning. Both types of putative moderation effects may be promising hypotheses as the body of research on social cognitive theory expands. Self-efficacy and similar constructs have been incorporated into many theories and models such as the reasoned action approach (see Chapter 2,

this volume), the health action process approach (see Chapter 7, this volume), and a number of integrated approaches to behavior change (Chapter 15, this volume), which highlights the value of the theory going forward as a means to inform efforts to predict and change behavior.

3.5 Summary and Conclusion Social cognitive theory is one of the most wellcited and applied theories of behavior change. It has been applied to multiple behaviors performed across contexts and settings (e.g., education, occupational and workplace, mental and physical health, environmental settings; Aro et al., 2018; Cherian & Jacob, 2013; French et al., 2014; Hays et al., 2014; Lent et al., 2016, 2017; Schunk & Zimmerman, 2007; Talsma et al., 2018; Unrau et al., 2018; Te Velde et al., 2018). As a consequence, a substantive body of evidence has accumulated that has provided support for the application of the theory to predict and change behavior (Luszczynska & Schwarzer, 2015). Central to the theory are the constructs of selfefficacy and outcome expectancies, both of which have been shown to be reliably related to the initiation and maintenance of behavior in formative research applying social cognitive theory (see Chapter 32, this volume). Based on accumulated empirical evidence, interventions targeting these cognitions should translate into changes in behavioral outcomes. Successful social cognitive theory–based behavior change interventions may use such strategies as providing experiences of success with the behavior of interest, modeling successful performance of the behavior, providing instructions on how to perform a behavior successfully, using encouragement and positive feedback on performance, prompting identification of barriers and how to overcome them, assisting in setting graded tasks and appropriate goals, or providing information on the benefits and consequences of performing the behavior (see Table 3.1).

Changing Behavior Using Social Cognitive Theory

41

Table 3.1 Social cognitive theory behavior change techniques Behavior Change Technique

Example

(1) Modeling or demonstration Have clients witness other people (similar to themselves) successfully of behavior master a difficult situation; social comparison and imitation of the behavior can strengthen self-efficacy beliefs. (2) Provision of instructions on Point to skills and strategies on how to perform a particular task and gain how to perform a behavior mastery experience. (3) Provision of general Reassure clients that they can adhere to a demanding new behavior, due to encouragement their competence and ability to plan. Tell them that they have what it takes to succeed in anything they put their efforts into. (4) Prompting barrier Support clients to be aware of obstacles and make realistic judgments about identification coping options. (5) Setting graded tasks Guide clients to perform small steps that are likely to be achieved successfully. Such graded tasks can be useful in clinical settings such as in physiotherapy (e.g., gradual progression of balance and strength exercises). (6) Prompting intention Underscore level of competence that is required to set a realistic goal. formation (7) Provision of information on Identify positive and negative outcomes of various types of behavior. consequences

A possible weakness of the theory lies in the lack of consideration of emotion-related and nonconscious processes (Beauchamp et al., 2019; see also Chapters 12, 15, and 34, this volume). Moreover, the causal mechanism, the hypothesized interplay of constructs, is not clearly documented, making it difficult to derive clear and specific predictions. There is no detailed set of hypotheses or predictions that provides a definitive, comprehensive representation of the entire theory on which appropriate behavior change interventions can be developed. Some aspects of social cognitive theory that have not been clearly described or mapped out in detail and perhaps the theory should, therefore, be considered a framework to guide ideas on effects on behavior rather than a formal theory specifying particular predictions. Moreover, the construct of self-efficacy is often not well understood and researchers often use misleading operationalizations in their assessment tools, which may be one source of inconsistency in the empirical literature (e.g., Beauchamp, Crawford, & Jackson, 2019; Williams & Rhodes, 2016). For example, a suggested rule for the

behavior-specific assessment of self-efficacy is: “I am confident that I can . . . (perform an action), even if . . . (a barrier)” (Luszczynska & Schwarzer, 2015). An example of a self-efficacy statement is: “I am confident that I can skip dessert after meals even if my family continues to eat it.” In contrast, one finds assessment items such as “for me, learning Spanish is easy” in some studies. Such easy/difficult statements do not have a good fit with the definition of the self-efficacy construct (Schwarzer & Luszczynska, 2016). Future research should, therefore, seek to resolve these inconsistencies by using semantic rules for rational item construction and by clarifying the nature of the self-efficacy construct. It is also important to avoid the possible confound of self-efficacy and motivation (Williams & Rhodes, 2016) by targeting clearly different operationalizations of these two constructs. Moreover, future research should go beyond the mere analysis of linear direct effects of social cognitive theory constructs on behaviors and consider also more complex operating mechanisms of change as reflected by mediation or moderation effects.

42

ALEKSANDRA LUSZCZYNSKA AND RALF SCHWARZER

References Abraham, C., & Michie, S. (2008). A taxonomy of behavior change techniques used in intervention. Health Psychology, 27, 379–387. https://doi.org /10.1037/0278-6133.27.3.379 Aro, T., Viholainen, H., & Koponen, T. et al. (2018). Can reading fluency and self-efficacy of reading fluency be enhanced with an intervention targeting the sources of self-efficacy? Learning and Individual Differences, 67, 53–66. https://doi.org /10.1016/j.lindif.2018.06.009 Ashford, S., Edmunds, J., & French, D. P. (2010). What is the best way to change self-efficacy to promote lifestyle and recreational physical activity? A systematic review with meta-analysis. British Journal of Health Psychology, 15, 265–88. https:// doi.org/10.1348/135910709X461752 Bandura, A. (1977a). Social Learning Theory. Englewood Cliffs, NJ: Prentice-Hall. Bandura, A. (1977b). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84, 191–215. Bandura, A. (1986). Social Foundations of Thought and Action. Englewood Cliffs, NJ: Prentice Hall. Bandura, A. (1997). Self-Efficacy: The Exercise of Control. New York: Freeman. Bandura, A. (2000a). Cultivate self-efficacy for personal and organizational effectiveness. In E. A. Locke (Ed.), The Blackwell Handbook of Principles of Organizational Behavior (pp. 120–136). Oxford: Blackwell. Bandura, A. (2000b). Exercise of human agency through collective efficacy. Current Directions of Psychological Science, 9, 75–78. https://doi.org /10.1111/1467-8721.00064. Bandura, A. (2001). Social cognitive theory: An agentic perspective. Annual Review of Psychology, 52, 1–26. https://doi.org/10.1146/annurev.psych.52.1.1 Beauchamp, M. R., Crawford, L. L., & Jackson, B. (2019). Social cognitive theory and physical activity: Mechanisms of behavior change, critique and legacy. Psychology of Sport and Exercise, 42, 110–117. https://doi.org/10.1016/j.psychsport .2018.11.009 Benight, C. C., & Bandura, A. (2004). Social cognitive theory of posttraumatic recovery: The role of

perceived self-efficacy. Behavioral Research and Therapy, 42, 1129–1148. https://doi.org/10.1016/j .brat.2003.08.008 Brockway, M., Benzies, K., & Hayden, K. A. (2017). Interventions to improve breastfeeding self-efficacy and resultant breastfeeding rates: A systematic review and meta-analysis. Journal of Human Lactation, 33, 486–499. https://doi.org/10 .1177/0890334417707957 Brown, S. D., Tramayne, S., Hoxha, D. et al. (2008). Social cognitive predictors of college students’ academic performance and persistence: A metaanalytic path analysis. Journal of Vocational Behavior, 72, 298–308. https://doi.org/10.1016/j .jvb.2007.09.003 Chen, L., Ho, S. S., & Lwin, M. O. (2017). A meta-analysis of factors predicting cyberbullying perpetration and victimization: From the social cognitive and media effects approach. New Media and Society, 19, 1194–1213. https://doi.org/10 .1177/1461444816634037 Cherian, J., & Jacob, J. (2013). Impact of self-efficacy on motivation and performance of employees. International Journal of Business and Management, 8, 80–88. https://doi.org/10.5539 /ijbm.v8n14p80 Contento, I. R., Randell, J., & Basch, C. E. (2002). Review and analysis of education measures used in nutrition education intervention research. Journal of Nutrition Education and Behavior, 34, 2–25. Davis, R., Campbell, R. Hildon, Z., Hobbs, L., & Michie, S. (2015). Theories of behaviour and behaviour change across the social and behavioural sciences: A scoping review. Health Psychology Review, 9, 323–344. https://doi.org/10 .1080/17437199.2014.941722 Feltz, D. L., Chase, M. A., Moritz, S. E., & Sullivan, P. J. (1999). A conceptual model of coaching efficacy: Preliminary investigation and instrument development. Journal of Educational Psychology, 91, 765–776. https://doi.org/10.1037/ 0022-0663.91.4.765 Fishbein, M., & Ajzen, I. (2010). Predicting and Changing Behavior: The Reasoned Action Approach. New York: Psychology Press. French, D. P., Olander, E. K., Chisholm, A., & McSharry, J. (2014). Which behaviour change

Changing Behavior Using Social Cognitive Theory

techniques are most effective at increasing older adults’ self-efficacy and physical activity behaviour? A systematic review. Annals of Behavioral Medicine, 48, 225–234. https://doi.org /10.1007/s12160-014-9593-z. Galipeau, R., Baillot, A., Trottier, A., & Lemire, L. (2018). Effectiveness of interventions on breastfeeding self-efficacy and perceived insufficient milk supply: A systematic review and meta-analysis. Maternal and Child Nutrition, 14, e12607. https://doi.org/10.1111/mcn.12607 Ghazi, C., Nyland, J., Whaley, R., Rogers, T, Wera, J., & Henzman, C. (2018). Social cognitive or learning theory use to improve self-efficacy in musculoskeletal rehabilitation: A systematic review and meta-analysis. Physiotherapy Theory and Practice, 34, 495–504. https://doi.org/10 .1080/09593985.2017.1422204 Hankonen, N., Absetz, P, Kinnunen, M., Haukkala, A., & Jallinoja, P. (2013). Toward identifying a broader range of social cognitive determinants of dietary intentions and behaviors. Applied Psychology: Health and Well-Being, 5, 118–135. https://doi.org /10.1111/j.1758-0854.2012.01081.x Hays, L. M., Finch, E. A., Saha, C., Marrero, D. G., & Ackermann, R. T. (2014). Effect of self-efficacy on weight loss: A psychosocial analysis of a community-based adaptation of the diabetes prevention program lifestyle intervention. Diabetes Spectrum, 27, 270–275. https://doi.org /10.2337/diaspect.27.4.270. Honicke, T., & Broadbent, J. (2016). The influence of academic self-efficacy on academic performance: A systematic review. Educational Research Review, 17, 63–84. https://doi.org/10.1016/j .edurev.2015.11.002 Huang, C. (2016). Achievement goals and self-efficacy: A meta-analysis. Educational Research Review, 19, 119–137. https://doi.org/10 .1016/j.edurev.2016.07.002. Hull, C. L. (1943). Principles of Behavior: An Introduction to Behavior Theory. New York: Appleton-Century-Crofts. Jekauc, D., Völkle, M., Wagner, M. O., Mess, F., Reiner, M., & Renner, B. (2015). Prediction of attendance at fitness center: A comparison between the theory of planned behavior, the social

43

cognitive theory, and the physical activity maintenance theory. Frontiers in Psychology, 6, e121. https://doi.org/10.3389/fpsyg.2015.00121 Kok, G., Peters, G.-J. Y., Kessels, L. T. E., ten Hoor, G., & Ruiter, R. A. C. (2018). Ignoring theory and misinterpreting evidence: The false belief in fear appeals. Health Psychology Review, 12, 111–125. https://doi.org/10.1080/17437199.2017.1415767 Langebeek, N., Gisolf, E. H., Reiss, P. et al. (2014). Predictors and correlates of adherence to combination antiretroviral therapy (ART) for chronic HIV infection: A meta-analysis. BMC Medicine, 12. https://doi.org/10.1186 /s12916-014-0142-1 Lent, R. W., Sheu, H. B., Miller, M. J., Cusick, M. E., Penn, L. T., & Truong, N. N. (2018). Predictors of science, technology, engineering, and mathematics choice options: A meta-analytic path analysis of the social-cognitive choice model by gender and race/ethnicity. Journal of Counseling Psychology, 65, 17–35. https://doi.org/10.1037 /cou0000243 Lent, R. W., Brown, S. D., Schmidt, J., Brenner, B., Lyons, H., & Treistman, D. (2003). Relation of contextual supports and barriers to choice behavior in engineering majors: Test of alternative social cognitive models. Journal of Counseling Psychology, 50, 458–465. https://doi.org/10.1037 /0022-0167.50.4.458 Lent, R. W., Ezeofor, I., Morrison M. A., Penn, L.T., & Ireland, G. W. (2016). Applying the social cognitive model of career self-management to career exploration and decision-making. Journal of Vocational Behavior, 93, 47–57. https://doi.org /10.1016/j.jvb.2015.12.007. Lent, R. W., Ireland, G. W., Penn, L. T., Morris, T. R., & Sappington, R. (2017). Sources of self-efficacy and outcome expectations for career exploration and decision-making: A test of the social cognitive model of career self-management. Journal of Vocational Behavior, 99, 107–117. https://doi.org /10.1016/j.jvb.2017.01.002 Liyana Amin, N. A., Tam, W. W. S., & Shorey, S. (2018). Enhancing first-time parents’ self-efficacy: A systematic review and meta-analysis of universal parent education interventions’ efficacy. International Journal of Nursing Studies, 82,

44

ALEKSANDRA LUSZCZYNSKA AND RALF SCHWARZER

149–162. https://doi.org/10.1016/j.ijnurstu.2018 .03.021 Lowry, P. B., Zhang, J., & Wu, T. (2017). Nature or nurture? A meta-analysis of the factors that maximize the prediction of digital piracy by using social cognitive theory as a framework. Computers in Human Behavior, 68, 104–120. https://doi.org/10.1016/j.chb.2016.11.015 Luszczynska, A., Benight, C. C., & Cieslak, R. (2009). Self-efficacy and health-related outcomes of collective trauma: A systematic review. European Psychologist, 14, 49–60. https://doi.org/ 10.1027/1016-9040.14.1.51 Luszczynska, A., Hagger, M. S., Banik, A., Horodyska, K., Knoll, N., & Scholz, U. (2016). Self-efficacy, planning, or a combination of both? A longitudinal experimental study comparing effects of three interventions on adolescents’ body fat. PLoS ONE, 11, e0159125. https://doi.org/10.1371/journal .pone.0159125. Luszczynska, A., Horodyska, K., Zarychta, K., Liszewska, N., Knoll, N., & Scholz, U. (2016). Planning and self-efficacy interventions encouraging replacing energy-dense foods intake with fruit and vegetable: A longitudinal experimental study. Psychology and Health, 31, 40–64. https://doi.org/10.1080/08870446.2015 .1070156 Luszczynska, A., & Schwarzer, R. (2015). Socialcognitive theory. In M. Conner, & P. Norman (Eds.), Predicting Health Behaviours (3rd ed., pp. 225–251). Maidenhead: McGraw-Hill Open University Press. McCormick, M. J. (2001). Self-efficacy and leadership effectiveness: Applying social cognitive theory to leadership. Journal of Leadership and Organizational Studies, 8, 22–33. https://doi.org/10 .1177/107179190100800102 Myers, N. D., Park, S. E., Ahn, S., & Lee, S. (2017). Proposed sources of coaching efficacy: A meta-analysis. Journal of Sport and Exercise Psychology, 39, 261–276. https://doi.org/10.1123 /jsep.2017-0155 Pajares F., & Schunk, D. H. (2001). Self-beliefs and school success: Self-efficacy, self-concept, and school achievement. In R. Riding & S. Rayner

(Eds.), Perception (pp. 239–266). London: Ablex Publishing. Prestwich, A., Kellar, I., Parker, R. et al. (2013). How can self-efficacy be increased? Meta-analysis of dietary interventions. Health Psychology Review, 8, 270–285. https://doi.org/10.1080/17437199 .2013.813729 Prestwich, A., Sniehotta, F. F., Whittington C., Dombrowski, S. U., Roggers, L., & Michie, S. (2014). Does theory influence the effectiveness of health behavior interventions? meta-analysis. Health Psychology, 33, 465–474. https://doi.org /10.1037/a0032853 Schunk, D. H., & Zimmerman. B. J. (2007). Influencing children’s self-efficacy and self-regulation of reading and writing through modeling. Reading and Writing Quarterly, 23, 7–25. https://doi.org /10.1080/10573560600837578 Schwarzer, R., & Luszczynska, A. (2016). Self-efficacy and outcome expectancies. In Y. Benyamini, M. Johnston, & E. C. Karademas (Eds.), Assessment in Health Psychology (pp. 31–44). Göttingen: Hogrefe. Short, S., & Ross-Stewart, L. (2009). A review of selfefficacy-based interventions. In S. D. Mellalieu & S. Hanton (Eds.), Applied Sport Psychology Advances: A Review (pp. 221–281). London: Routledge. Skinner, B. F. (1938). The Behavior of Organisms. New York: Appleton-Century-Crofts. Stajkovic, A. D., & Luthans, F. (1998). Self-efficacy and work-related performance: A meta-analysis. Psychological Bulletin, 124, 240–261. https://doi .org/10.1037//0033-2909.124.2.240 Stacey, F. G., James, E. L., Chapman K., Courneya, K. S., & Lubans, D. R. (2015). A systematic review and meta-analysis of social cognitive theory-based physical activity and/or nutrition behavior change interventions for cancer survivors. Journal of Cancer Survivorship, 9, 305–338. https://doi.org/10.1007 /s11764-014-0413-z Talsma, K., Schüz, B., Schwarzer, R., & Norris, K. (2018). I believe, therefore I achieve (and vice versa): A meta-analytic cross-lagged panel analysis of self-efficacy and academic performance. Learning and Individual

Changing Behavior Using Social Cognitive Theory

Differences, 61, 136–150. https://doi.org/10.1016 /j.lindif.2017.11.015 Tang, M.Y., Smith, D. M., Mc Sharry, J., Hann, M., & French, D. P. (2019). Behavior change techniques associated with changes in postintervention and maintained changes in self-efficacy for physical activity: A systematic review with meta-analysis. Annals of Behavioral Medicine, 53, 801–815. https://doi.org/10.1093/abm/kay090 Te Velde, S. J., Lankhorst, K., Zwinkels, M., Verschuren, O., Takken, T., de Groot, J., & HAYS Study Group (2018). Associations of sport participation with self-perception, exercise self-efficacy and quality of life among children and adolescents with a physical disability or chronic disease-a cross-sectional study. Sports Medicine – Open, 4. https://doi.org/10.1186 /s40798-018-0152-1 Unrau, N. J., Rueda, R., Son, E., Polanin, J. R., Lundeen, R. J., & Muraszewski, A. K. (2018). Can reading self-efficacy be modified? A meta-analysis of the impact of interventions on reading self-efficacy. Review of Educational

45

Research, 88, 167–204. https://doi.org/10.3102 /0034654317743199 Warner, L. M., Schüz, B., Wolff, J. K., Parschau, L., Wurm, S., & Schwarzer, R. (2014). Sources of self-efficacy for physical activity. Health Psychology, 33, 1298–1308. https://doi.org/10 .1037/hea0000085. Williams, D. M., & Rhodes, R. E. (2016). The confounded self-efficacy construct: Conceptual analysis and recommendations for future research. Health Psychology Review, 10, 113–128. https:// doi.org/10.1080/17437199.2014.941998 Williams, S. L., & French, D. P. (2011). What are the most effective intervention techniques for changing physical activity self-efficacy and physical activity behaviour – and are they the same? Health Education Research, 26, 308–322. https://doi.org/10.1093/her/cyr005 Young, M. D., Plotnikoff, R. C., Collins, C. E., Callister, R., & Morgan, P. J. (2014). Social cognitive theory and physical activity: A systematic review and meta-analysis. Obesity Reviews, 15, 983–995.

4

Changing Behavior Using the Health Belief Model and Protection Motivation Theory Sheina Orbell, Hina Zahid, and Caroline J. Henderson

Practical Summary The health belief model and protection motivation theory are long-established models of behavior change. They underpin many health communications, such as warnings on cigarette packets, and are often the basis of commercial advertisements. According to these models, behavior change follows from inducing a person to perceive threat if they do not act and simultaneously providing information about an effective behavior they can easily enact in order to address that threat (coping appraisal). Threat can be manipulated by increasing a sense of vulnerability to a serious outcome. Coping appraisal involves providing information that behavior will be effective in reducing threat and by illustrating that behavior, showing that it is easy and not unpleasant to enact. If behavior change interventions manipulate threat appraisal alone, they may backfire, because they create anxiety without a solution, and, in this case, people will cope by discounting or denying the evidence or engage in wishful thinking.

4.1 Introduction During the twentieth century, developed countries witnessed a significant shift in major causes of mortality from those predominantly related to bacterial and viral infection, which were exacerbated by living conditions such as overcrowding and poor sanitation, to mortality that was attributable to what is now commonly referred to as “lifestyle”-related. Several factors contributed to this shift, including improvements in living conditions; the development and introduction of new medical technologies such as screening for tuberculosis and vaccination programs for diseases such as polio; and economic development associated with declining working hours, jobs that were less physically demanding, and greater disposable income. Over time, cancer, cardiovascular disease, and stroke became major sources of

early mortality (e.g., CDC, 2013). These changes set the scene for development of scientific interest in understanding how best to promote behavior that would maximize the effectiveness of new public health preventive measures such as screening and modify behaviors such as smoking, alcohol use, poor diet, and insufficient physical activity that are major causes of ill health. Psychological science also underwent a major paradigm shift toward the middle of the twentieth century. During the early decades of the twentieth century, psychology was dominated by behaviorism and the view that people learn stimulusresponse associations so that their actions are controlled by external factors, specifically contingent rewards or punishments (Hull, 1943; Skinner, 1938). However, behaviorism was soon replaced https://doi.org/10.1017/9781108677318.004

Changing Behavior Using the Health Belief Model and Protection Motivation Theory

by a view that people acquire mental representations of their world and expectations concerning the outcomes of their action. This shift from external to internal cognitive control of behavior led to the development of a class of models referred to as expectancy value theories. The health belief model (Becker, Haefner, Kasl et al., 1977; Hochbaum, 1958; Rosenstock, 1966, 1974) and protection motivation theory (Rogers, 1975, 1983) are two of the earliest social cognitive accounts of personal behavior that were developed primarily in the context of understanding and changing health behavior. They established enduring theoretical principles for behavior change that persist in some more recent theoretical developments (see, e.g., Chapters 2, 7, and 10, this volume). In the following sections, the components and structure of the health belief model and protection motivation theory are outlined, empirical evidence is evaluated, and the (enduring) use and efficacy of the models in informing intervention to modify behavior is considered.

Socio-demographic variables

47

4.2 Overview of Theory and Evidence 4.2.1 The Health Belief Model The health belief model is illustrated in Figure 4.1. The model focused on two aspects of people’s cognitive representations of health behavior. Two key beliefs, perceived susceptibility to illness and expected severity or impact of illness, comprise the mental representation of illness itself. Perceived benefits of action and perceived barriers/costs to action comprised the mental representation of action. The model also proposed that cues to action serve to trigger behavior when appropriate beliefs are held. Cues to action might include such things as a reminder letter or text message, a new symptom, or a television advertisement. An individual’s general health motivation was also included in later versions of the model (e.g., Becker, Haefner, & Maiman, 1977). These six constructs together comprised the health belief model. No formal propositions

Perceived susceptibility Perceived severity Health motivation Perceived costs or barriers Perceived benefits

Action

Cues to Action

Figure 4.1 Schematic representation of the health belief model

48

SHEINA ORBELL, HINA ZAHID, AND CAROLINE J. HENDERSON

regarding the operationalization of constructs, nor combinatorial rules, were proposed. Typically, the constructs are viewed and operationalized as separate independent variables that additively predict behavior. These features of the model may be viewed as weaknesses in the status of the model (Harrison, Mullen, & Green, 1992) but may also be seen as strengths in the context of a developing understanding of beliefs associated with a wide range of different behaviors with implications for health. For example, the severity of a condition or illness might be conceptualized in relation to a range of expectancies regarding a wide range of potentially relevant outcomes such as disfigurement, change in personal or sexual relations, inability to work, loss of income or loss of the ability to engage in desired leisure activities, and so on, as well as early death, and perceived benefits might correspondingly be related to potential prevention of these outcomes. Relatedly, barriers will be behavior-specific. For example, barriers to a behavior that requires appointment keeping may include such things as time or distance or means of transport, whereas unpleasantness or lack of skill may be more relevant in another context such as making complex dietary changes. The generic conceptualization of health beliefs permits researchers to undertake effective pilot work in order to identify context-appropriate beliefs about illness and barriers to action to measure and to target in behavior change intervention. A further important feature of the health belief model concerns the relationship of health beliefs to socioeconomic status and other demographic indices such as gender, ethnicity, and age. Figure 4.1 shows these variables on the left-hand portion of the model. Early accounts of the health belief model were grounded in the need to explain and intervene to achieve improved health equality in society via uptake of preventive health services or modifications of personal behavior with implications for health (Rosenstock, 1974). Sociodemographic variables are not considered modifiable variables by psychologists, but it was hypothesized that

identification of modifiable health beliefs that distinguish between individuals even of the same background would inform evidence-based health education to achieve greater health equality (e.g., Orbell, Crombie, & Johnstone, 1996). Currently, morbidity, mortality, and virtually all health behaviors are persistently associated with socioeconomic status, even in contexts where health care interventions such as screening are publicly financed and free at point of delivery (e.g., Orbell et al., 2017). The health belief model proposes that health beliefs mediate the relationship of sociodemographic variables to health action. For example, social experience engendered by social structural position might shape beliefs and subsequent behavior (e.g., do perceived costs/barriers loom larger for those with limited economic resources, or less daily experience of job discretion?) (Orbell et al., 1996; Salloway, Pletcher, & Collins, 1978). From a theoretical point of view, an important moderator hypothesis can also be derived from the model; modification of health beliefs will moderate (attenuate) the relationship of sociodemographic variables to behavior, thereby narrowing sociodemographic inequality.

4.2.2 Protection Motivation Theory Protection motivation theory is represented in Figure 4.2. The theory formalizes the idea that protection motivation, that is, motivation to take action to promote health and prevent illness, or some other potential threat, is derived from threat appraisal combined with coping appraisal. Coping appraisal is viewed as the mental representation of the “recommended” behavioral response to threat. Unlike the health belief model, the mental representations of threat and coping appraisal are not proposed as direct determinants of action but as direct determinants of protection motivation, usually operationalized as intention. Intention, in turn, is a proximal determinant of action. Threat appraisal comprises perceived susceptibility to, and perceived severity of, an outcome if action is not taken. Coping

Changing Behavior Using the Health Belief Model and Protection Motivation Theory

Threat Appraisal Perceived susceptibility Perceived severity Fear

49

Cope with Threat (e.g., Avoidance Denial, Fatalism, Wishful thinking, Hopelessness)

Coping Appraisal Perceived responseefficacy Perceived response-cost Perceived self-efficacy

Protection motivation (Intention to act to manage threat)

Threat protective behavior

Figure 4.2 Schematic representation of protection motivation theory

appraisal comprises response efficacy (beliefs that the recommended response will be effective in reducing threat, aka “perceived benefits”), response costs (expectancies that performing the recommended response will result in negative experiences, aka “perceived barriers”), and self-efficacy (beliefs in one’s competence and confidence to perform the recommended action; Bandura, 1977) (see also Chapters 3 and 7, this volume). Threat appraisal is proposed to enhance motivation to act, in circumstances where response efficacy is high, response costs are low, and self-efficacy is high. Protection motivation theory also makes theoretical predictions regarding the possibility that an individual may not respond to threat by taking appropriate protective action to address it but, instead, attempt to manage and reduce feelings of threat. These alternate responses are often referred to as “maladaptive coping responses” (Rippetoe & Rogers, 1987), illustrated in Figure 4.2 by the box labeled “Cope with threat.” Where threat appraisal is high, but response efficacy and self-efficacy are low and response costs high, rather than embark on the goal of avoiding threat, an individual may employ a range of strategies directed at discounting evidence or denying threat or avoiding thinking about threat in order to minimize unpleasant emotions.

4.2.3 Empirical Evidence Unsurprisingly, given their longevity and continued use, the health belief model and protection motivation theory have been subject to multiple research syntheses and meta-analyses. Janz and Becker’s (1984) review of forty-six studies employing the health belief model using a votecount procedure provided strong support for the model components: susceptibility (81 percent significant tests of association), severity (65 percent), benefits (78 percent), and barriers (89 percent). Harrison, Mullen, and Green (1992) and Carpenter (2010) provided full meta-analyses that were capable of not only establishing if relations were significant but also estimating the size and variability in the effects. Two metaanalytic syntheses of protection motivation theory (Floyd, Prentice-Dunn, & Rogers, 2000; Milne, Sheeran, & Orbell, 2000) estimated the size and variability of effects among the theory constructs. These meta-analyses have tended to focus on health behaviors but the models have also been employed in occupational and marketing contexts (see Sidebars 4.1 and 4.2). Previous summaries of evidence have tended to rely on tests of association derived from prospective designs, wherein beliefs and behavior are measured

50

SHEINA ORBELL, HINA ZAHID, AND CAROLINE J. HENDERSON

Sidebar 4.1 Protection motivation in an occupational setting

Safety is a major concern in many workplace settings and occupational injury contributes substantially to absence from work. Lower back pain is a widespread occupational hazard facing health care workers. Patient handling is often unpredictable, the human load bulky and unstable, and often lifting is required in awkward and unplanned situations – factors that contribute to manual handling risks not found in other occupational settings. In addition, protective behavior via the use of a hoist requires the cooperation of coworkers. Rickett, Orbell, and Sheeran (2006) investigated the role of threat and coping appraisal specified by protection motivation theory in explaining consistent hoist usage by health care workers. They developed scales to assess threat appraisal via perceived susceptibility to back problems (the likelihood of acquiring back problems as a consequence of moving patients), perceived severity (the extent to which back problems would interfere with career and social roles), and fear of back problems. Coping appraisal comprised response efficacy (the extent to which using a hoist would reduce the chances of acquiring or worsening back problems), response costs of using the hoist (such as the extent to which hoist use was perceived as time-consuming or would make the health care worker difficult to work with), response costs of not using the hoist (such as annoying my boss or colleagues), and self-efficacy (control over, competency, and ease of hoist use). Findings showed that threat appraisal variables were unrelated to the intention to use a hoist, and actual hoist use, during the following six weeks, whereas coping appraisal variables were important predictors, including self-efficacy (r = 0.43), response costs (r = −0.41), response benefits (r = 0.23), and social costs of not using a hoist (r = 0.35). The study pointed to the importance of workplace culture in ensuring safety equipment was available and staff competent in its use and to the promotion of shared beliefs regarding the appropriateness and normative expectation of hoist use.

Sidebar 4.2 Protection motivation in advertising to promote purchasing behavior

Adverts that aim to create new markets or increase product sales often combine elements that seek to enhance threat, followed by a “recommended response” to buy and use a product, consistent with protection motivation theory. These adverts typically exploit social anxieties. This form of advertising is common in relation to dental hygiene products such as mouthwash, home germ-cleansing products, and products related to home or family security such as life insurance. For example, the manufacturers of a mouthwash (ListerineTM) created an advert that self-evidently changed behavior by creating a remarkable new multimillion-dollar market for a previously unknown product in the 1950s.1 One of these adverts showed an

1

See “Listerine Mouthwash to the Rescue – 1950s,” Online video clip. www.youtube.com/watch? v=_39gkDAbMaI.

Changing Behavior Using the Health Belief Model and Protection Motivation Theory

51

attractive young woman who was “at risk” (susceptibility) of failing to attract a man and being left lonely and without companionship (undesired consequence: severity) because of her bad breath. The advert went on to introduce a coping response (bottle of mouthwash) that was portrayed as effective (in curing halitosis and attracting men: response efficacy) and easy to use (illustrated by images of placing the bottle conveniently by the sink and a demonstration of its use: self-efficacy). Modern advertising often follows a similar format by first arousing or making salient the threat that might arise from neglecting funeral insurance, or allowing children to come into contact with unclean surfaces, or failing to get the boiler serviced, followed by product information that emphasizes elements of coping appraisal designed to enhance perceptions that the product is easy to obtain and use and highly effective in reducing threat. The purpose of adverts is to create novel behavior and increase sales so they often employ effective behavior change techniques such as modeling to illustrate the perceived ease of use and efficacy of the product. Greater use of these coping appraisal techniques in interventions might enhance behavior change in many domains.

at different points in time and measurement of beliefs precedes measurement of behavior. Many early tests employed what epidemiologists might call “case-control studies,” or retrospective designs, in which people enacting a particular behavior are compared with those not enacting a previous behavior. Such approaches may have value in epidemiology through identifying the location of an outbreak of disease by answering key questions (e.g., “Have people with salmonella poisoning been in a recent location more than people who do not have salmonella poisoning?”). However, when applied to the measurement of beliefs, that is, comparing the current beliefs of people who have or have not performed a behavior in the past, they are fraught with theoretical and empirical limitation. For example, a person who has undergone screening for cancer and received a normal result, or who always uses a condom during sexual intercourse, may perceive low personal susceptibility to illness, even if perceived susceptibility to illness was originally a motivating force for action (Weinstein & Nicolich, 1993). The effect of perceived susceptibility on behavior might also be attenuated if an individual

responds to threat by engaging in a threat management coping response such as denial of risk or fatalism. In such cases, a measure of perceived susceptibility might reflect the outcome of this cognitive process. Relatedly, a person who has performed a given behavior repeatedly and found it straightforward and easy will probably perceive few barriers and high self-efficacy, based on that experience. Second, beliefs may change over time, as a consequence of new experience and knowledge. Prospective studies, in which the measurement of beliefs precedes the measurement of behavior, are likely to provide more reliable estimates of the relative importance of beliefs in promoting motivation and behavior change. A summary of evidence derived from prospective studies of the health belief model and protection motivation theory is provided in Table 4.1. Taken together, the table shows that all variables specified by the health belief model and protection motivation theory show small consistent relationships in the theoretically predicted direction with future behavior. Given that barriers may be equated to response costs and benefits to response efficacy, it may also be inferred that these variables

52

SHEINA ORBELL, HINA ZAHID, AND CAROLINE J. HENDERSON

Table 4.1 Summary of prospective relations between constructs specified by the health belief model and protection motivation theory and behavior Meta-analysis Construct

Harrison et al. (1992) Carpenter (2010)

Susceptibility 0.19 Severity 0.13 Benefits 0.10 Barriers −0.16 Response Efficacy Response Costs Self-Efficacy Protection Motivation (Intention)

0.05 0.15 0.27 −0.30

Milne et al. (2000) 0.12 0.07

0.09 −0.25 0.22 0.44

Note. Coefficients are bias-corrected averaged correlation coefficients.

appear to be more strongly associated with future behavior than threat appraisal variables. Selfefficacy also emerged as an important variable not included in the health belief model. The authors of all meta-analytic synthesis do, however, note that there is large unexplained heterogeneity in effects across studies. In addition, the metaanalyses did not include consideration of maladaptive responses to threat predicted by protection motivation theory. A few studies have examined these predictions (e.g., Abraham et al., 1994; BenAhron, White, & Phillips, 1995; Hodgkins & Orbell, 1998). For example, Ben-Ahron and colleagues observed that low-perceived severity and self-efficacy predicted avoidance coping and lowresponse efficacy and self-efficacy predicted religious faith coping in relation to binge drinking. Abraham and colleagues showed that low response and self-efficacy were associated with wishful thinking in relation to condom use to prevent HIV infection. Hodgkins and Orbell observed that high response costs predicted avoidance of breast self-examination. These studies provide evidence consistent with the theoretical prediction that, in the absence of high coping appraisal, perceived threat may prompt maladaptive coping responses.

These reviews of evidence are derived only from studies conducted in the health behavior domain. However, protection motivation theory, in particular, has also guided research in occupational settings (e.g., Melamed et al., 1996; Rickett, Orbell, & Sheeran, 2006) (see Sidebar 4.1) and more recently in relation to cybersecurity (e.g., Ashley et al., 2016; Hanus & Wu, 2016). The meta-analyses provide valuable estimates of the size of the relationship between cognitive representations (beliefs) and behavior. However, it should be borne in mind that these syntheses provide summaries of the strength of the relation between single variables and the behavior in question. They do not provide an assessment of the extent to which the individual variables, in combination, explain variance in future behavior. Future research may consider combining meta-analysis with techniques to test the combined effects of the theory variables in predicting behavior (e.g., Cheung & Hong, 2017; Hagger et al., 2017). Researchers should also note the possibility of ceiling effects, a class of response invariance that occurs when all participants in a study endorse a belief to an almost identical extent (e.g., the belief that cancer is a serious disease). Invariance will render a variable statistically incapable of explaining variance in behavior, underlining

Changing Behavior Using the Health Belief Model and Protection Motivation Theory

the importance of considering the wording of relevant items carefully.

4.3 How Have the Theories Been Used to Change Behavior? The health belief model and protection motivation theory have both been tested in interventions to change behavior. However, health belief model interventions have more often been employed in public health or clinical contexts, whereas protection motivation theory has been subject to more laboratory experimental tests (Milne, Orbell, & Sheeran, 2002). Interventions should aim to modify constructs specified by the theory. Typically, health belief model and protection motivation theory interventions take the form of written communications (leaflets or booklets) (e.g., Strecher et al., 1994), video presentations, or verbal communications via telephone (e.g., Anderson et al., 2011), or in person (e.g., Tola et al., 2016). Modern approaches might also involve text or e-communications (see Chapter 29, this volume). Some interventions employ health professionals and intensive education programs lasting over several sessions. For example, Tola et al. (2016) report a health belief model intervention in Ethiopia to improve adherence to treatment in patients with tuberculosis. The intervention involved seven sessions of health education delivered by health professionals that included psychological distress counseling to facilitate the processing of information and material to address perceived susceptibility and response efficacy, methods to overcome barriers to adherence and to enhance self-efficacy. The intervention resulted in a substantial increase in adherence that was supported by a change in targeted health beliefs. Interventions designed to demonstrate that a change in beliefs results in a change in behavior should follow baseline assessment of beliefs specified by the theory, followed by reassessment of the beliefs to determine if the manipulation has been successful in changing beliefs (see Chapters 19 and 22, this volume).

53

Final assessment of behavior at an appropriate point in time to observe behavior change should follow. Good practice to test theory would also establish that the changes in beliefs engendered by the education mediate the impact of the intervention on behavior change. Recommended elements of an intervention to change beliefs are presented in Appendix 4.1 (supplemental materials).

4.4 Evidence Base for the Use of the Theories in Changing Behavior 4.4.1 Evaluation of Behavior Change Interventions Based on the Health Belief Model and Protection Motivation Theory The first reported intervention based on the health belief model was published in 1970 (Haefner & Kirscht, 1970). The study concerned check-up visits to the doctor with an impressive eight-month follow-up. Just one systematic review of interventions employing the health belief model has been published to date (Jones et al., 2013) and there is, to date, no review of interventions based on protection motivation theory, although Milne et al. (2002) reviewed the impact of experimental manipulations on changing cognitive representations. Jones and colleagues reviewed eighteen studies, including sixteen randomized controlled trials, in contexts such as alcoholism, sleep apnea, asthma, and diabetes. The authors concluded that 77 percent of interventions applying the health belief model resulted in significant changes in behavior but noted that few studies provided evaluation of the mediation hypothesis discussed in section 4.2.1. While there is considerable support for the idea that behavior change interventions based on the health belief model and protection motivation theory are capable of changing behavior, it is perhaps timely, given that severity, benefits, and barriers are threat-specific, to consider reviews of studies that address behavior change in particular

54

SHEINA ORBELL, HINA ZAHID, AND CAROLINE J. HENDERSON

contexts in those contexts alone rather than attempt to aggregate them across contexts. As noted in Section 4.2.1, the measurement (and appropriate manipulation) of relevant constructs will always be behavior- and context-specific. The assumption that “off-the-shelf” measures, or pre-validated measures similar to those available for constructs such as anxiety or depression, represent superior measures is very unhelpful in the context of establishing the specific cognitive change brought about by an intervention targeting specific social cognition variables (see Chapter 19, this volume). For example, Jones et al. (2013) note the use of a wide range of intervention evaluation measures that are validated but irrelevant to establishing the veridicality of manipulation of health belief model variables. Further limitations of the extant evidence base relate to inadequate transparency and reporting of intervention content in research reports and articles, practices that stymie evaluation and replication efforts (see also Chapter 23, this volume), and inadequate follow-up, in terms of both time lag and the objective measurement of behavioral outcomes.

4.4.2 Theoretical and Empirical Relationship to Fear Appeals Perceived “threat” is a fundamental variable in both the health belief model and protection motivation theory and in both theories comprises perceptions of susceptibility to an adverse outcome and perceived severity of that outcome. Both accounts are premised on the idea that motivation follows from awareness that there is a threat to be addressed (see also Chapter 5, this volume). Response to threat is a function of beliefs about the likelihood of a response being effective, and beliefs that the response is easy to perform, and will not result in unpleasant consequences that undermine the positive consequences or benefits. Thus threat arousal might be seen as fundamental to interventions to promote behavior change. A useful summary term for this idea is “fear

appeal.” This section considers current evidence regarding the importance of “appealing to fear.” The history of fear appeals is a long one and the importance of threat in fear appeals is a contentious and much debated issue. In recent years, several meta-analyses have tackled the question of when appealing to fear may or may not effect behavior change (Kok et al., 2018; Peters, Ruiter, & Kok, 2013; Tannenbaum et al., 2015; Witte & Allen, 2000). Peters and colleagues conducted a review of studies that had manipulated both threat (via perceived susceptibility or severity or a combination thereof) and efficacy (via response or self-efficacy or a combination thereof). Their review adopted very stringent inclusion criteria, requiring that threat and coping efficacy be manipulated in a 2 × 2 experimental design (low vs. high or present vs. absent), and assessed a “real” behavioral outcome (e.g., studies where behavior involved requesting more information were excluded). Surprisingly, given that no date restrictions were applied to their search criteria, just six studies were eligible for inclusion. They encompassed a range of behaviors such as taking roundworm medication, earthquake preparation, tetanus vaccination, truancy, and condom use. The key hypothesis was that there would be a significant interaction between threat and efficacy manipulations, and this was supported in the analysis. Effect sizes (Cohen’s d) from an analysis of simple effects of the manipulations are presented in Appendix 4.2 (supplemental materials). Threat manipulations significantly increased behavior only when efficacy was high. In fact, the effect of manipulating threat in combination with low efficacy was negative, suggesting that this combination of intervention components might undermine attempts to change behavior. The effect of manipulating efficacy under conditions of low threat was nonsignificant, whereas the effect of manipulating efficacy under conditions of high threat was highly significant and large (d = 0.71). This analysis provides important clarification of behavior change processes not available either

Changing Behavior Using the Health Belief Model and Protection Motivation Theory

from meta-analyses of bivariate relations between constructs in prospective studies or from existing reviews of behavior change interventions employing protection motivation theory and the health belief model. Although the analysis is based on the small number of rigorous tests available, findings indicate that behavior change interventions should not target threat appraisals unless either the population is already high in self-efficacy and response efficacy (low barriers, high benefits) or the intervention also includes elements that will substantially increase response and self-efficacy. For example, if an intervention seeks to address accessibility to a service by providing the service in the worksite without loss of income, manipulating threat perceptions may be adequate. More often, however, personal self-efficacy needs to be addressed. Findings are consistent with recent meta-analyses with admittedly less stringent inclusion criteria that also propose that interventions that promote threat are unlikely to be effective in isolation. For example, in a meta-analysis of interventions to promote condom use to protect against sexually transmitted disease, there was minimal evidence that threat-inducing arguments were effective (Albarracín et al., 2005). Related to these findings, Sheeran et al. (2014) suggest that interventions that increased perceptions of susceptibility have larger effects on behavior change if they also increase response efficacy and selfefficacy (d = 0.52). Behavior change studies need to employ effective methods to target these constructs and demonstrate, in pilot work with similar populations, that the intervention has powerful effects on these constructs prior to implementation (see Appendix 4.2). Further work is also warranted that employs 2 (threat manipulation/intervention) × 2 (selfefficacy/response efficacy manipulation), or even 2 (severity manipulation) × 2 (susceptibility manipulation) × 2 (response efficacy manipulation) × 2 (self-efficacy manipulation), designs with sufficiently large samples to examine the effects of manipulations of severity, susceptibility, response

55

efficacy, and self-efficacy independently on behavior and their interactive effects. Protection motivation theory provides a theoretical rationale for people making “threat control” or emotion control responses to health communications (see Figure 4.2). Peters and colleagues’ (2013) observation that threat combined with low efficacy had a negative effect on behavior is consistent with the prediction from protection motivation theory that people will engage in denial, refutation of evidence, or emotion management responses when the recommended coping response is difficult or psychologically costly (Rippetoe & Rogers, 1987). Behavior change interventions that manipulate threat alone are not indicated by current evidence and may do more harm than good. Whereas health communications such as graphic threat images on cigarette packets tend to neglect the importance of selfefficacy of recommended action on behavior change (interested readers might like to consider recent debate of this issue: Borland, 2018; Brewer, Hall, & Noar, 2018; Kok et al., 2018; Malouff, 2018; Niederdeppe & Kemp, 2018; Peters & ShootsReinhard, 2018; Peters et al., 2018; Roberto, Mongeau, & Liu, 2018; White & Albarracín, 2018), those communications focused primarily on behavior change (that leads to economic gains by selling products) tend to employ threat as a means to introduce clear visual and narrative representations of response efficacy, and self-efficacy, including behavioral modeling (see Sidebar 4.2). However, it should also be noted that a good deal of behavior change in health contexts concerns changing existing behavioral habits that require specific skills and techniques (see Chapter 13, this volume).

4.4.3 Relationship of Beliefs to Sociodemographic Variables The health belief model (Figure 4.1) proposed that beliefs mediate the effects of social structural variables such as socioeconomic status, income, and education and demographic variables such as gender, age, and ethnicity on behavior. Relatively

56

SHEINA ORBELL, HINA ZAHID, AND CAROLINE J. HENDERSON

few studies have formally examined this mediation hypothesis (e.g., Orbell et al., 1996). More recently, Orbell et al. (2017) provided a unique evaluation of the ability of health beliefs to mediate the effects of both socioeconomic status and ethnicity on objectively observed colorectal screening behavior. Findings provided strong support for the role of perceived costs (barriers) and self-efficacy as mediator variables. As noted in Section 4.2.1, the identification of modifiable cognitive representations not only permits intervention to change behavior but such intervention should also moderate (attenuate) the relationship of sociodemographic variables on future behavior where such prior relationships between sociodemographic variables and behavior have been established (see also Chapter 27, this volume). This hypothesis remains largely untested. Some recent research does suggest one possible route by which manipulation of perceived benefits (response efficacy) and perceived barriers (response costs) might modify sociodemographic variability. Manipulations in which the benefits of action are framed as more immediate and the costs more distal have been shown to increase motivation for preventive health behavior across multiple domains (Orbell, Perugini, & Rakow, 2004; Orbell & Hagger, 2006; Orbell & Kyriakai, 2008; see also Whitaker et al., 2011). A recent study showed that these manipulations modified the motivation of people with Pakistani ethnicity to take up screening (Zahid, 2019). People of Pakistani ethnicity may share a form of predeterministic fatalism that results in low perceived efficacy to modify the long-term life course. Manipulating response efficacy as an immediate benefit served to promote motivation for screening. Further research is required that investigates the moderation hypothesis.

4.5 Conclusions The health belief model and protection motivation theory provided an important delineation of key

variables in motivating behavior change, particularly in the context of health threat, and were two of the earliest expectancy value models of behavior change. Variables specified by the models predict behavior prospectively and many interventions guided by these models show effects on behavior change. Use of the models to change behavior should ensure adequate consideration of coping efficacy, publication of content of manipulations, evidence that the manipulations impact on coping efficacy, and long-term follow-up of behavior.

References Abraham, C. S., Sheeran, P., Abrams, D., & Spears, R. (1994). Exploring teenagers’ adaptive and maladaptive thinking in relation to the threat of HIV infection. Psychology and Health, 9, 253–272. https://doi.org/10.1080/08870449408407485 Albarracín, D., Gillete, J. C., Earl, A. N., Glasman, L. R., & Durantini, M. R. (2005). A test of major assumptions about behaviour change: A comprehensive look at the effects of passive and active HIV interventions since the beginning of the epidemic. Psychological Bulletin, 131, 856–897. https://doi.org/10.1037/0033-2909.131.6.856 Anderson, R. T., Camacho, F., Iaconi, A. I., Tegeler, C. H., & Balkrishnan, R. (2011). Enhancing the effectiveness of community stroke risk screening: A randomised controlled trial. Journal of Stroke and Cerebrovascular Diseases, 20, 330–335. https://doi.org/10.1016/j .jstrokecerebrovasdis.2010.02.002 Ashley, D. N., Boothe, L. G., & Pearson, M. R. (2016). Risky electronic communication behaviors and cyberbullying victimisation: An application of Protection Motivation Theory. Computers in Human Behavior, 60, 508–513. https://doi.org/10 .1016/j.chb.2016.02.010 Bandura, A. (1977). Self-efficacy: Towards a unifying theory of behavioral change. Psychological Review, 84, 191–215. https://doi.org/10.1037 /0033-295X.84.2.191 Becker, M. H., Haefner, D. P., Kasl, S. V., Kirscht, J. P., Maiman, L. A., & Rosenstock, I. M. (1977). Selected psychosocial models and correlates of

Changing Behavior Using the Health Belief Model and Protection Motivation Theory

individual health-related behaviors. Medical Care, 15, 27–46. https://doi.org/10.1097/ 00005650-197705001-00005 Becker, M. H., Haefner, D. P., & Maiman, L. A. (1977). The health belief model in the prediction of dietary compliance: A field experiment. Journal of Health and Social Behavior, 18, 348–366. https://doi.org /10.2307/2955344 Ben-Ahron, V., White, D., & Phillips, K. (1995). Encouraging drinking at safe limits on single occasions: The potential contribution of protection motivation theory. Alcohol and Alcoholism, 30, 633–639. https://doi.org/10.1093 /oxfordjournals.alcalc.a045774 Borland, R. (2018). Misinterpreting theory and ignoring evidence: Fear appeals can actually work: A comment on Kok et al. (2018). Health Psychology Review, 12, 126–128. https://doi.org /10.1080/17437199.2018.1445545 Brewer, N. T., Hall, M. G., & Noar, S. M. (2018). Pictorial cigarette pack warnings increase quitting: A comment on Kok et al. Health Psychology Review, 12, 129–132. https://doi.org /10.1080/17437199.2018.1445544 Carpenter, C. J. (2010). A meta-analysis of the effectiveness of health belief model variables in predicting behavior. Health Communication, 25, 661–669. https://doi.org/10.1080/10410236 .2010.521906 CDC (Centers for Disease Control). (2013). Health Behaviors of Adults: United States, 2008–2010. Washington, DC: US Department of Health and Human Services. Cheung, M. W. L., & Hong, R. Y. (2017). Applications of meta-analytic structural equation modeling in health psychology: Examples, issues, and recommendations. Health Psychology Review, 11, 265–279. https://doi.org/10.1080/17437199 .2017.1343678 Floyd, D. L., Prentice-Dunn, S., & Rogers, R. W. (2000). A meta-analysis of research on protection motivation theory. Journal of Applied Social Psychology, 30, 407–429. https://doi.org/10.1111 /j.1559-1816.2000.tb02323.x Haefner, D. P., & Kirscht, J. P. (1970). Motivational and behavioral effects of modifying health beliefs. Public Health Reports, 85, 478–484.

57

Hagger, M. S., Koch, S., Chatzisarantis, N. L. D., & Orbell, S. (2017). The common-sense model of self-regulation: Meta-analysis and test of a process model. Psychological Bulletin, 143, 1117–1154. https://doi.org/10.1037/bul0000118 Hanus, B., & Wu, Y. A. (2016). Impact of users’ security awareness on desktop security behavior: A protection motivation theory perspective. Information Systems Management, 33, 2–16. https://doi.org/10.1080/10580530.2015.1117842 Harrison, J. A., Mullen, P. D., & Green, L. W. (1992). A meta-analysis of studies of the health belief model with adults. Health Education Research, 7, 107–116. https://doi.org/10.1093/her/7.1.107 Hochbaum, G. M. (1958). Public Participation in Medical Screening Programs: A Sociopsychological Study. Public Health Service Publication 572. Washington, DC: US Government Printing Office. Hodgkins, S. & Orbell, S. (1998) Does protection-motivation theory predict behavior? A longitudinal test and exploration of the role of previous behavior. Psychology and Health, 13, 237–250. https://doi.org/10.1080 /08870449808406749 Hull, C. L. (1943). Principles of Behaviour: An Introduction to Behaviour Theory. Oxford: Appleton-Century. Janz, N., & Becker, M. H. (1984). The health belief model: A decade later. Health Education Quarterly, 11, 1–47. https://doi.org/10.1177 /109019818401100101 Jones, C. J., Smith, H., & Llewellyn, C. (2013). Evaluating the effectiveness of interventions in improving adherence: A systematic review. Health Psychology Review, 8, 253–269. Kok, G., Peters, G.-J. Y., Kessels, L. T. E., Hoor, G. A., & Ruiter, R. A. C. (2018). Ignoring theory and misinterpreting evidence: The false belief in fear appeals. Health Psychology Review, 12, 111–125. https://doi.org/10.1080/17437199.2017.1415767. Malouff, J. M. (2018). What constitutes evidence that fear appeals have positive effects on health behaviour? Commentary on Kok, Peters, Kessels, ten Hoor, and Ruiter (2018). Health Psychology Review, 12, 133–135. https://doi.org/10.1080 /17437199.2018.1445541

58

SHEINA ORBELL, HINA ZAHID, AND CAROLINE J. HENDERSON

Melamed, S., Rabinowitz, S., Feiner, M., & Weisberg, E. (1996). Usefulness of protection motivation theory in explaining hearing protection device use among male industrial workers. Health Psychology, 15, 209–215. https://doi.org/10.1037 //0278-6133.15.3.209 Milne, S., Orbell, S., & Sheeran, P. (2002). Combining motivational and volitional interventions to promote exercise participation: Protection motivation theory and implementation intentions. British Journal of Health Psychology, 7, 163–184. https://doi.org/10.1348/135910702169420 Milne, S., Sheeran, P., & Orbell, S. (2000). Prediction and intervention in health-related behavior: A meta-analytic review of protection motivation theory. Journal of Applied Social Psychology, 30, 106–143. https://doi.org/10.1111/j.1559-1816 .2000.tb02308.x Niederdeppe, J., & Kemp, D. (2018). Ignoring theory and evidence: Commentary on Kok et al. (2018). Health Psychology Review, 12, 136–139. https:// doi.org/10.1080/17437199.2018.1445543 Orbell, S., Crombie, I., & Johnston, G. (1996) Social cognition and social structure in the prediction of cervical screening uptake. British Journal of Health Psychology, 1, 35–50. https://doi.org/10 .1111/j.2044-8287.1996.tb00490.x Orbell, S., & Hagger, M. S. (2006). Temporal framing and the decision to take part in Type 2 diabetes screening: Effects of individual differences in consideration of future consequences on persuasion. Health Psychology, 25, 537–548. https://doi.org/10.1037/0278-6133 .25.4.537 Orbell, S., & Kyriakai, M. (2008). Temporal framing and persuasion to adopt preventive health behavior: Moderating effects of individual differences in consideration of future consequences on sunscreen use. Health Psychology, 27, 770–779. https://doi.org/10.1037 /0278-6133.27.6.770 Orbell, S., Perugini, M., & Rakow, T. (2004). Individual differences in sensitivity to health communications: Consideration of future consequences. Health Psychology, 23, 388–396. https://doi.org/10.1037/ 0278-6133.23.4.388 Orbell, S., Sczepura, A., Gumber, A., Weller, D., & Hagger, M. S. (2017). South Asian ethnicity,

socio-economic status and psychological mediators of faecal occult colorectal screening participation: A test of a process model. Health Psychology, 36, 1161–1172. https://doi.org/10 .1037/hea0000525 Peters, G.-J. Y., Ruiter, R. A. C., & Kok, G. (2013). Threatening communication: A critical re-analysis and a revised meta-analytic test of fear appeal theory. Health Psychology Review, 7, S8–S31. https://doi.org/10.1080/17437199.2012.703527 Peters, G.-J. Y., Ruiter, R. A. C., ten Hoor, G. A., Kessels, L. T. E., & Kok, G. (2018). Towards consensus on fear appeals: A rejoinder to the commentaries on Kok, Peters, Kessels, ten Hoor, and Ruiter (2018). Health Psychology Review, 12, 151–156. https://doi.org/10.1080/17437199 .2018.1454846 Peters, E., & Shoots-Reinhard, B. (2018). Don’t throw the baby out with the bath water: Commentary on Kok, Peters, Kessels, ten Hoor, and Ruiter (2018). Health Psychology Review, 12, 140–143. https:// doi.org/10.1080/17437199.2018.1445542 Rickett, B., Orbell, S., & Sheeran, P. (2006). Socialcognitive determinants of joist usage among health care workers. Journal of Occupational Health Psychology, 11, 182–196. https://doi.org /10.1037/1076-8998.11.2.182 Rippetoe, P. A., & Rogers, R. W. (1987). Effects of components of protection motivation theory on adaptive and maladaptive coping with a health threat. Journal of Personality and Social Psychology, 52, 596–604. https://doi.org/10.1037 /0022-3514.52.3.596 Roberto, A. J., Mongeau, P. A., & Liu, Y. (2018). A (re) defining moment for fear appeals: A comment on Kok et al. (2018). Health Psychology Review, 12, 144–146. https://doi.org/10.1080/17437199 .2018.1445546 Rogers, R. W. (1975). A protection motivation theory of fear appeals and attitude change. Journal of Psychology, 91, 93–114. https://doi.org/10.1080 /00223980.1975.9915803 Rogers, R. W. (1983). Cognitive and physiological processes in fear appeals and attitude change: A revised theory of protection motivation. In J. T. Cacioppo & R. E. Petty (Eds.). Social Psychophysiology: A Source Book (pp. 153–176). New York: Guilford Press.

Changing Behavior Using the Health Belief Model and Protection Motivation Theory

Rosenstock, I. M. (1966). Why people use health services. Millbank Memorial Fund Quarterly, 44, 94–124. https://doi.org/10.1111/j.1468-0009 .2005.00425.x Rosenstock, I. M. (1974). Historical origins of the health belief model. Health Education Monographs, 2, 328–335. https://doi.org/10.1177 /109019817400200403 Salloway, J. C., Pletcher, W. R., & Collins, J. J. (1978). Sociological and social psychological models of compliance with prescribed regimen: In search of a synthesis. Sociological Symposium, 23, 100–121. Sheeran, P., Harris, E., & Epton, T. (2014). Does heightening risk appraisals change people’s intentions and behaviour? A meta-analytic review of the experimental evidence. Psychological Bulletin, 140, 511–543. https://doi.org/10.1037/a0033065 Skinner, B. F. (1938). The Behaviour of Organisms: An Experimental Analysis. New York: AppletonCentury. Strecher, V. J., Kreuter, M., DenBoer, D. J., Kobrin, S., Hospers, H. J., & Skinner, C. S. (1994). The effects of computer tailored smoking cessation messages in family practice settings. Journal of Family Practice, 39, 262–270. Tannenbaum, M. B., Hepler, J., Zimmerman, R. S. et al. (2015). Appealing to fear: A meta-analysis of fear appeal effectiveness and theories. Psychological Bulletin, 141, 1178–1204. https://doi.org/10.1037 /a0039729.

59

Tola, H. H., Shojaeizadeh, D., Tol, A. et al. (2016). Psychological and educational intervention to improve tuberculosis treatment adherence in Ethiopia based on health belief model: A cluster randomised controlled trial. PLoS ONE, 11, e0155147. https://doi.org/10.1371/journal .pone.0155147. Weinstein, N. D., & Nicolich, M. (1993). Correct and incorrect interpretations of correlations between risk perceptions and risk behaviors. Health Psychology, 12, 235–245. https://doi.org/10.1037 //0278-6133.12.3.235 Whitaker, K. L., Good, A., Miles, A., Robb, K., & Wardle, J. (2011). Socioeconomic inequalities in colorectal cancer screening uptake: Does time perspective play a role? Health Psychology, 30, 702–709. https://doi.org/10.1037/a0023941 White, B. X., & Albarracín, D. (2018). Investigating belief falsehood. Fear appeals do change behaviour in experimental laboratory studies. A commentary on Kok et al. (2018). Health Psychology Review, 12, 147–150. https://doi.org /10.1080/17437199.2018.1448292 Witte, K., & Allen, M. (2000). A meta-analysis of fear appeals: Implications for effective public health campaigns. Health Education and Behavior, 27, 591–615. https://doi.org/10.1177/10901981000 2700506 Zahid, H. (2019). Cultural influences on representation, coping and prevention of illness. Unpublished PhD thesis, University of Essex.

5

Changing Behavior Using the Common-Sense Model of Self-Regulation Linda D. Cameron, Sara Fleszar-Pavlović, and Tenie Khachikian

Practical Summary People often grapple with the challenge of motivating themselves or others to engage in adaptive behaviors in response to anxiety-arousing threats to their wellbeing. Such challenges can emerge in contexts such as illness diagnoses, health risks, environmental threats, sports competitions, public performances, or job interviews. Conversely, practitioners often face the challenge of persuading individuals to engage in behaviors that are in their best interests or the interests of the public good when the individuals either are unmotivated to do so or have stronger preferences for alternative actions such as those providing immediate gratification or avoidance of a threat. The common-sense model of self-regulation describes how beliefs, emotions, action plans, and appraisals of progress toward behavioral goals can influence behaviors in threatening or risky situations. This chapter describes the model and reviews its strategies for changing behavior to improve health-related outcomes as well as changing behaviors in other life domains.

5.1 Introduction The common-sense model of self-regulation (Leventhal, Brissette, & Leventhal, 2003; Leventhal, Meyer, & Nerenz, 1980) delineates cognitive and emotional processes influencing the self-regulation of behavior. In the context of the model, self-regulation refers to the processes used to control or direct thoughts, emotions, and actions toward the attainment of goals. These processes involve dynamic iterations of setting goals, planning and enacting behaviors, evaluating progress toward goal attainment, and revising goals and actions accordingly. The model was initially developed to focus on managing health threats such as somatic symptoms, health-risk communications, and illness experiences.

The common-sense model integrates and builds on theory and research from multiple scientific fields. First, it draws on control theory principles that behavior is construed as constantly changing in order to move one’s status toward one’s goals, standards, and ideals (Miller, Galanter, & Pribram, 1960; see Chapter 9, this volume). Second, it incorporates emotion regulation processes delineated by theories of emotion and research on how fear arousal motivates protective actions (Leventhal, 1970; Leventhal, Singer, & Jones, 1965). The key understanding arising from this emotions research, that communications must elicit both fear arousal and action planning, is central to the model (see also Chapter 34, this volume). Third, the model https://doi.org/10.1017/9781108677318.005

Changing Behavior Using the Common-Sense Model of Self-Regulation

integrates anthropological research on explanatory models of illness, revealing how these mental models have common structures across cultures (Kleinman, 1980, 1988). Specifically, these mental models encompass attributes reflecting beliefs about symptoms, causes, timeline, consequences, and control through treatments. That these attributes hold consistently across cultures suggests a universal need to understand these features of an illness experience. Finally, the model draws on cognitive science, delineating abstract, conceptual processes and concrete-experiential processes that are triggered in response to stimuli and operate in parallel (Epstein, 1994). Abstract, conceptual cognition involves reasoned, linguistic processing of information, whereas concrete-experiential processes involve perceptual, image-based contents. The integration of these multiple processes involving control dynamics, emotion regulation, mental models, and cognitive processes is a key strength of the common-sense model and underscores its potential for guiding behavior change techniques. The model is also used to understand and predict outcomes such as quality of life and distress (e.g., Richardson et al., 2016); however, this literature is beyond the scope of this chapter. This chapter reviews the common-sense model of self-regulation and its application to

61

interventions for behavior change. Given that the primary focus of the model is health-related behaviors, the chapter focuses largely on its application to changing behaviors in response to health threats. Importantly, however, the model incorporates selfregulation processes involved in response to threats and challenges more generally, such as environmental risks, education-related challenges, and political threats to social justice. The chapter thus considers applications of the model to changing behaviors in other arenas and highlights the need for theory development and research to broaden the model’s relevance and use to behavior more generally.

5.2 Brief Overview of the Common-Sense Model and Evidence The common-sense model of self-regulation (see Figure 5.1) emphasizes the roles of individuals’ common-sense beliefs about threats and rules for coping decisions. Perceptions of threat cues simultaneously activate problem-focused self-regulation via efforts to control the threat itself and emotionfocused self-regulation via efforts to manage distress-related arousal. In the problem-focused arm of the model, the activation of a mental schema or

Risk-Action Link Coherence

T H R E A T C U E

P E R C E P T I O N S

Threat Representation

Abstract Conceptual

Coping for Threat Control

Appraisal of Coping Outcomes

Coping for Emotional Control

Appraisal of Coping Outcomes

Identity Consequences Control Cause Timeline Coherence

Concrete Experiential Representation of Emotional Reaction (e.g., fear or worry)

Attentional Deployment Cognitive Reappraisal Proactive Behaviour Response Modulation

Figure 5.1 The common-sense model of self-regulation

62

´ , AND TENIE KHACHIKIAN LINDA D. CAMERON, SARA FLESZAR-PAVLOVIC

representation of the threat elicits emotional responses and guides action plans for threatreducing behavior. For example, an unusual lump can activate a representation of cancer, which can trigger worry and motivate scheduling a medical appointment. The emotion regulation arm involves efforts that can often fuel protective behavior; for example, worry about a rapid resting pulse can stimulate physical activity efforts. When fear is high, however, it can have the counterproductive effects of inhibiting behaviors that themselves pose threats of yielding negative consequences, such as when someone has a detection test – for example, a mammogram (Lerman et al., 1991) or chest x-ray (Leventhal & Watts, 1966). Appraisals of problemfocused and emotion-focused coping outcomes feedback to revise the representations and emotional arousal. Problem-focused and emotionfocused regulation involve both abstract conceptual processes and concrete-experiential processes. Whereas abstract processes can lead to relatively “cool” deliberative appraisals and decision-making, concrete-experiential processes can evoke relatively “hot” responses and impulsive reactions (Epstein, 1994). The model provides further elaboration on three features of the self-regulation process: representations, coherence, and emotion regulation behaviors. These features are discussed in turn, followed by considerations regarding variations of the model and general principles for applying it to behavior change.

5.2.1 Representations Representations activated by illness cues (e.g., sore throat) incorporate the five attributes of explanatory models (Cameron, Durazo, & Rus, 2016; Kleinman, 1988): (1) identity, including its label (e.g., strep throat) and symptoms (e.g., fever, fatigue); (2) causes (e.g., exposure to bacteria); (3) timeline or duration, which may be acute, cyclical (it comes and goes), or chronic; (4) consequences (e.g., need to miss work); and

(5) actions for its control or cure (e.g., antibiotics, rest). This representation guides action planning (e.g., seeking medical care, taking the prescribed antibiotics) and provides reference points for appraising outcomes of actions (e.g., whether the sore throat has subsided). A representation’s conceptual content (e.g., abstract knowledge about strep throat as caused by bacterial infection) and concrete-experiential content (e.g., memories of languishing in bed with strep throat) can differentially influence emotional and behavioral reactions. Concrete-experiential content can be more powerful than conceptual content in stimulating distress and motivating protective efforts (Cameron & Chan, 2008). Similar to representations of illness, representations of illness risk are activated by threat cues from media messages (e.g., about sun exposure and skin cancer), tests of susceptibility (e.g., genetic testing for skin cancer risk), personal characteristics (e.g., pale skin), and other sources. Risk representations include attributes comparable to those of illness representations: identity risk (label and characteristics conferring risk), causal risk (factors causing the condition), timeline (likely times in one’s life for its onset, likely duration), consequences (intrapersonal and interpersonal outcomes), and control/cure (whether and how the condition can be treated or controlled).

5.2.2 Coherence The common-sense model identifies coherence, or how representational attributes and their links with protective actions are understandable and “make sense,” as critical for motivating adaptive behaviors. Representational coherence, or having a clear understanding of the threat, can reduce distress caused by confusion and increase protection motivations (Durazo & Cameron, 2019). For example, someone diagnosed with gingivitis could feel calmer on learning that it is a controllable bacterial infection and the reason for one’s bleeding gums, and this understanding could motivate regular

Changing Behavior Using the Common-Sense Model of Self-Regulation

toothbrushing and flossing. Risk-action link coherence, or a clear understanding of how protective actions work to control the threat, can further motivate protective action (e.g., Cameron et al., 2012; Lee et al., 2011; see Appendix 5.1, supplemental materials). For example, women smokers might be motivated to quit if they understand how smoking increases cervical cancer risk by introducing chemicals into the bloodstream that reach and alter cervical cells.

5.2.3 Emotion Regulation The common-sense model also delineates four types of emotion regulation behaviors (Cameron & Jago, 2008; Gross, 1998): attentional deployment, cognitive reappraisal, proactive behavior, and response modulation. Attentional deployment involves efforts to focus on a threat (e.g., by either attending to or perseverating about an upcoming job interview) or avoid it (e.g., by distracting one’s thoughts away from it). Cognitive reappraisal involves reinterpreting the threat to be more benign (e.g., reinterpreting the interview as an opportunity or exciting challenge). Proactive behavior includes efforts to prepare for or manage the threat (e.g., searching the Internet for information about the organization) and includes seeking social support (e.g., asking others for interviewing tips). Response modulation includes modifying communications about emotional experiences (e.g., expressing or suppressing emotions to others) and acting to reduce emotional arousal (e.g., via relaxation exercises or alcohol or substance use). Through these processes, reactions such as fear or worry can motivate adaptive behavior or they can foster efforts that impede adaptive behavior (e.g., avoidance and substance misuse).

5.2.4 Variations of the Common-Sense Model Variations of the common-sense model take into account specific contextual factors (e.g., cancer survivorship; Durazo & Cameron, 2019),

63

behavior-related factors such as medication beliefs (e.g., Hagger et al., 2017; Horne et al., 2019), and factors that alter or moderate proposed common-sense model relationships (Hagger et al., 2017). For example, the revised commonsense model developed by Hagger and colleagues (2017) incorporates treatment beliefs, behavioral beliefs such as self-efficacy, and the moderating influences of illness characteristics (e.g., type of illness), dispositional and personality factors (e.g., dispositional optimism), and emotional representations (e.g., fear levels). A later version (Hagger & Orbell, 2019) elaborates further on model features such as mediational processes, behavior and treatment beliefs as determinants of coping and outcomes that are independent of threat representations, and additional moderators (e.g., trait negative affectivity, perfectionism).

5.2.5 Applications of the Common-Sense Model to Behavior Change The common-sense model can guide efforts to change a range of behaviors for managing health and other threats. Types of behaviors (with examples from the health field) include prevention efforts (e.g., Lee et al., 2011), informed decisions (e.g., Cameron et al., 2012), detection and management efforts (e.g., Levine et al., 2016), and emotion regulation behaviors influencing wellbeing (e.g., Cameron et al., 2007). It can be used to promote single behaviors (e.g., genetic testing for disease risk; Marteau & Weinman, 2006), cyclical behaviors (e.g., flu vaccinations; Parker et al., 2016), and lifestyle habits (e.g., sunscreen use; Hubbard et al., 2018). Which features of the model to target in interventions will vary according to characteristics of the behavior. Specifically, determining which common-sense model components to target requires considerations of whether the desired behavior involves (1) the prevention, detection, or management of a threat; (2) singular, cyclical, or ongoing

64

´ , AND TENIE KHACHIKIAN LINDA D. CAMERON, SARA FLESZAR-PAVLOVIC

enactment; (3) the need to manage distress interfering with behavior enactment; and (4) the need for feedback or training in appraising behavioral outcomes. Threat representations, coherence, and imagery processes can be useful targets for most behaviors. Emotion regulation can be particularly important in contexts evoking high fear, such as diagnoses of serious conditions or stressful public performances. Appraisal processes are critical targets for cyclical and ongoing behaviors, for which enabling individuals to receive and interpret feedback can motivate continued behavior and helpful revisions to behavior. Examples include insulin monitoring for diabetes control (Levine et al., 2016), performance appraisal in occupational settings (Fletcher, 2001), and provision of feedback to teachers in schools (Stormont & Reinke, 2013). Table 5.1 illustrates how common-sense model constructs can be operationalized as techniques and intervention components.

5.3 Strategies Targeting Representations to Elicit Behavior Change Strategies for changing representational attributes (identity, cause, timeline, consequences, controllability, and coherence) to promote behaviors include psychoeducational approaches, communication skills training for clinicians and practitioners, and motivational interviewing. These approaches and their use within health settings are discussed in the following sections.

5.3.1 Psychoeducational Approaches One approach for instilling adaptive illness representations is through personal sessions during which an educator reviews the individual’s illness beliefs and corrects inaccurate perceptions. These approaches range in intensity and requisite training from cognitive behavior therapy (e.g., Christensen et al., 2015) to short messages in communications such as pamphlets (e.g., Scott

et al., 2012). In one intervention (Petrie et al., 2002), patients hospitalized for myocardial infarction participated in three sessions with a trained educator prior to discharge. They received information about the pathophysiology of myocardial infarction, along with illustrations to instill concrete images of the condition; and engaged in discussions about their representational beliefs and the recovery process. Compared to usualcare participants, intervention participants exhibited positive changes in illness beliefs and returned to work more quickly. Subsequent research replicated and extended these findings by demonstrating that the intervention increased exercise and reduced telephone consultations with doctors (Broadbent et al., 2009). Similar psychoeducational approaches apply the common-sense model to develop letters and pamphlets promoting representational beliefs that motivate healthy behaviors. For instance, an intervention letter providing information about controllability and consequences increased cardiac rehabilitation attendance among myocardial infarction patients (Mosleh et al., 2014). Similarly, dental services utilization was increased for children whose parents received letters with information addressing dental caries identity, consequences, causes, control, timeline, coherence, and emotional representations (Nelson et al., 2017). As another example, a pamphlet with information about symptoms, timeline, and consequences of oral cancer decreased anticipated delay in seeking care for oral cancer symptoms and increased self-examination intentions (Scott et al., 2012). Importantly, the pamphlet was as effective as a comparable in-person session in motivating these behaviors relative to a control, no-information group. New social media and computer graphics technologies extend the scope of these psychoeducational approaches. For example, text messaging programs targeting illness representations can promote adherence with medications for controlling asthma (Petrie et al., 2012) and HIV/AIDS (Perera et al., 2014). Computer animations

Changing Behavior Using the Common-Sense Model of Self-Regulation

65

Table 5.1 Translation of common-sense model constructs into techniques and intervention components for promoting physical activity and a healthy diet to lose weight Construct

Technique example

Example message or task

Representations

Communication to change representations

Identity: Description of the levels of excess body fat identified as obesity. Cause: Inactivity and overeating foods high in sugar or fat contribute to weight gain. Consequences: Obesity increases the risk of diabetes, heart disease, and cancer. Control: A combination of physical activity and a low-calorie diet can reduce weight. Timeline: It will take about 5 months to lose about 20 pounds. Coherence: Images illustrating visceral and subcutaneous fat and how they affect bodily organs. Explanation of how physical activity promotes fat loss by burning calories and increasing metabolism.

Risk-Action Link Coherence

Communication to instill understanding of the relationship between specific actions and health risk Coping for Threat Action planning Develop a detailed dietary intake plan and physical Control activity schedule. Emotion Regulation Recognize internal emotional Identify typical situations that elicit obesity-related states distress and stress-induced eating. Place motivational cues, such as pictures of oneself Coping for Attentional deployment: at an ideal weight, healthy foods, and exercise Emotional efforts to focus on reducing gear, around the home. Control the threat Use emotional distress as a prompt to engage in Proactive behavior: physical activity to feel positive about oneself. management of the threat Reinterpret weight loss as a challenge that can be Cognitive reappraisal: achieved. reinterpretation of the threat Seek social support; use relaxation exercises to as more benign reduce obesity-related distress. Response modulation: reduction of emotional arousal Appraisal of Coping Self-monitoring Use an app to record daily weight status, diet, and Outcomes physical activity.

demonstrating physiological changes induced by behaviors can instill mental images that enhance representational and risk-action link coherence and, in turn, motivate protective action. For example, three-dimensional animation showing cardiac changes in response to healthy versus unhealthy habits can improve representational

beliefs, coherence, and behaviors among sedentary adults (Lee et al., 2011; see Appendix 5.1, supplemental materials). A trial of an animated intervention for patients with acute coronary syndrome yielded similar, beneficial effects on illness beliefs and adaptive behaviors (Jones et al., 2015).

66

´ , AND TENIE KHACHIKIAN LINDA D. CAMERON, SARA FLESZAR-PAVLOVIC

5.3.2 Communication Skills Training for Clinicians Clinicians often do not know or acknowledge that their patients’ illness representations may not align with their own knowledge (Salmon, Peters, & Stanley, 1999). Consequently, clinicians often direct consultations in ways that fail to elicit their patients’ inaccurate beliefs, thereby missing opportunities to address and modify them (Davison & Pennebaker, 1997). In an intervention designed to improve communication behaviors during consultations about hypertension management (de Ridder, Theunissen, & van Dulmen, 2007), clinicians were trained to discuss illness representations with their patients. Compared to care-as-usual consultations, consultations by trained clinicians were of a higher quality in terms of communication behaviors, information exchanges about representational beliefs, and affective utterances. Elliott et al. (2008) also found that a commonsense model–guided intervention for training pharmacists to discuss illness representations in telephone consultations with patients receiving new medications improved medication adherence.

5.3.3 Common-Sense Model–Guided Motivational Interviewing Motivational interviewing, defined as a “collaborative, person-centered form of guiding to elicit and strengthen motivation for change” (Miller & Rollnick, 2009, p. 137), has been particularly effective with individuals hesitant to change their behavior (see Chapter 45, this volume). The goal is to increase their intrinsic motivation to modify behavior using persuasive and supportive strategies. Structuring motivational interviews to address illness representations can be particularly effective in galvanizing behavior for individuals who are ambivalent to behavior change, such as individuals with poorly controlled diabetes

(Keogh et al., 2011) or periodontitis due to poor oral hygiene (Godard, Dufour, & Jeanne, 2011).

5.4 Strategies Targeting Emotion Regulation Processes to Change Behaviors The common-sense model identifies three categories of strategies for targeting emotion regulation processes to change behaviors: (1) increasing worry and related emotions that motivate adaptive behavior; (2) reducing worry or distress that inhibits adaptive behavior; and (3) changing emotion regulation behaviors that influence well-being. The following subsections discuss each category, along with examples of strategies and relevant evidence.

5.4.1 Arousing Worry and Related Emotions Communications and interventions can focus on enhancing worry to motivate adaptive behaviors in conditions where worry is low. Indeed, the original version of the common-sense model was specifically developed to account for the motivational effects of fear appeals and conditions under which they elicit behavior change (Leventhal, Meyer, & Nerenz, 1980; Leventhal, Singer, & Jones, 1965; Leventhal, Watts, & Pagano, 1967). Worry positively predicts adaptive actions (e.g., Cameron, 2008; Magnan & Cameron, 2015; McCaul, Schroeder, & Reid, 1996), and intervention strategies that increase worry enhance protection motivations (e.g., Lee et al., 2011; Magnan & Cameron, 2015). Fear appeals, such as graphic warnings on tobacco products, are frequently used to change behavior (Magnan & Cameron, 2015; Noar et al., 2016; see Chapter 34, this volume). Importantly, fear appeals motivate behavior change primarily when individuals maintain sufficient beliefs

Changing Behavior Using the Common-Sense Model of Self-Regulation

about self-efficacy, that is, that one can change the behavior, and response-efficacy, that is, that the behavior change will reduce risk (Witte & Allen, 2000). Otherwise, fear appeals can induce defensive avoidance and reactance, including rejection of message credibility. Other strategies for increasing worry to motivational levels include communications that enhance risk-action link coherence by providing individuals with a clear understanding of how a behavior reduces risk (Lee et al., 2011). In addition, framing risk messages with metaphors that arouse worry can serve to transfer worry associated with well-known threats to underappreciated risks. For example, one study demonstrated that a message describing sun exposure with an enemy combat metaphor of the sun as an enemy whose rays attack skin and metaphorically framing sun protection products as superheroes increased worry and sunscreen intentions, with worry mediating the metaphoric message effects on intentions (Landau, Arndt, & Cameron, 2018). These effects did not hold for individuals with low fear about enemy combat and for whom the metaphor did not activate worry.

5.4.2 Reducing Worry That Impedes Behavior For situations in which fear or worry impedes adaptive behaviors, interventions can focus on enhancing emotion regulation behaviors to promote behavior change. The common-sense model points to three such strategies: mindfulness practices, emotional expression, and cognitive reappraisal. Mindfulness exercises can reduce distress by reshaping attentional focus and promoting nonthreatening, nonjudgmental appraisals of stressful experiences (Goldin et al., 2017). Mindfulness practices can facilitate changes of numerous behaviors including binge eating (Kristeller, Wolever, & Sheets, 2014), excessive internet gaming (Li et al., 2017), athletic performance (Perry et al., 2017), and weight loss (Tapper et al., 2009).

67

Expressive writing tasks, through which individuals disclose feelings about stressful experiences, can also reduce threat-related distress (Frattaroli, 2006; Pennebaker, 1997). Expressive writing can improve desired behaviors and performance, although current evidence remains limited to selected behaviors such as academic performance (e.g., Cameron & Nicholls, 1998), aggressive behavior (Kliewer et al., 2011), and sleep behavior (Arigo & Smyth, 2012). Other behaviors likely to benefit from expressive writing include treatment adherence and athletic performance, although these effects remain untested. Cognitive reappraisal strategies, which are central to cognitive behavioral therapies, can reduce anxieties blocking adaptive behavior. For example, a message encouraging students to reappraise test anxiety as facilitating their cognitive abilities improves exam performance; moreover, its benefits generalize to performance in other academic pursuits (e.g., Brady, Hard, & Gross, 2018). Similar benefits of anxiety reappraisal hold for public speaking (Beltzer et al., 2014), singing (Brooks, 2014), reducing discriminatory behavior toward minority groups (Schultz et al., 2015), and reducing disordered eating (McLean, Paxton, & Wertheim, 2011).

5.4.3 Changing Emotion Regulation Behaviors A third set of efforts targeting emotion regulation aims to change emotion regulation habits to facilitate general well-being. Some emotion regulation interventions target a specific strategy (e.g., promoting mindfulness tendencies in daily life; Quaglia et al., 2016). Others target multiple emotional regulation habits. For example, interventions targeting mindfulness, cognitive reappraisal, and emotional expression tendencies have demonstrated improvements in these habits and, in turn, adjustment and quality of life (Cameron et al., 2007; Cameron, Carroll, & Hamilton, 2018; GieseDavis et al., 2002).

68

´ , AND TENIE KHACHIKIAN LINDA D. CAMERON, SARA FLESZAR-PAVLOVIC

5.5 Strategies Targeting ProblemFocused Coping with Action Plans to Elicit Behavior Change

and sleep behavior (Conroy & Hagger, 2018; Pham & Taylor, 1999; Loft & Cameron, 2013; see Chapter 33, this volume).

Early studies guided by theoretical precursors of the common-sense model revealed the importance of developing action plans for achieving behavior change (Leventhal, Singer, & Jones, 1965; Leventhal & Watts, 1966, Leventhal, Watts, & Pagano, 1967). They demonstrated that a combination of threat arousal, via messages about the dangers of tetanus or cigarette smoking, and a specific plan for taking action, such as for obtaining a tetanus vaccination or reducing smoking, was needed to induce behaviors (vaccinations or reductions in smoking). Subsequent research substantiated the efficacy of action planning for behavior change (Gollwitzer, 1999; Norman & Conner, 2005; see Chapters 6, 7, and 39, this volume). Efficacious action planning requires (1) setting realistic expectations for goal attainment; (2) clearly detailing actions; (3) specifying schedules and conditions for action; (4) instilling prompts or reminders; and (5), if needed, obtaining behavior skills training (e.g., training in how to study for an examination or maintain a compost pile). Mental imagery techniques offer potent means for creating and implementing action plans. Mental imagery engages concrete-experiential processing of visual and perceptual mental contents that link strongly with emotional processes (Cameron & Chan, 2008). Once formed and rehearsed, mental imagery schema or scripts engage automatic information processes that can prompt behavior independently of deliberative, volitional processes (Epstein, 1994). Mental simulation exercises guiding one to vividly imagine the situation, the process through which a behavior is enacted, and the attainment of the behavioral goal can be particularly effective in improving performance (Pham & Taylor, 1999). Mental simulations of action plans have been shown to improve numerous behaviors, including physical activity, academic performance,

5.6 Strategies Targeting Appraisals to Elicit Behavior Change Self-monitoring and feedback provision are commonly used to improve appraisal processes and change numerous behaviors, including gambling and dietary habits (e.g., Hsu, Rouf, & AllmanFarinelli, 2018; Rodda et al., 2018; see Chapter 37, this volume). Self-monitoring and feedback components are prevalent in computer (e.g., Chambliss et al., 2011), smartphone (e.g., Burke et al., 2017), and social media–based programs (e.g., Cavallo et al., 2012), with wearable activity trackers (e.g., Fitbit; Cadmus-Bertram et al., 2015) as notable examples. A common-sense model–guided intervention for improving self-monitoring and feedback appraisals to enhance Type 2 diabetes selfmanagement provides an example of the benefits of harnessing appraisal processes (Levine et al., 2016). Intervention participants utilized an automated, self-management monitor that provided reminders to self-test blood glucose, feedback on test results, interpretations of the results, and “action tips” based on results (e.g., “take one glucose tablet” or “increase physical activity”). Intervention compared to control participants exhibited higher self-testing rates and better glycated hemoglobin levels over the subsequent year. The research team recently supplemented the intervention with procedures for monitoring associations between blood glucose levels and behavior (McAndrew et al., 2008). Most interventions implementing selfmonitoring or feedback provision do so as part of multicomponent programs without testing their independent influences on behavior change. However, studies testing self-monitoring or feedback provision as sole interventions demonstrate their unique efficacy (e.g., Hutchesson et al., 2016;

Changing Behavior Using the Common-Sense Model of Self-Regulation

69

Sidebar 5.1 Need help with asthma control? There’s a common-sense model app for that!

The utilization of mHealth platforms provides new opportunities for delivering common-sense–guided interventions. The Adolescent Adherence Patient Tool (ADAPT; Kosse et al., 2019) for helping adolescents with asthma improve their adherence to their medications is a notable example of this approach. The intervention addresses the low rates of adherence to asthma medications during the adolescent years (Orrell-Valente et al., 2008). Causes of poor adherence include inaccurate understanding of asthma’s chronicity, symptoms of poor control, and the necessity of the medications; lack of worry or concern; poor regulation of asthmarelated distress; and forgetting to take medications (Holley et al., 2016). ADAPT targets these multiple factors through coordinated mechanisms. Short videos and chats with pharmacists promote accurate illness representations as well as action plans for adherence. Weekly measures to monitor symptoms and adherence are included to improve appraisal processes and adjust worry to motivational levels (e.g., when symptoms are elevated or minimized). Individually tailored reminders prompt medication use at appropriate times and prevent forgetting. The program also provides a peer chat function for sharing experiences and emotional support. ADAPT was tested in a cluster randomized controlled trial involving 66 Dutch pharmacies, with 234 adolescents aged 12–18 years assigned to 6 months of either app use or usual care. ADAPT improved medication adherence for adolescents with poor adherence rates at baseline, who tended to be older and have poorer initial control of their asthma. ADAPT thus appears to be helpful for those adolescents who are particularly vulnerable to problems with medication use and asthma control. ADAPT holds considerable promise as an mHealth intervention that can be easily implemented and widely disseminated within health care systems. It also provides a useful model for apps aimed at changing habitual, ongoing behaviors in other domains such as education, the workplace, and environmental conservation.

Zabatiero et al., 2013). For example, parents who received in-the-moment feedback from clinicians about their parenting behavior during parent-child interactions reduced their intrusive behavior and improved their sensitivity to their child’s experiences (Caron, Bernard, & Dozier, 2018).

5.7 Evidence Base for the Use of the Common-Sense Model in Changing Behavior As highlighted in prior sections, specific commonsense model–based behavior change strategies (e.g., mindfulness, fear arousal, action planning, self-

monitoring) have received extensive testing through trials that examine the mechanism by which each strategy affects behavior. However, trials testing behavior change interventions specifically guided by the model and targeting or testing multiple components of the model remain limited. An mHealth intervention for asthma control, described in Sidebar 5.1, provides a compelling example of the utility of common-sense model–guided interventions targeting multiple mechanisms. Most interventions guided by the common-sense model have focused on changing illness representations to promote behavior change; of these, most also target medication beliefs (Broadbent et al.,

70

´ , AND TENIE KHACHIKIAN LINDA D. CAMERON, SARA FLESZAR-PAVLOVIC

2009; Mosleh et al., 2014; Petrie et al., 2012; Scott et al., 2012) or problem-focused coping through action-planning strategies (Broadbent et al., 2009; Petrie et al., 2002; Scott et al., 2012). Such interventions adhere to common-sense model principles that optimal change arises from interventions targeting multiple facets of the self-regulation system; however, they do not provide evidence of the independent effects of representational changes on behavior. Further, studies yielding evidence that interventions altered representational attributes did not test whether these changes mediated the intervention effects on behavior. Disentangling the unique contribution of each representational attribute to behavior change represents another area demanding empirical attention. Such tests for mediation require appropriate assessments of changes in measures of mechanisms and behaviors over time (MacKinnon, Fairchild, & Fritz, 2010). Evidence on the efficacy of common-sense model–guided interventions to improve clinician communication skills or motivational interviewing to promote behavior change also remains limited. While the findings from the few trials are promising, more research is needed to determine their efficacy, effectiveness, and capacity for reach. Similarly, research on common-sense model– guided interventions enhancing emotion regulation skills or appraisal processes to improve behaviors remains scant. To date, emotion regulation interventions are limited in their reliance on intensive, group-based programs (Cameron et al., 2007, 2018). Future research efforts should focus on developing less-intensive programs delivered individually or through internet formats to enhance their feasibility and capacity for reach. This research area is further limited by a preponderance of studies employing small samples (e.g., Jones et al., 2015; Perera et al., 2014; Petrie et al., 2002). Studies testing moderators of intervention effects are also lacking, constraining our understanding of the boundary conditions of intervention effects and which social groups

might benefit most. The sole study testing for moderation of common-sense model–based interventions (Cameron et al., 2005) revealed that the Petrie et al. (2002) intervention targeting myocardial infarction illness perceptions induced detrimental behavioral responses for patients high in negative affectivity, that is, the patient’s tendency to experience anxiety and related emotions. That the intervention lowered cardiac rehabilitation attendance and exercise while increasing dietary fat intake for anxiety-prone individuals underscores the need to screen for this characteristic to determine intervention fit. Finally, more research is needed in which the common-sense model is used to take cultural differences in illness representations and coping beliefs into account when designing behavior change interventions. A core strength of the common-sense model is its reliance on common-sense beliefs about illnesses and treatments, both of which are shaped by cultural belief systems (Cameron et al., 2016). The power of the model for guiding culturally tailored interventions is demonstrated by an intervention to increase shoe wearing among Ethiopians at risk for podoconiosis (“mossy foot”) by promoting coherent understandings of the links between genetic susceptibility, walking barefoot, and exposure to affected soil (McBride et al., 2019). Much research is needed to develop behavior change interventions that incorporate culturally relevant beliefs. Despite these limitations of the evidence base, research offers promise for using the commonsense model to design behavior change interventions with benefits demonstrated in diverse illness settings (e.g., chronic, acute, and risk for illnesses) and populations (e.g., adolescents, racial/ethnic minorities, and older populations). Considerable scope exists for extending the model to change behaviors in other domains involving threats to psychosocial well-being, goal attainment, and the public good. For example, it has been applied to understand and predict behavior in response to environmental risks (Severtson, Baumann, &

Changing Behavior Using the Common-Sense Model of Self-Regulation

Brown, 2006), with implications for specific common-sense beliefs that could be targeted to increase protective actions.

5.8 Summary and Conclusions The common-sense model of self-regulation provides a rich framework for understanding behavior in threatening or risky situations and identifies cognitive, emotional, and behavioral mechanisms that can be targeted to change behavior. Growing evidence supports the use of multiple strategies for changing threat representations, emotional arousal, problem-focused coping, and appraisal processes to promote initiation and maintenance of behaviors that lead to adaptive outcomes such as reduced risk, better functioning, and better psychological well-being. Most research testing common-sense model–based strategies for behavior change has focused on behaviors for managing illness and other health threats. Similar approaches can be taken to inform interventions to improve academic performance by anxious or struggling students, public performance affected by social anxiety, job performance, pro-environmental behaviors, and political behaviors such as increasing votes for policies to improve the public good.

References Arigo, D., & Smyth, J. M. (2012). The benefits of expressive writing on sleep difficulty and appearance concerns for college women. Psychology and Health, 27, 210–226. https://doi.org /10.1080/08870446.2011.558196 Beltzer, M. L., Nock, M. K., Peters, B. J., & Jamieson, J. P. (2014). Rethinking butterflies: The affective, physiological, and performance effects of reappraising arousal during social evaluation. Emotion, 14, 761–768. https://doi.org/10.1037 /a0036326 Brady, S. T., Hard, B. M., & Gross, J. J. (2018). Reappraising test anxiety increases academic performance of first-year college students.

71

Journal of Educational Psychology, 110, 395–406. https://doi.org/10.1037/edu0000219 Broadbent, E., Ellis, C. J., Thomas, J., Gamble, G., & Petrie, K. J. (2009). Further development of an illness perception intervention for myocardial infarction patients: A randomized controlled trial. Journal of Psychosomatic Research, 67, 17–23. https://doi.org/10.1016/j.jpsychores.2008.12.001 Brooks, A. W. (2014). Get excited: Reappraising pre-performance anxiety as excitement. Journal of Experimental Psychology: General, 143, 1144–1158. https://doi.org/10.1037/a0035325 Burke, L. E., Zheng, Y., Ma, Q. et al. (2017). The SMARTER pilot study: Testing feasibility of real-time feedback for dietary self-monitoring. Preventive Medicine Reports, 6, 278–285. https:// doi.org/10.1016/j.pmedr.2017.03.017 Cadmus-Bertram, L., Marcus, B. H., Patterson, R. E., Parker, B. A., & Morey, B. L. (2015). Use of the Fitbit to measure adherence to a physical activity intervention among overweight or obese, postmenopausal women: Self-monitoring trajectory during 16 weeks. Journal of Medical Internet Research MHealth and UHealth, 3. https://doi.org10.2196/mhealth.4229 Cameron, L. D. (2008). Illness risk representations and motivations to engage in protective behavior: The case of skin cancer risk. Psychology and Health, 23, 91–112. https://doi.org/10.1080/147683207 01342383 Cameron, L. D., Booth, R. J., Schlatter, M., Ziginskas, D., & Harman, J. E. (2007). Changes in emotion regulation and psychological adjustment following use of a group psychosocial support program for women recently diagnosed with breast cancer. Psycho-Oncology, 16, 171–180. https://doi.org/10.1002/pon.1050 Cameron, L. D., Carroll, P., & Hamilton, W. K. (2018). Evaluation of an intervention promoting emotion regulation skills for adults with persisting distress due to adverse childhood experiences. Child Abuse and Neglect, 79, 423–433. https://doi.org /10.1016/j.chiabu.2018.03.002 Cameron, L. D., & Chan, C. K. Y. (2008). Designing health communications: Harnessing the power of affect, imagery, and self-regulation. Social and Personality Psychology Compass, 2, 262–282.

72

´ , AND TENIE KHACHIKIAN LINDA D. CAMERON, SARA FLESZAR-PAVLOVIC

https://doi.org/10.1111/j.1751-9004 .2007.00057.x Cameron, L. D., Durazo, A., & Rus, H. (2016). Illness representations. In Y. Benyamini, M. Johnston, & V. Karademas (Eds.), Assessment in Health Psychology (pp. 45–59). Abingdon: Hogrefe Publishing. Cameron, L. D., & Jago, L. (2008). Emotion regulation interventions: A common-sense model approach. British Journal of Health Psychology, 13, 215–221. https://doi.org/10.1348/135910708 X288800 Cameron, L. D., Marteau, T. M., Brown, P. M., Klein, W. M. P., & Sherman, K. A. (2012). Communication strategies for enhancing understanding of the behavioral implications of genetic and biomarker tests for disease risk: The role of coherence. Journal of Behavioral Medicine, 35, 286–298. https://doi.org/10.1007 /s10865-011-9361-5 Cameron, L. D., & Nicholls, G. (1998). Expression of stressful experiences through writing: Effects of a self-regulation manipulation for pessimists and optimists. Health Psychology, 17, 84–92. https:// doi.org/10.1037/0278-6133.17.1.84 Cameron, L. D., Petrie, K. J., Ellis, C., Buick, D., & Weinman, J. (2005). Trait negative affectivity and responses to a health education intervention for myocardial infarction patients. Psychology and Health, 20, 1–18. https://doi.org/10.1080 /08870440412331300011 Caron, E. B., Bernard, K., & Dozier, M. (2018). In vivo feedback predicts parent behavior change in the attachment and biobehavioral catch-up intervention. Journal of Clinical Child and Adolescent Psychology,47, S35–S46. https://doi .org/10.1080/15374416.2016.1141359 Cavallo, D. N., Tate, D. F., Ries, A. V., Brown, J. D., DeVellis, R. F., & Ammerman, A. S. (2012). A social media–based physical activity intervention: A randomized controlled trial. American Journal of Preventive Medicine, 43, 527–532. https://doi.org/10.1016/j.amepre.2012 .07.019 Chambliss, H. O., Huber, R. C., Finley, C. E., Mcdoniel, S. O., Kitzman-Ulrich, H., & Wilkinson, W. J. (2011). Computerized

self-monitoring and technology-assisted feedback for weight loss with and without an enhanced behavioral component. Patient Education and Counseling, 85, 375–382. https://doi.org/10.1016 /j.pec.2010.12.024 Christensen, S. S., Frostholm, L., Ørnbøl, E., & Schröder, A. (2015). Changes in illness perceptions mediated the effect of cognitive behavioural therapy in severe functional somatic syndromes. Journal of Psychosomatic Research, 78, 363–370. https://doi.org/10.1016/j.jpsychores.2014.12.005 Conroy, D., & Hagger, M. S. (2018). Imagery interventions in health behavior: A meta-analysis. Health Psychology, 37, 668–679. https://doi.org /10.1037/hea0000625 Davison, K. P., & Pennebaker, J. W. (1997). Illness representations in online support groups. In K. J. Petrie & J. Weinman (Eds.), Perceptions of Health and Illness (pp. 463–486). Amsterdam: Harwood Academic Publishers. de Ridder, D. T., Theunissen, N. C., & Dulmen, S. M. (2007). Does training general practitioners to elicit patients’ illness representations and action plans influence their communication as a whole? Patient Education and Counseling, 66, 327–336. https:// doi.org/10.1016/j.pec.2007.01.006 Durazo, A., & Cameron, L. D. (2019). Representations of cancer recurrence risk, recurrence worry, and health-protective behaviours: An elaborated, systematic review. Health Psychology Review, 13, 447–476. https://doi.org/10.1080/17437199 .2019.1618725 Elliott, R. A., Barber, N., Clifford, S., Horne, R., & Hartley, E. (2008). The cost effectiveness of a telephone-based pharmacy advisory service to improve adherence to newly prescribed medicines. Pharmacy World and Science, 30, 17–23. https:// doi.org/10.1007/s11096-007-9134-y Epstein, S. (1994). Integration of the cognitive and the psychodynamic unconscious. American Psychologist, 49, 709–724. https://doi.org/10 .1037/0003-066X.49.8.709 Fletcher, C. (2001). Performance appraisal and management: The developing research agenda. Journal of Occupational and Organizational Psychology, 74, 473–487. https://doi.org/10.1348 /096317901167488

Changing Behavior Using the Common-Sense Model of Self-Regulation

Frattaroli, J. (2006). Experimental disclosure and its moderators: A meta-analysis. Psychological Bulletin, 132, 823–865. https://doi.org/10.1037/ 0033-2909.132.6.823 Giese-Davis, J., Koopman, C., Butler, L. D. et al. (2002). Change in emotion-regulation strategy for women with metastatic breast cancer following supportive-expressive group therapy. Journal of Consulting and Clinical Psychology, 70, 916–925. https://doi.org/10.1037/0022-006X.70.4.916 Godard, A., Dufour, T., & Jeanne, S. (2011). Application of self-regulation theory and motivational interview for improving oral hygiene: A randomized controlled trial. Journal of Clinical Periodontology, 38, 1099–1105. https:// doi.org/10.1111/j.1600-051X.2011.01782.x Goldin, P. R., Morrison, A. S., Jazaieri, H., Heimberg, R. G., & Gross, J. J. (2017). Trajectories of social anxiety, cognitive reappraisal, and mindfulness during an RCT of CBGT versus MBSR for social anxiety disorder. Behaviour Research and Therapy, 97, 1–13. https://doi.org/10.1016/j.brat.2017.06.001 Gollwitzer, P. M. (1999). Implementation intentions: Strong effects of simple plans. American Psychologist, 54, 493–503. https://doi.org/10 .1037/0003-066x.54.7.493 Gross, J. J. (1998). The emerging field of emotion regulation: An integrative review. Review of General Psychology, 2, 271–299. https://doi.org /10.1037/1089-2680.2.3.271 Hagger, M. S., Koch, S., Chatzisarantis, N. L. D., & Orbell, S. (2017). The common-sense model of self-regulation: Meta-analysis and test of a process model. Psychological Bulletin, 143, 1117–1154. https://doi.org/10.1037/bul0000118 Hagger, M. S., & Orbell, S. (2019). A Revised Common-Sense Model of Illness Self-Regulation. Unpublished manuscript, University of California, Merced. Holley, S., Morris, R., Knibb, R. et al. (2016). Barriers and facilitators to asthma self-management in adolescents: A systematic review of qualitative and quantitative studies. Pediatric Pulmonology, 52, 430–442. https://doi.org/10.1002/ppul.23556 Horne, R., Cooper, V., Wileman, V., & Chan, A. (2019). Supporting adherence to medicines for long-term

73

conditions: A perceptions and practicalities approach based on an extended common-sense model. European Psychologist, 24, 82–96. https:// doi.org/10.1027/1016-9040/a000353 Hsu, M. S. H., Rouf, A., & Allman-Farinelli, M. (2018). Effectiveness and behavioral mechanisms of social media interventions for positive nutrition behaviors in adolescents: A systematic review. Journal of Adolescent Health, 63, 531–545. https://doi.org/10.1016/j.jadohealth.2018.06.009 Hubbard, G., Kyle, R. G., Neal, R. D., Marmara, V., Wang, Z., & Dombrowski, S. U. (2018). Promoting sunscreen use and skin self-examination to improve early detection and prevent skin cancer: Quasi-experimental trial of an adolescent psycho-educational intervention. BMC Public Health, 18, 1–15. https://doi.org/10 .1186/s12889-018-5570-y Hutchesson, M. J., Tan, C. Y., Morgan, P., Callister, R., & Collins, C. (2016). Enhancement of self-monitoring in a web-based weight loss program by extra individualized feedback and reminders: Randomized trial. Journal of Medical Internet Research, 18, e82. https://doi.org10.2196/jmir.4100 Jamieson, J. P., Peters, B. J., Greenwood, E. J., & Altose, A. J. (2016). Reappraising stress arousal improves performance and reduces evaluation anxiety in classroom exam situations. Social Psychological and Personality Science, 7, 579–587. https://doi.org/10.1177/1948550616644656 Jones, A. S., Ellis, C. J., Nash, M., Stanfield, B., & Broadbent, E. (2015). Using animation to improve recovery from acute coronary syndrome: A randomized trial. Annals of Behavioral Medicine, 50, 108–118. https://doi.org/10.1007 /s12160-015-9736-x Keogh, K. M., Smith, S. M., White, P. et al. (2011). Psychological family intervention for poorly controlled type 2 diabetes. American Journal of Managed Care, 17, 105–113. Kleinman, A. (1980). Patients and Healers in the Context of Culture: An Exploration of the Borderland between Anthropology, Medicine, and Psychiatry, Vol. 3. Berkeley: University of California Press. Kleinman, A. (1988). The Illness Narratives: Suffering, Healing, and the Human Condition. New York: Basic Books.

74

´ , AND TENIE KHACHIKIAN LINDA D. CAMERON, SARA FLESZAR-PAVLOVIC

Kliewer, W., Lepore, S. J., Farrell, A. D. et al. (2011). A school-based expressive writing intervention for at-risk urban adolescents’ aggressive behavior and emotional lability. Journal of Clinical Child and Adolescent Psychology, 40, 693–705. https:// doi.org/10.1080/15374416.2011.597092 Kosse, R. C., Bouvy, M. L., de Vries, T. W., & Koster, S. (2019). Effect of a mHealth intervention on adherence in adolescents with asthma: A randomized controlled trial. Respiratory Medicine, 149, 45–51. https://doi.org/10.1016/j .rmed.2019.02.009 Kristeller, J., Wolever, R. Q., & Sheets, V. (2014). Mindfulness-based eating awareness training (MB-EAT) for binge eating: A randomized clinical trial. Mindfulness, 5, 282–297. https://doi .org/10.1007/s12671-012-0179-1 Landau, M. J., Arndt, J., & Cameron, L. D. (2018). Do metaphors in health messages work? Exploring emotional and cognitive factors. Journal of Experimental Social Psychology, 74, 135–149. https://doi.org/10.1016/j.jesp.2017.09.006 Lee, T. J., Cameron, L. D., Wünsche, B., & Stevens, C. (2011). A randomized trial of computer-based communications using imagery and text information to alter representations of heart disease risk and motivate protective behaviour. British Journal of Health Psychology, 16, 72–91. https://doi.org/10.1348 /135910710X511709 Lerman, C., Trock, B., Rimer, B. K., Jepson, C., Brody, D., & Boyce, A. (1991). Psychological side effects of breast cancer screening. Health Psychology, 10, 259–267. https://doi.org/10.1037 /0278-6133.10.4.259 Leventhal, H. (1970). Findings and theory in the study of fear communications. In L. Berkowitz (Ed.), Advances in Experimental Social Psychology, Vol. 5 (pp. 119–186). New York: Academic Press. Leventhal, H., Brissette, I., & Leventhal, E. A. (2003). The common-sense model of self-regulation of health and illness. In L. D. Cameron & H. Leventhal (Eds.), The Self-Regulation of Health and Illness Behaviour (pp. 42–65). New York: Routledge. Leventhal, H., Meyer, D., & Nerenz, D. (1980). The common sense model of illness danger. In

S. Rachman (Ed.), Medical Psychology Vol. 2 (pp. 7–30). New York: Pergamon Press. Leventhal, H., Singer, R., & Jones, S. (1965). Effects of fear and specificity of recommendations upon attitudes and behavior. Journal of Personality and Social Psychology, 2, 20–29. https://doi.org/10 .1037/h0022089 Leventhal, H., & Watts, J. C. (1966). Sources of resistance to fear-arousing communications on smoking and lung cancer. Journal of Personality, 34, 155–175. https://doi.org/10.1111/j.1467-6494 .1966.tb01706.x Leventhal, H., Watts, J. C., & Pagano, F. (1967). Effects of fear and instructions on how to cope with danger. Journal of Personality and Social Psychology, 6, 313–321. https://doi.org/10.1037 /h0021222 Levine, J. C., Burns, E., Whittle, J. et al. (2016). Randomized trial of technology-assisted self-monitoring of blood glucose by low-income seniors: Improved glycemic control in type 2 diabetes mellitus. Journal of Behavioral Medicine, 39, 1001–1008. https://doi.org/10.1007 /s10865-016-9763-5 Li, W., Garland, E. L., McGovern, P., O’Brien, J. E., Tronnier, C., & Howard, M. O. (2017). Mindfulness-oriented recovery enhancement for internet gaming disorder in U.S. adults: A stage I randomized controlled trial. Psychology of Addictive Behaviors, 31, 393–402. https://doi.org /10.1037/adb0000269 Loft, M. H., & Cameron, L. D. (2013). Using mental imagery to deliver self-regulation techniques to improve sleep behaviors. Annals of Behavioral Medicine, 46, 260–272. https://doi.org/10.1007 /s12160-013-9503-9 MacKinnon, D. P., Fairchild, A. J., & Fritz, M. (2010). Meditation analysis. Annual Review of Psychology, 58, 593–614. https:// https://doi.org10.1146/annurev .psych.58.110405.085542 Magnan, R. E., & Cameron, L. D. (2015). Do young adults perceive that cigarette graphic warnings provide new knowledge about the harms of smoking? Annals of Behavioral Medicine, 49, 594–604. https://doi.org/10.1007/s12160-0159691-6

Changing Behavior Using the Common-Sense Model of Self-Regulation

Marteau, T. M., & Weinman, J. (2006). Self-regulation and the behavioural response to DNA risk information: A theoretical analysis and framework for future research. Social Science and Medicine, 62, 1360–1368. https://doi.org/10.1016 /j.socscimed.2005.08.005 McAndrew, L. M., Musumeci-Szabó, T. J., Mora, P. A. et al. (2008). Using the common sense model to design interventions for the prevention and management of chronic illness threats: From description to process. British Journal of Health Psychology, 13, 195–204. https://doi.org/10.1348 /135910708X295604 McBride, C., De Heer, H., Tekola Ayele, F., & Davey, G. (2019). Using genomic knowledge to improve health promotion interventions in the developing world. In E. B. Fisher, L. D. Cameron, A. J. Christensen et al. (Eds.), Principles and Concepts of Behavioral Medicine: A Global Handbook (pp. 713–736). New York: Springer. McCaul, K. D., Schroeder, D. M., & Reid, P. A. (1996). Breast cancer worry and screening: Some prospective data. Health Psychology, 15, 430–433. https://doi.org/10.1037/0278-6133.15.6.430 McLean, S. A., Paxton, S. J., & Wertheim, E. H. (2011). A body image and disordered eating intervention for women in midlife: A randomized controlled trial. Journal of Consulting and Clinical Psychology, 79, 751–758. https://doi.org/10.1037 /a0026094 Miller, G. A., Galanter, E., & Pribram, G. H. (1960). Plans and the Structure of Behavior. New York: Holt. Miller, W. R., & Rollnick, S. (2009). Ten things that motivational interviewing is not. Behavioural and Cognitive Psychotherapy, 37, 129–140. https://doi .org/10.1017/S1352465809005128 Mosleh, S. M., Bond, C. M., Lee, A. J., Kiger, A., & Campbell, N. C. (2014). Effectiveness of theory-based invitations to improve attendance at cardiac rehabilitation: A randomized controlled trial. European Journal of Cardiovascular Nursing, 13, 201–210. https://doi.org/10.1177 /1474515113491348 Nelson, S., Slusar, M. B., Albert, J. M., & Riedy, C. A. (2017). Do baby teeth really matter? Changing parental perception and increasing dental care

75

utilization for young children. Contemporary Clinical Trials, 59, 13–21. https://doi.org/10.1016 /j.cct.2017.05.002 Noar, S. M., Hall, M. G., Francis, D. B., Ribisl, K. M., Pepper, J. K., & Brewer, N. T. (2016). Pictorial cigarette pack warnings: A meta-analysis of experimental studies. Tobacco Control, 25, 341. https://doi.org10.1136/tobaccocontrol-2014-051978 Norman, P., & Conner, M. (2005). The theory of planned behavior and exercise: Evidence for the mediating and moderating roles of planning on intention-behavior relationships. Journal of Sport and Exercise Psychology, 27, 488–504. https://doi .org/10.1123/jsep.27.4.488 Orrell-Valente, J. K., Jarlsberg, L. G., Hill, L. G., & Cabana, M. D. (2008). At what age do children start taking daily asthma medicines on their own? Pediatrics, 122, 1186–1192. https://doi .org10.1542/peds.2008-0292 Parker, P. A., Banerjee, S. C., Matasar, M. J. et al. (2016). Protocol for a cluster randomised trial of a communication skills intervention for physicians to facilitate survivorship transition in patients with lymphoma. BMJ Open, 6. https://doi .org10.1136/bmjopen-2016-011581 Pennebaker, J. W. (1997). Writing about emotional experiences as a therapeutic process. Psychological Science, 8, 162–166. https://doi.org /10.1111/j.1467-9280.1997.tb00403.x Perera, A. I., Thomas, M. G., Moore, J. O., Faasse, K., & Petrie, K. J. (2014). Effect of a smartphone application incorporating personalized health-related imagery on adherence to antiretroviral therapy: A randomized clinical trial. AIDS Patient Care and STDs, 28, 579–586. https://doi.org/10.1089 /apc.2014.0156 Perry, J. E., Ross, M., Weinstock, J., & Weaver, T. (2017). Efficacy of a brief mindfulness intervention to prevent athletic task performance deterioration: A randomized controlled trial. The Sport Psychologist, 31, 410–421. https://doi.org /10.1123/tsp.2016-0130 Petrie, K. J., Cameron, L. D., Ellis, C. J., Buick, D., & Weinman, J. (2002). Changing illness perceptions after myocardial infarction: An early intervention randomized controlled trial. Psychosomatic

76

´ , AND TENIE KHACHIKIAN LINDA D. CAMERON, SARA FLESZAR-PAVLOVIC

Medicine, 64, 580–586. https://doi.org/10.1097/ 00006842-200207000-00007 Petrie, K. J., Perry, K., Broadbent, E., & Weinman, J. (2012). A text message programme designed to modify patients’ illness and treatment beliefs improves self-reported adherence to asthma preventer medication. British Journal of Health Psychology, 17, 74–84. https://doi.org/10.1111/j .2044-8287.2011.02033.x Pham, L. B., & Taylor, S. E. (1999). From thought to action: Effects of process- versus outcome-based mental simulations on performance. Personality and Social Psychology Bulletin, 25, 250–260. https://doi.org/10.1177/0146167299025002010 Quaglia, J. T., Braun, S. E., Freeman, S. P., McDaniel, M. A., & Brown, K. W. (2016). Metaanalytic evidence for effects of mindfulness training on dimensions of self-reported dispositional mindfulness. Psychological Assessment, 28, 803–818. https://doi.org/10.1037 /pas0000268 Richardson, E. M., Schüz, N., Sanderson, K., Scott, J. L., & Schüz, B. (2016). Illness representations, coping, and illness outcomes in people with cancer: A systematic review and meta-analysis. Psycho-Oncology, 26, 724–737. https://doi.org/10.1002/pon.4213 Rodda, S., Merkouris, S. S., Abraham, C., Hodgins, D. C., Cowlishaw, S., & Dowling, N. A. (2018). Therapist-delivered and self-help interventions for gambling problems: A review of contents. Journal of Behavioral Addictions, 7, 211–226. https://doi.org/10.1556/2006.7.2018.44 Salmon, P., Peters, S., & Stanley, I. (1999). Patients perceptions of medical explanations for somatisation disorders: Qualitative analysis. BMJ: British Medical Journal, 318, 372–376. https://doi .org/10.1136/bmj.318.7180.372

Schultz, J. R., Gaither, S. E., Urry, H. L., & Maddox, K. B. (2015). Reframing anxiety to encourage interracial interactions. Translational Issues in Psychological Science, 1, 392–400. https://doi.org/10.1037/tps0000048 Scott, S. E., Khwaja, M., Low, E. L., Weinman, J., & Grunfeld, E. A. (2012). A randomised controlled trial of a pilot intervention to encourage early presentation of oral cancer in high risk groups. Patient Education and Counseling, 88, 241–248. https://doi.org/10.1016/j.pec.2012.03.015 Severtson, D. J., Baumann, L. C., & Brown, R. L. (2006). Applying a health behavior theory to explore the influence of information and experience on arsenic risk perceptions, policy beliefs, and protective behavior. Risk Analysis, 26, 353–368. https://doi.org/10.1111/j.1539-6924 .2006.00737.x Stormont, M., & Reinke, W. M. (2013). Providing performance feedback for teachers to increase treatment fidelity. Intervention in School and Clinic, 49, 219–224. https://doi.org/10.1177 /1053451213509487 Tapper, K., Shaw, C., Ilsley, J., Hill, A. J., Bond, F. W., & Moore, L. (2009). Exploratory randomised controlled trial of a mindfulness-based weight loss intervention for women. Appetite, 52, 396–404. https://doi.org/10.1016/j.appet.2008.11.012 Witte, K., & Allen, M. (2000). A meta-analysis of fear appeals: Implications for effective public health campaigns. Health Education and Behavior, 27, 591–615. https://doi.org/10.1177/10901981000 2700506 Zabatiero, J., Kovelis, D., Furlanetto, K. C., Mantoani, L. C., Proença, M., & Pitta, F. (2013). Comparison of two strategies using pedometers to counteract physical inactivity in smokers. Nicotine and Tobacco Research, 16, 562–568.

6

Changing Behavior Using the Model of Action Phases Lucas Keller, Peter M. Gollwitzer, and Paschal Sheeran

Practical Summary The model of action phases highlights the distinction between the “why” and “how” questions related to goal pursuit. It implies that changing the behavior of individuals who have not yet decided to pursue a goal needs a “why”-oriented approach focusing on motivation, reasons, and rationale, whereas changing the behavior of people who have decided to pursue a goal needs a “how”-oriented approach focusing on strategies needed to enact the behaviors to attain the goal. The mindset theory of action phases proposes that different mental procedures need to be activated before a decision to perform a behavior to obtain a distal goal has been made compared to the procedures needed for goal pursuit after a decision has been made. These distinct deliberative and implemental mindsets influence the effectiveness of interventions geared at behavior change. Furthermore, implementation intentions (i.e., plans that specify in advance how the person will strive for chosen goals) are a particularly useful implemental tool and have been found to facilitate goal attainment.

6.1 Introduction In the mid-1980s, Heckhausen and Gollwitzer set out to analyze how people control their actions (see Heckhausen, Gollwitzer, & Weinert, 1987). They quickly realized that breaking action control down into different phases greatly benefited its understanding. Heckhausen and Gollwitzer’s analysis was heavily influenced by the work of Kurt Lewin (e.g., Lewin et al., 1944), for whom there was never any doubt that motivational phenomena can only be properly understood and analyzed from an action perspective that distinguishes the processes of goal setting from those of goal striving, an insight that went unheeded for several decades. Accordingly, Heckhausen and Gollwitzer (1987) proposed the “Rubicon” model of action phases, which describes the course of action as a temporal, linear path

starting with a person’s wishes or desires and ending with the evaluation of the action outcomes achieved. The model was designed to raise and help answer the following questions: How do people select their goals? How do they plan the execution of goal striving? How do they enact these plans? Moreover, how do they evaluate their accomplishments? According to the Rubicon model, a course of action involves a phase of deliberating the desirability and feasibility of one’s wishes at the outset in order to arrive at a binding decision regarding which of them one wants to pursue as a goal (pre-decisional phase), a phase of planning concrete strategies for achieving this goal Lucas Keller and Peter M. Gollwitzer gratefully acknowledge support from the German Research Foundation (DFG; FOR 2374). https://doi.org/10.1017/9781108677318.006

78

LUCAS KELLER, PETER M. GOLLWITZER, AND PASCHAL SHEERAN

(the pre-actional phase), a phase of enacting these plans (actional phase), and finally a phase of evaluating the achieved outcomes (post-actional phase). The Rubicon model was subsequently superseded by the mindset theory of action phases, in which each of the four phases was proposed to be associated with a distinct mindset (Gollwitzer, 1990, 2012). Defining the mindset in each phase required specifying in detail what kinds of phase-specific tasks need to be solved and whether engagement with these tasks initiates a distinct mindset that facilitates task performance. Research on mindset theory of action phases targeted the pre-decisional phase and the pre-actional phase and thus analyzed the features of the deliberative and implemental mindsets, respectively (for an interesting assessment of mindsets in the postactional phase, see McCrea & Vann, 2018). For instance, with respect to the pre-actional phase, it was postulated that the individual’s task is to prepare to “strive” for the upcoming goal. In line with this task demand, it was hypothesized and found in extensive experimental research (e.g., Heckhausen & Gollwitzer, 1987) using cognitive paradigms that the pre-actional individual is indeed focused on planning out the upcoming goal striving prospectively and that the respective implemental mindset carries cognitive features that facilitate meeting this task demand. Subsequent research (Gollwitzer, 1999, 2014) going beyond the implemental mindset notion started to explore what kind of planning is particularly effective in helping people to realize their goals. This research revealed that implementation intentions, specific cue-contingent plans linking context and action, qualify as a powerful self-regulation tool when it comes to striving for one’s goals, no matter to which domain these goals pertain (e.g., health, achievement, interpersonal; Gollwitzer & Sheeran, 2006). Research on mindset theory of action phases and implementation intentions are both inspired by the science of motivation. However, this research does not limit itself to expectancy-value type theorizing (see also Chapter 2, this volume). Rather, concepts

used by theorists in the early days of motivation science such as Narziss Ach, William James, and Kurt Lewin (see summary by Gollwitzer, 2018), including goals, plans, and mindsets, were revived to better understand what determines people’s actions and, in particular, how to change established action patterns. Moreover, this research also restored the distinction between motivation and volition by differentiating motivational phases of action that are occupied with the why of pursuing a certain goal and whether goal attainment did actually satisfy the person’s needs versus volitional phases that are occupied with planning out the how of striving for a chosen goal and getting involved with effectively realizing one’s goal commitments. The purpose of this chapter is to lay out the basic tenets and characteristics of mindset theory of action phases and implementation intention theory. Furthermore, it will provide an overview of how to induce respective mindsets and how planning with implementation intentions can help people attain their goals. The chapter is thus structured as follows: First, the mindset theory of action phases is presented in Section 6.2.1 alongside results of experimental research highlighting the differences in information processing in the different phases of the model. Next, the behavior change strategy of implementation intentions is outlined in Section 6.2.2 and the psychological processes by which implementation intentions affect behavior change described. Section 6.3 addresses the question of how the notion of mindsets (Section 6.3.1) and implementation intentions (Section 6.3.2) can be applied to understand and instigate behavior change. In Section 6.4, a meta-analysis of meta-analyses is presented that assesses the effectiveness of implementation intentions to instigate behavior change across a wide range of different domains and samples. Finally, Section 6.5 presents an outlook on potential future studies applying both mindsets and implementation intentions to instigate behavior change and fill gaps in the knowledge base regarding the underlying processes.

Changing Behavior Using the Model of Action Phases

6.2 Brief Overview of the Theory and Evidence 6.2.1 Mindset Theory of Action Phases Mindset theory of action phases describes successful goal pursuit as the smooth transition through the four consecutive but distinct action phases postulated in the Rubicon model (see Figure 6.1). In the first action phase, individuals have not yet decided what to do and must weigh the pros and cons and feasibility of their wishes. Once a wish has been turned into a binding goal that one wants to realize (i.e., the proverbial “Rubicon” has been crossed), the process enters the second action phase in which individuals plan the implementation of their decision. In the third action phase, individuals act on their goal by initiating goal striving and overcoming obstacles to stay on track. In the fourth and final action phase, after having completed their goal striving, individuals evaluate their progress, deeming further goaldirected action to be either necessary or futile. In each of these phases, individuals face different task demands that activate a typical set of beneficial cognitive procedures – termed “mindsets.” Once activated, these mindsets can carry over to

unrelated tasks – tasks that are different to those used to instigate the mindsets in the first place. This feature of action-phase–related mindsets allows for testing the presumed distinct task demands of the four action phases and answering the question of whether the four action phases of the Rubicon model are indeed distinct. Moreover, because of their trans-situational stability, inducing action-phase–related mindsets can be used to instigate behavior change (for overviews, see Gollwitzer, 2012; Gollwitzer & Keller, 2016). Making a goal decision has striking consequences for both information search and information processing. Before a decision to strive for a given goal is made, individuals need to process all available information in a relatively accurate manner with regard to both the feasibility (i.e., realistic assessments; e.g., Puca, 2001) and the desirability (i.e., impartial assessments; e.g., Taylor & Gollwitzer, 1995) of striving for this goal. Therefore, a certain open-mindedness concerning available information is beneficial in this early phase as well as accurate processing of this information (e.g., Fujita, Gollwitzer, & Oettingen, 2007). Once a decision has been made and the metaphorical Rubicon has been crossed, however, information on desirability and feasibility that may

Initiating goal-directed behavior

Phase I

Phase II

Phase III

Phase IV

Weighing the pros and cons of different options

Planning the when, where, and how to act

Acting on the goal and shielding it from distractions

Evaluating one’s goal striving

Crossing the Rubicon by making the decision

Figure 6.1 The model of action phases

79

Completing goal-directed behavior

80

LUCAS KELLER, PETER M. GOLLWITZER, AND PASCHAL SHEERAN

threaten the basis of the initial decision has to be disregarded (e.g., Bayer & Gollwitzer, 2005); now one needs to become more closed-minded in processing potentially relevant information. The task is to get started on striving for the chosen goal and, once started, one needs to stay on track; accordingly, shielding one’s goal striving from potential interferences is called for. The optimal focus on the implementation of the goal involves planning out when, where, and how to act; doing so activates a strong implemental mindset with cognitive features that help to meet the task demands of the pre-actional phase. Research on the features of the implemental mindset has in turn sparked interest in finding out what are particularly effective plans in facilitating goal striving and thus enhancing goal attainment (i.e., so-called implementation intentions).

6.2.2 Implementation Intentions Implementation intentions (see Gollwitzer, 1993, 1999, 2014) are plans that specify when, where, and how one will initiate a goal-directed response and involve creating an if (critical situation) and then (goal-directed response) contingency. Both the critical situation (opportunity or obstacle) and the goal-directed responses can take on quite different forms. For instance, the critical situation can be either inside (e.g., a certain feeling) or outside (e.g., a certain point in time or a certain event) the person, and the goal-directed response can pertain not only initiating or inhibiting a simple behavior (e.g., eating an apple, ignoring snacks) but also to thinking about things in a certain way (e.g., a positive evaluation) as well as engaging in or regulating a feeling (e.g., feeling pride, ignoring one’s negative mood). The effects of implementation intentions rest on two key processes: (1) they enhance the perception of, and attention to, the specified critical situation, and (2) they allow for automatic initiation of the specified goal-directed response on encountering the critical situation. Individuals who have

formed an implementation intention that specifies a critical situation in which a planned response is to be enacted are faster and more efficient in detecting this situation and in enacting the respective goal-directed response (e.g., Brandstätter, Lengfelder, & Gollwitzer, 2001; Orbell & Sheeran, 2000) – and this all without the need for further conscious involvement (e.g., self-talk such as “Oh, the critical situation is here; I’d better get going now!”; e.g., Bayer et al., 2009). Forming implementation intentions switches a person’s action control by goals (i.e., effortful, top-down control) to action control by specified critical situations (i.e., automatic, bottom-up control).

6.3 How Does Research Stimulated by Mindset Theory of Action Phases Inform Changing Behavior? 6.3.1 Deliberative and Implemental Mindsets People with a deliberative mindset have been found to become more open-minded with respect to processing available information, more impartial in evaluating pros and cons, and more realistic in judging probabilities of success than people with an implemental mindset. Examples of how to induce deliberative and implemental mindsets are provided in Sidebar 6.1. There is also evidence that these distinct cognitive orientations have different downstream consequences in terms of actual goal striving. For example, individuals with an implemental mindset evinced comparatively higher persistence in the face of difficulties and were more eager to work on their goals as expressed in less time needed for task completion than those in a deliberative mindset (Brandstätter et al., 2015). More recent mindset research showed that participants in an implemental mindset not only exhibit relative closed-mindedness with respect to processing available information but also show a more

Changing Behavior Using the Model of Action Phases

81

Sidebar 6.1 How to induce deliberative and implemental mindsets

Mindset inductions are key tools in empirical research on mindset theory of action phases. In a typical deliberative mindset induction, participants are asked to think about an unresolved personal concern (e.g., “Should I stop smoking?”; “Shall I move to a less noisy part of the city?”) that is currently on their mind but they have not yet decided on whether to act or not. Participants then go on to list short- and long-term, positive and negative consequences of both making a change decision and not making a change decision (i.e., maintaining the status quo). These personal wishes are commonly self-chosen by participants and often bear little resemblance to the behavior that is supposed to be changed. In a typical implemental mindset induction, participants are asked to think about a current personal project that they have decided to realize but have not acted on yet (e.g., “I want to drink less alcohol!”; “I want to exercise at least once a week!”). They are then asked to list a series of steps necessary to implement this project and to write down when, where, and how they plan to enact each of these envisioned steps. Again, these unresolved projects are self-chosen and often do not share any overlap with the behavior that is targeted for change. Experimental research on action phase mindsets (see an overview by Gollwitzer, 2012) has demonstrated that the more the unresolved concerns or chosen projects are of personal importance, the stronger the respective mindset effects (Taylor & Gollwitzer, 1995). Thus, to allow individuals enough time to really ponder their concerns or projects, the mindset inductions should take around fifteen minutes. Inductions can be paper- or computer-based as long as people are guided by precise instructions.

focused, narrower breadth of visual attention (determined by tracking eye movements) compared to participants in a deliberative mindset (Büttner et al., 2014). In line with implemental mindset effects on illusory feelings of control, Hügelschäfer and Achtziger (2014) found that participants in an implemental mindset are also more confident in having correctly answered questions in a general knowledge test than participants in a deliberative mindset. Strikingly, Dennehy, Ben-Zeev, and Tanigawa (2014) found that an implemental mindset helps people to shield themselves from the detrimental effects of stereotype threat. The induction of an implemental mindset helped participants from a low socioeconomic status background to overcome performance anxiety in a speeded mental arithmetic task, thus attenuating performance deficits caused by the experience of stereotype threat.

Recent research also tested whether deliberative versus implemental mindsets influence risk-taking behavior (Keller & Gollwitzer, 2017) using the Balloon Analogue Risk Task (BART; Lejuez et al., 2002). In the BART, one has to decide after each pump whether to go on pumping a balloon one more time or to save its current value and opt out. Each pump increases the balloon’s current value but also risks the balloon popping and the loss of the entire current value of the balloon. The BART mirrors risk-taking in the real world quite well (i.e., has a high ecological validity) as, for instance, indicated by the fact that it can differentiate smokers from nonsmokers (Lejuez, Aklin, Jones et al., 2003) and correlates with a variety of real-world risk-taking behaviors in adolescents (Lejuez, Aklin, Zvolensky et al., 2003). Keller and Gollwitzer (2017) found that participants in a deliberative mindset on average

82

LUCAS KELLER, PETER M. GOLLWITZER, AND PASCHAL SHEERAN

stopped pumping earlier on each balloon than participants in an implemental mindset, thus saving more balloons but foregoing potential monetary rewards by being too risk-averse. In line with these findings, Winterich and Nenkov (2015) reported four studies showing that people in a deliberative mindset exhibited an increased open-mindedness with respect to information about high saving rates of others. This, in turn, led to higher saving as compared to participants who did not receive such information and participants who received the information but were not in a deliberative mindset. This pattern of findings suggests that the deliberative mindset per se did not increase savings but rather increased openmindedness for information in favor of high savings.

6.3.2 Implementation Intentions Research on implementations intentions can be grouped into effect studies and process studies (for an overview, see Gollwitzer, 2014; see also

Sidebar 6.2 for how to study implementation intentions). In the effect studies, research has shown that implementation intentions facilitate goal attainment by promoting the initiation of goal striving, the shielding of ongoing goal striving from unwanted disruptive influences, the disengagement from failing courses of goal-directed action, and the conservation of capability for future goal striving (Gollwitzer & Sheeran, 2006). These findings held true for goals of different behavioral domains (e.g., health, achievement, interpersonal) and samples (e.g., age groups, cultures, various clinical samples). In the process studies, it was demonstrated that the effectiveness of implementation intentions is rooted in bottom-up control of the specified goaldirected responses by the respective situational cues (e.g., Gilbert et al., 2009; see also Chapter 39, this volume). These process studies explored the mental representation of the situational cues and the link between the specified cues and the respective response; it was found that

Sidebar 6.2 Experimenting with implementation intentions

In a typical implementation intention study, a between-participants design used: In the control group the participants are only given information about the task at hand (e.g., performing a math test), in the goal condition the participants are asked to form the intention to do well on the task (e.g., to succeed on the math test), and in the implementation intention condition participants are asked in addition to forming this goal to plan out in advance how they want to implement this goal (e.g., “As soon as I have solved one of the test items, then I will immediately move on to the next.”). The planning is triggered by questions on when, where, and how they want to act on their goal or by asking them to fill in a prepared statement: “If (opportunity/obstacle/ critical condition/cue) . . ., then I will (enact goal-directed response) . . .!” In some cases, the if-then plan is already specified by the experimenter/interventionist and provided to the research participant. Crucially, individuals in the mere goal condition and individuals in the implementation intention condition share the same underlying goal and significant parts of knowledge about how people can act to facilitate goal attainment. Forming implementation intentions is an easy-to-use self-regulation tool, as individuals commonly come up with a suitable plan on their own or readily adopt the given one (see also Chapter 39, this volume).

Changing Behavior Using the Model of Action Phases

implementation intentions heighten the cognitive accessibility of the critical cue and they strengthen the associative link between the cue and the planned response (e.g., Webb & Sheeran, 2007, 2008). Moreover, numerous experimental studies demonstrated that performing the planned response in the presence of the critical cues runs off in an automatic fashion; it is fast, efficient, and does not require conscious intent (e.g., Bayer et al., 2009; Sheeran, Webb, & Gollwitzer, 2005). Finally, brain studies indicate that action control by implementation intentions activates those brain regions that are known to be involved in automatic, bottom-up control by situational cues (e.g., Hallam et al., 2015). Recent research on implementation intentions has addressed the question of whether the influence of critical situations that cause unwanted impulsive and habitual responses can be countered by the formation of implementation intentions. It has done so by targeting cognitive, affective, and behavioral responses. With respect to cognitive responses, it has been shown that implicit stereotyping can be successfully controlled by forming implementation intentions like “whenever I see a black face on the screen, I will think the word ‘safe’ ” or “if I see a person, then I will ignore his race” (e.g., Stewart & Payne, 2008; Mendoza, Gollwitzer, & Amodio, 2010). With respect to affective responses, it was found that forming the plan “if I see a spider, then I will remain calm and relaxed” managed to curb fear responses in participants with spider phobia (Schweiger Gallo et al., 2009); and, with respect to the control of habitual behavioral responses, Marquardt and colleagues (2017) recently found that stroke patients with a mild-to-moderate hand paresis who formed task-specific plans like “when the green arrow points to the right, then I will press the left key instantly” performed better on the Simon Task, which assesses a person’s control over habitual hand movements, and this was true for the affected as well as the nonaffected hand. Taken together, these lines of

83

research show the effective regulation of affective, behavioral, and cognitive responses toward critical stimuli. A further new line of research focuses on the question of whether implementation intentions can also control social phenomena known to be enacted automatically, such as mimicry (Wieber, Gollwitzer, & Sheeran, 2014) or social projection (A. Gollwitzer et al., 2017). In all of the social phenomena studied, implementation intentions successfully down-regulated, and sometimes even up-regulated, these phenomena. Finally, recent research has asked whether a plan that specifies a switch to reflective thinking once the critical situation is encountered can be used to halt acting on one’s habits and impulses (e.g., Doerflinger, Martiny-Huenger, & Gollwitzer, 2017; Bieleke et al., 2017). Findings suggested that implementation intentions can automate the initiation of deep thinking; behavioral guidance by unwanted impulses and habits was prevented by planning to think when thinking was needed.

6.4 Evidence Base for Use of Theory in Changing Behavior To gain insight into the overall effectiveness of forming implementation intentions in promoting behavioral performance and goal attainment, a meta-analysis of published meta-analyses of implementation intention effects was conducted. Six reviews were located, which focused on multiple behaviors (Gollwitzer & Sheeran, 2006), specific behavioral categories (diet – Adriaanse et al., 2011; physical activity – Bélanger-Gravel, Godin, & Amireault, 2013; da Silva et al., 2018), affective outcomes (Webb et al., 2012), and clinical/psychiatric samples (Toli, Webb, & Hardy, 2016). The number of tests in the primary meta-analyses ranged from 13 to 93, and average effect sizes ranged from d+ = 0.24 to d+ = 0.99 (see Figure 6.2). The sampleweighted average effect size across these metaanalyses was d+ = 0.54 (95% CI = 0.51 to 0.57). The What Works Clearinghouse (WWC), a federal

84

LUCAS KELLER, PETER M. GOLLWITZER, AND PASCHAL SHEERAN

Figure 6.2 Meta-analysis of meta-analyses of implementation intention effects on goal achievement

Sidebar 6.3 Strong tests of the behavioral impact of implementation intentions

Although there appears to be no formal definition of a “strong test” of a behavioral intervention, several criteria can be proposed (see Chapter 22, this volume). The intervention should be effective (1) at scale (i.e., among large, representative samples); (2) using objective outcomes; (3) over extended time periods; and (4) for “difficult” or “complex” behaviors. Implementation intention research has met each of these criteria. Neter et al.’s (2014) memorably titled paper “From the bench to public health” reported a field experiment of if-then planning among ~30,000 people eligible for colorectal cancer screening. Electronic health records showed a substantial increase in fecal occult blood test adherence (OR = 1.17, p < 0.001). Martin et al. (2011) observed a 42 percent relative reduction in rates of clinically verified pregnancy two years later among socioeconomically deprived teenage women who formed implementation intentions in relation to their contraceptive use. Conner et al. (2019) tested the impact of if-then planning on smoking uptake among 6,155 adolescents over a time period of four years. The intervention led to a 6.5 percent reduction in the number of adolescents who had ever smoked (p < 0.001). Statistically significant effects of implementation intentions have thus been observed even in strong tests of intervention effects and for complex behaviors that have considerable importance in terms of public health. repository of “gold-standard” evidence on education programs, characterizes effect sizes of d+ ≥ 0.25 as “substantively important” (WWC, 2014, p. 23). It thus seems fair to conclude that forming implementation intentions is, at minimum, “substantively important” for promoting behavior change (see Sidebar 6.3).

6.5 Summary, Conclusion, and Outlook Both mindset inductions and implementation intentions have been used to instigate behavior change in various domains. Nevertheless, there are still many open questions. With respect to

Changing Behavior Using the Model of Action Phases

mindsets (Gollwitzer, 2012), for instance, the role of affect and affect-rich versus affect-poor information has not yet been systematically addressed. In contexts such as health risk communication, medical or moral decision-making, or when facing anxiety-provoking situations, individuals are exposed to affective information regarding negative future outcomes or have to make decisions in emotionally laden situations. Another issue relates to the emergence of a motivational task in a volitional action phase (Brandstätter & Schüler, 2013; Nenkov & Gollwitzer, 2012). What happens when people have to engage in renewed deliberation after they have crossed the Rubicon? Research addressing this question could contribute to a better understanding of how linear transitions between the four action phases are. Finally, implementation intentions can be combined with mental contrasting (Oettingen, 2014; Cross & Sheffield, 2019) to generate a powerful self-regulation strategy that enables people to deal with new and changing demands by themselves. Behavior change brought on by an omniscient choice architect (e.g., nudging; Thaler & Sunstein, 2008) cannot be generalized to new contexts as individuals are often oblivious to the effects of choice architecture or are not in a position to change choice architecture by themselves. In contrast, mental contrasting with implementation intentions (Oettingen, 2019) offers a meta-cognitive self-regulation tool that can be used as needed and in relation to any wishes the person may have. Mental contrasting prompts participants to first think about their desired futures, thus clarifying what they want to attain in the future, and then contrast the desired positive outcomes with their current personal obstacles standing in the way of attaining them. Mental contrasting leads to higher energization and stronger commitments given that individuals are confident that they can actually reach the desired outcome. Mental contrasting also helps participants to identify key obstacles to the realization of their wishes, obstacles that can be specified in

85

implementation intentions that link the obstacle with a goal-directed response that can overcome it. Mental contrasting with implementation intentions has been found to be even more effective than implementation intentions or mental contrasting on its own (Adriaanse et al., 2010), and warrants testing in future behavior change interventions.

References Adriaanse, M. A., Oettingen, G., Gollwitzer, P. M., Hennes, E., de Ridder, D. T. D., & de Wit, J. B. F. (2010). When planning is not enough: Fighting unhealthy snacking habits by mental contrasting with implementation intentions (MCII). European Journal of Social Psychology, 40, 1277–1293. https://doi.org/10.1002/ejsp.730 Adriaanse, M. A., Vinkers, C. D. W., de Ridder, D. T. D., Hox, J. J., & de Wit, J. B. F. (2011). Do implementation intentions help to eat a healthy diet? A systematic review and meta-analysis of the empirical evidence. Appetite, 56, 183–193. https:// doi.org/10.1016/j.appet.2010 .10.012 Bayer, U. C., Achtziger, A., Gollwitzer, P. M., & Moskowitz, G. B. (2009). Responding to subliminal cues: Do if-then plans facilitate action preparation and initiation without conscious intent? Social Cognition, 27, 183–201. https://doi .org/10.1521/soco.2009.27.2.183 Bayer, U. C., & Gollwitzer, P. M. (2005). Mindset effects on information search in self-evaluation. European Journal of Social Psychology, 35, 313–327. https://doi.org/10.1002/ejsp.247 Bélanger-Gravel, A., Godin, G., & Amireault, S. (2013). A meta-analytic review of the effect of implementation intentions on physical activity. Health Psychology Review, 7, 23–54. https://doi .org/10.1080/17437199.2011.560095 Bieleke, M., Gollwitzer, P. M., Oettingen, G., & Fischbacher, U. (2017). Social value orientation moderates the effects of intuition versus reflection on responses to unfair ultimatum offers. Journal of Behavioral Decision Making, 30, 569–581. https://doi.org/10.1002/bdm.1975 Brandstätter, V., Giesinger, L., Job, V., & Frank, E. (2015). The role of deliberative versus

86

LUCAS KELLER, PETER M. GOLLWITZER, AND PASCHAL SHEERAN

implemental mindsets in time prediction and task accomplishment. Social Psychology, 46, 104–115. https://doi.org/10.1027/1864-9335/a000231 Brandstätter, V., Lengfelder, A., & Gollwitzer, P. M. (2001). Implementation intentions and efficient action initiation. Journal of Personality and Social Psychology, 81, 946–960. https://doi.org /10.1037//0022-3514.81.5.946 Brandstätter, V., & Schüler, J. (2013). Action crisis and cost-benefit thinking: A cognitive analysis of a goal-disengagement phase. Journal of Experimental Social Psychology, 49, 543–553. https://doi.org/10.1016/j.jesp.2012.10.004 Büttner, O. B., Wieber, F., Schulz, A. M., Bayer, U. C., Florack, A., & Gollwitzer, P. M. (2014). Visual attention and goal pursuit: Deliberative and implemental mindsets affect breadth of attention. Personality and Social Psychology Bulletin, 40, 1248–1259. https://doi.org/10.1177/014616721 4539707 Conner, M., Grogan, S., West, R. et al. (2019). Effectiveness and cost-effectiveness of repeated implementation intention formation plus anti-smoking messages on adolescent smoking initiation: A cluster randomized controlled trial. Journal of Consulting and Clinical Psychology, 87, 422–432. https://dx.doi.org/10.1037%2Fccp 0000387 Cross, A., & Sheffield, D. (2019). Mental contrasting for health behaviour change: A systematic review and meta-analysis of effects and moderator variables. Health Psychology Review, 13, 209–225. https:// doi.org/10.1080/17437199.2019.1594332 da Silva, M. A. V., São-João, T. M., Brizon, V. C., Franco, D. H., & Mialhe, F. L. (2018). Impact of implementation intentions on physical activity practice in adults: A systematic review and meta-analysis of randomized clinical trials. PLoS ONE, 13, e0206294. https://doi.org/10.1371 /journal.pone.0206294 Dennehy, T. C., Ben-Zeev, A., & Tanigawa, N. (2014). “Be prepared”: An implemental mindset for alleviating social-identity threat. British Journal of Social Psychology, 53, 585–594. https://doi.org /10.1111/bjso.12071 Doerflinger, J., Martiny-Huenger, T., & Gollwitzer, P. M. (2017). Planning to deliberate

thoroughly: If-then planned deliberation increases the adjustment of decisions to newly available information. Journal of Experimental Social Psychology, 69, 1–12. https://doi.org/10.1016/j .jesp.2016.10.006 Fujita, K., Gollwitzer, P. M., & Oettingen, G. (2007). Mindsets and pre-conscious open-mindedness to incidental information. Journal of Experimental Social Psychology, 43, 48–61. https://doi.org/10 .1016/j.jesp.2005.12.004 Gilbert, S. J., Gollwitzer, P. M., Cohen, A.-L., Burgess, P. W., & Oettingen, G. (2009). Separable brain systems supporting cued versus self-initiated realization of delayed intentions. Journal of Experimental Psychology: Learning, Memory, and Cognition, 35, 905–915. https://doi .org/10.1037/a0015535 Gollwitzer, A., Schwoerer, B., Stern, C., Gollwitzer, P. M., & Bargh, J. A. (2017). Down and up regulation of a highly automatic process: Implementation intentions can both decrease and increase social projection. Journal of Experimental Social Psychology, 70, 19–26. https://doi.org/10.1016/j.jesp.2016.12.006 Gollwitzer, P. M. (1990). Action phases and mind-sets. In E. T. Higgins & R. Sorrentino (Eds.), The Handbook of Motivation and Cognition: Foundations of Social Behavior, Vol. 2 (pp. 53–92). New York: Guilford Press. Gollwitzer, P. M. (1993). Goal achievement: The role of intentions. European Review of Social Psychology, 4, 141–185. https://doi.org/10.1080 /14792779343000059 Gollwitzer, P. M. (1999). Implementation intentions: Strong effects of simple plans. American Psychologist, 4, 493–503. https://doi.org/10.1037 /0003-066X.54.7.493 Gollwitzer, P. M. (2012). Mindset theory of action phases. In P. Van Lange, A. W. Kruglanski, & E. T. Higgins (Eds.), Handbook of Theories of Social Psychology (pp. 526–545). London: SAGE. Gollwitzer, P. M. (2014). Weakness of the will: Is a quick fix possible. Motivation and Emotion, 38, 305–22. https://doi.org/10.1007/s11031-0149416-3 Gollwitzer, P. M. (2018). The goal concept: A helpful tool for theory development and testing in

Changing Behavior Using the Model of Action Phases

motivation science. Motivation Science, 18, 185–205. https://doi.org/10.1037/mot0000115 Gollwitzer, P. M., & Keller, L. (2016). Mindset theory. In V. Zeigler-Hill & T. K. Shackelford (Eds.), Encyclopedia of Personality and Individual Differences (pp. 1–8). Cham: Springer International. https://doi.org/10.1007/978-3-319-28099-8_1141-1 Gollwitzer, P. M., & Sheeran, P. (2006). Implementation intentions and goal achievement: A meta-analysis of effects and processes. Advances in Experimental Social Psychology, 38, 69–119. https://doi.org/10.1016/S0065-2601 (06)38002–1 Hallam, G. P., Webb, T. L., Sheeran, P. et al. (2015). The neural correlates of emotion regulation by implementation intentions. PLoS ONE, 10, e0119500. https://doi.org/10.1371/journal .pone.0119500 Heckhausen, H., & Gollwitzer, P. M. (1987). Thought contents and cognitive functioning in motivational versus volitional states of mind. Motivation and Emotion, 11, 101–120. https://doi.org/10.1007 /BF00992338 Heckhausen, H., Gollwitzer, P. M., & Weinert, F. E. (1987). Jenseits des Rubikon: Der Wille in den Humanwissenschaften [Beyond the Rubicon: The Will in the Humanities]. Heidelberg: Springer. Hügelschäfer, S., & Achtziger, A. (2014). On confident men and rational women: It’s all on your mind(set). Journal of Economic Psychology, 41, 31–44. https://doi.org/10.1016/j.joep.2013.04.001 Keller, L., & Gollwitzer, P. M. (2017). Mindsets affect risk perceptions and risk-taking behavior: Optimistic bias and the BART. Social Psychology, 48, 135–147. https://doi.org/10.1027/1864-9335/ a000304 Lejuez, C. W., Aklin, W. M., Jones, H. A. et al. (2003). The Balloon Analogue Risk Task (BART) differentiates smokers and nonsmokers. Experimental and Clinical Psychopharmacology, 11, 26–33. https://doi .org/10.1037/1064-1297.11.1.26 Lejuez, C. W., Aklin, W. M., Zvolensky, M. J., & Pedulla, C. M. (2003). Evaluation of the Balloon Analogue Risk Task (BART) as a predictor of adolescent real-word risk-taking behaviours. Journal of Adolescence, 26, 475–479. https://doi .org/10.1016/S0140-1971(03)00036–8

87

Lejuez, C. W., Read, J. P., Kahler, C. W. et al. (2002). Evaluation of a behavioral measure of risk taking: The Balloon Analogue Risk Task (BART). Journal of Experimental Psychology: Applied, 8, 75–84. https://doi.org/10.1037//1076-898X.8.2.75 Lewin, K., Dembo, T., Festinger, L., & Sears, P. S. (1944). Level of aspiration. In J. M. Hunt (Ed.), Personality and the Behavior Disorders (pp. 333–378). Oxford: Ronald Press. Marquardt, M. K., Cohen, A.-L., Gollwitzer, P. M., Gilbert, S. J., & Dettmers, C. (2017). Making if-then plans counteracts learned non-use in stroke patients: A proof-of-principle study. Restorative Neurology and Neuroscience, 35, 537–545. https://doi.org/10.3233/RNN-170748 Martin, J., Sheeran, P., Slade, P., Wright, A., & Dibble, T. (2011). Durable effects of implementation intentions: Reduced rates of confirmed pregnancy at two years. Health Psychology, 30, 368–373. https://doi.org/10.1037 /a0022739 McCrea, S. M., & Vann, R. J. (2018). Postactional goal pursuit: Consequences of task completion for thought content, affect, and behavioral intentions. Motivation and Emotion, 42, 852–870. https://doi .org/10.1007/s11031-018-9713-3 Mendoza, S. A., Gollwitzer, P. M., & Amodio, D. M. (2010). Reducing the expression of implicit stereotypes: Reflexive control through implementation intentions. Personality and Social Psychology Bulletin, 36, 512–523. https://doi.org /10.1177/0146167210362789 Nenkov, G. Y., & Gollwitzer, P. M. (2012). Pre- versus postdecisional deliberation and goal commitment: The positive effects of defensiveness. Journal of Experimental Social Psychology, 48, 106–121. https://doi.org/10.1016/j.jesp.2011.08.002 Neter, E., Stein, N., Barnett-Griness, O., Rennert, G., & Hagoel, L. (2014). From the bench to public health: Population-level implementation intentions in colorectal cancer screening. American Journal of Preventive Medicine, 46, 273–280. https://doi.org/10.1016/j.amepre.2013 .11.008 Oettingen, G. (2014). Rethinking Positive Thinking: Inside the New Science of Motivation. New York: Penguin Random House.

88

LUCAS KELLER, PETER M. GOLLWITZER, AND PASCHAL SHEERAN

Oettingen, G. (2019). WOOP my life. Website. www .woopmylife.org Orbell, S., & Sheeran, P. (2000). Motivational and volitional processes in action initiation: A field study of the role of implementation intentions. Journal of Applied Social Psychology, 30, 780–797. https://doi .org/10.1111/j.1559-1816.2000.tb02823.x Puca, R. M. (2001). Preferred difficulty and subjective probability in different action phases. Motivation and Emotion, 25, 307–326. https://doi.org/10 .1023/A:1014815716476 Schweiger Gallo, I., Keil, A., McCulloch, K. C., Rockstroh, B., & Gollwitzer, P. M. (2009). Strategic automation of emotion regulation. Journal of Personality and Social Psychology, 96, 11–31. https://doi.org/10.1037/a0013460 Sheeran, P., Webb, T. L., & Gollwitzer, P. M. (2005). The interplay between goal intentions and implementation intentions. Personality and Social Psychology Bulletin, 31, 87–98. https://doi.org/10 .1177/0146167204271308 Stewart, B. D., & Payne, B. K. (2008). Bringing automatic stereotyping under control: Implementation intentions as efficient means of thought control. Personality and Social Psychology Bulletin, 34, 1332–1345. https://doi .org/10.1177/0146167208321269 Taylor, S. E., & Gollwitzer, P. M. (1995). Effects of mindsets on positive illusions. Journal of Personality and Social Psychology, 69, 213–226. https://doi.org/10.1037/0022-3514.69.2.213 Thaler, R. H., & Sunstein, C. R. (2008). Nudge: Improving Decisions about Health, Wealth, and Happiness. New Haven, CT: Yale University Press. Toli, A., Webb, T. L., & Hardy, G. E. (2016). Does forming implementation intentions help people with mental health problems to achieve goals? A meta-analysis of experimental studies with clinical and analogue samples. British Journal of

Clinical Psychology, 55, 69–90. https://doi.org/10 .1111/bjc.12086 Webb, T. L., Schweiger Gallo, I., Miles, E., Gollwitzer, P. M., & Sheeran, P. (2012). Effective regulation of affect: An action control perspective on emotion regulation. European Review of Social Psychology, 23, 143–186. https://doi.org/10.1080 /10463283.2012.718134 Webb, T. L., & Sheeran, P. (2007). How do implementation intentions promote goal attainment? A test of component processes. Journal of Experimental Social Psychology, 43, 295–302. https://doi.org/10.1016/ j.jesp.2006.02.001 Webb, T. L., & Sheeran, P. (2008). Mechanisms of implementation intention effects: The role of goal intentions, self-efficacy, and accessibility of plan components. British Journal of Social Psychology, 47, 373–395. https://doi.org/10.1348 /014466607X267010 Wieber, F., Gollwitzer, P. M., & Sheeran, P. (2014). Strategic regulation of mimicry effects by implementation intentions. Journal of Experimental Social Psychology, 53, 31–39. https://doi.org/10.1016/j.jesp.2014.02.002 Winterich, K. P., & Nenkov, G. Y. (2015). Save like the Joneses: How service firms can utilize deliberation and informational influence to enhance consumer well-being. Journal of Service Research, 18, 384–404. https://doi.org/10.1177 /1094670515570268 WWC (What Works Clearinghouse). (2014). WWC Procedures and Standards Handbook (Version 3.0). US Department of Education, Institute of Education Sciences, National Center for Education Evaluation and Regional Assistance, and What Works Clearinghouse. https://ies.ed.gov /ncee/wwc/Docs/referenceresources/wwc_proce dures_v3_0_standards_handbook.pdf

7

Changing Behavior Using the Health Action Process Approach Ralf Schwarzer and Kyra Hamilton

Practical Summary When it comes to changing behavior, the first question to address is the level of motivation an individual possesses to attain a target behavior. If a person has not yet formed an intention to perform the behavior in question, then the individual is not currently motivated to change and is in need of motivational support. According to the health action process approach (HAPA), this support could involve motivational constructs such as action self-efficacy, positive outcome expectancies (benefits of behavior change), and some risk awareness for not changing. On the other hand, if a person has already formed an intention to participate in the behavior in question, then the individual is already motivated to attain the target behavior. Such individuals are unlikely to benefit from motivational support but are likely in need of behavioral support to overcome barriers that prevent them translating their good intentions into action. This support could involve volitional constructs such as coping self-efficacy, planning, and selfmonitoring. Moreover, if behavior is derailed and the person relapses to their previous pattern of behavior, instilling optimistic self-beliefs for successful reinitiation of action such as building recovery self-efficacy may be an effective strategy.

7.1 Introduction A key question in behavior change research is how to predict and modify the adoption and maintenance of the target behaviors. People have, in principle, control over their conduct. Accordingly, it is assumed that people are able to modify their behavior to change health-compromising behaviors (e.g., inactivity, snacking, alcohol misuse) through appropriate self-regulatory efforts and instead adopt health-enhancing alternatives (e.g., regular exercise, preventive nutrition, dental hygiene, condom use, vaccination). Although people do have this control and are able to change their behavior, there are still substantive problems with lack of adherence to health promoting behaviors and failure to give up health-compromising behaviors.

The problems likely involve a breakdown in capacity to self-regulate behavior and limitations in the skills necessary to do so. Health self-regulation refers to the motivational, volitional, and behavioral processes of abandoning health-risk behaviors in favor of adopting and maintaining health-enhancing behaviors. This chapter outlines the key theoretical constructs of a health behavior change model, the health action process approach (HAPA) describes how the model has been used to change behavior, and provides empirical evidence to serve as illustrations in the use of the HAPA to change behavior. The model has been designed as a general framework https://doi.org/10.1017/9781108677318.007

90

RALF SCHWARZER AND KYRA HAMILTON

to conceptualize health self-regulation as a process that can be subdivided into phases in which different psychological constructs work best to move people forward to improved health behaviors. This framework builds on older approaches that distinguish between a motivational phase and a volitional phase of change, characterizing individuals’ mindsets toward taking action. The level of behavioral intention can be seen as a separator between these phases, identifying less motivated persons in the first phase and more motivated persons in the second phase. The idea is that the factors that move people forward differ between the two phases. The model has been used to inform behavior change in many health behaviors and contexts but can also be an appropriate means to inform behavior change outside health domains such as academic performance, goal striving in sports, career development, and searching for a job.

effort. In the sections that follow, definitions and operationalization (how the constructs are used) of the constructs will be provided, the key mechanisms by which these constructs relate to behavior will be described, the evidence in support of the HAPA presented, and the limitations of the HAPA discussed.

7.3 Key Constructs of the HAPA The HAPA identifies six key constructs, which are proposed to be directly or indirectly related to behavior as the ultimate outcome. The HAPA psychological constructs are more or less important for either goal setting or goal pursuit and can be assigned to these two phases (see Table 7.1). The next sections provide definitions and operationalization of these constructs.

7.3.1 Intention

7.2 Overview of the Theory and Evidence The HAPA specifies six constructs that are considered the core antecedents of behavior: intention, risk perception, outcome expectancies, selfefficacy, planning, and action control (selfmonitoring). These constructs are proposed to be determinants of action through a set of proposed mechanisms. Together, the constructs and proposed mechanisms constitute the theoretical framework of the HAPA. Furthermore, and important to consider when designing interventions to change behavior, it is useful to distinguish phases of self-regulation and assess individuals according to their position within these phases. A useful distinction is the one between motivation and volition. In the motivational phase, individuals are in a deliberative mindset while setting a goal (intention), whereas, in the volitional phase, individuals are in an implementation mindset while pursuing their goal (see Chapter 6, this volume). Thus, goal setting and goal pursuit can be understood as two distinct processes that require self-regulatory

Changes in health behaviors can be influenced by opportunities and barriers, by explicit decisions, or by random events. Here, the discussion is constrained to intentional changes that happen when people become motivated to alter their previous way of life and set goals for a different course of action. For example, they may consider quitting smoking or they make an effort to do so. Thus, intention represents a key factor in health behavior change and behavior change more broadly. Table 7.1 Health action process approach (HAPA) constructs according to phases of behavior change Motivation Phase (Goal Setting)

Volition Phase (Goal Pursuit)

Personalized Risk Feedback Outcome Expectancies Action Self-Efficacy

Coping Self-Efficacy

Intention Formation, Goal Adjustment

Recovery Self-Efficacy Action Planning, Coping Planning Action Control

Changing Behavior Using the Health Action Process Approach

This construct was originally proposed by Fishbein and Ajzen (1975) based on Lewin’s (1951) seminal work on the topic, as outlined in the theory of reasoned action (see Chapter 2, this volume), to operate as a mediator between attitudes and behavior – theorizing that emanated from well-documented research that showed relatively modest relations between attitudes and behavior (e.g., Wicker, 1969). Empirical evidence has shown intention to be a stronger proximal predictor of behavior than attitudes for many behaviors and, thus, an indispensable variable when it comes to predicting behaviors (Abraham & Sheeran, 2000). In the process of motivation, intention has been regarded as a kind of “watershed moment” between an initial goal setting phase (motivational phase) and a subsequent goal pursuit phase (volitional phase) (see Chapter 6, this volume). Although the construct of intention is indispensable in explaining behavior change, it has some inherent limitations as a predictor of behavior. Meta-analytic research has shown relatively modest correlations for the intention-behavior link (Abraham & Sheeran, 2000; Sheeran & Webb, 2016; see also Chapter 6, this volume). When trying to translate intentions into behavior, individuals are faced with various obstacles such as distractions, forgetting, or competing bad habits. If not equipped with means to meet these obstacles, intention alone is not sufficient to change behavior. To overcome this limitation, further constructs are required that operate in concert with the intention.

7.3.2 Risk Perception Perceiving a health threat is often viewed as an important prerequisite for individuals to be motivated to change their risky behavior (Renner & Schupp, 2011). If an individual is unaware of the risky nature of their actions, they would not be motivated to change. Risk perception has two aspects: perceived severity of a health condition and personal vulnerability toward it (see

91

Chapter 4, this volume). The first refers to the amount of harm that might occur and the second pertains to the subjective probability that one could fall victim to that condition. Thus, it has been recommended that people should be informed about the existence of a health risk and, moreover, that they should imagine themselves as possible victims if they do not take the necessary precautions. Using fear strategies that aim to scare people into health behaviors, however, has not been shown to be effective unless other methods are used in conjunction with fear strategies (for a review, see Kok et al., 2018; see also Chapters 4 and 34, this volume). In general, initial risk perceptions may be sufficient to put some people on track for developing a motivation to change but, later on, other variables may be more influential in the self-regulation process.

7.3.3 Outcome Expectancies People not only need to be aware of the existence of a health threat; they also need to know how to regulate their behavior by understanding the contingencies between their actions and subsequent outcomes. These outcome expectancies are influential beliefs in the motivation to change (Bandura, 1997; see also Chapter 3, this volume). For example, a smoker may find more good reasons to quit than good reasons to continue smoking (e.g., “If I quit smoking then I will save money”). Similarly, positive outcome expectancies (e.g., “If I exercise five times per week, I will reduce my cardiovascular risk”) are chiefly seen as being important in the motivation phase, when a person balances the pros and cons of certain behavioral outcomes. Ambivalence in positive and negative outcome expectancies is typical in rational decision-making and has been proposed in other models of behavior prediction to underpin attitudinal beliefs (see Chapter 2, this volume). Ambivalence may not lead directly to action but can help to form an intention to perform a given behavior.

92

RALF SCHWARZER AND KYRA HAMILTON

7.3.4 Self-Efficacy Risk perceptions and outcome expectancies make it more likely that the individual will set goals and form an intention to perform a behavior to address the risk or achieve the identified outcomes. Perceived self-efficacy also contributes to the intention formation but has also been proposed to be important at all stages of the health behavior change process (Bandura, 1997; see also Chapter 3, this volume). Perceived self-efficacy reflects individuals’ beliefs in their capabilities to exercise control over challenging demands and over their own functioning. Although perceived self-efficacy has been found to be important for both intention formation and behavioral action (Hamilton, Vayro, & Schwarzer, 2015; Lhakhang et al., 2016; Zhang et al., 2019, 2020), it does not always constitute exactly the same construct. Its meaning depends on the particular situation of individuals who may be more or less advanced in the change process. The rationale for the distinction between several phase-specific self-efficacy beliefs is that, during the course of health behavior change, different tasks have to be mastered and different selfefficacy beliefs are required to master these tasks successfully. For example, individuals might be confident in their capability to be physically active in general (i.e., high action self-efficacy) but might not be very confident to resume physical activity after a setback (i.e., low recovery self-efficacy). In the HAPA, three types of self-efficacy are distinguished: action self-efficacy, coping self-efficacy, and recovery self-efficacy. Action self-efficacy, sometimes also referred to as pre-action self-efficacy or task self-efficacy, refers to the first phase of the process in which an individual does not yet act but develops a motivation to do so. It is an optimistic belief during the preactional (motivational) phase. Individuals high in action self-efficacy imagine success, anticipate potential outcomes of diverse strategies, and are more likely to initiate a new behavior. Those with less self-efficacy imagine failure, harbor self-

doubts, and tend to procrastinate. Coping selfefficacy, sometimes also called maintenance selfefficacy, represents optimistic beliefs about one’s capability to cope with barriers that arise during the period of behavioral maintenance. A new health behavior might turn out to be much more difficult to adhere to than expected but a self-efficacious person responds confidently with better strategies, more effort, and prolonged persistence to overcome such hurdles. Once an action has been taken, individuals with high coping self-efficacy try harder and persist longer than those who are less self-efficacious. Recovery self-efficacy addresses the experience of failure and recovery from setbacks. It pertains to one’s conviction to get back on track after being derailed; the person trusts in their competence to regain control after a setback or failure and to reduce harm. This distinction between phase-specific selfefficacy beliefs has proven useful in various domains of behavior change where action selfefficacy has tended to predict intentions, whereas coping self-efficacy and recovery self-efficacy have tended to predict behaviors such as breast selfexamination (Luszczynska & Schwarzer, 2003), dietary behaviors (Ochsner, Scholz, & Hornung, 2013; Schwarzer & Renner, 2000), depression prevention (Zarski et al., 2018), and physical exercise (Scholz, Sniehotta, & Schwarzer, 2005).

7.3.5 Action Planning and Coping Planning As discussed in the previous section, intention alone is not sufficient to change behavior. The HAPA thus proposes that intentions are more likely to be translated into behaviors when people anticipate detailed plans, imagine success scenarios, and develop preparatory strategies for tackling a challenging task (Schwarzer, 2016). Action planning and coping planning are theorized as proximal determinants of behavior and distinct mediators likely to ensure intentions are translated to behavior, previously referred to as a dual mediation model (Carraro & Gaudreau, 2013; see also

Changing Behavior Using the Health Action Process Approach

Chapter 6, this volume). This proposition is a central tenet of the HAPA (Schwarzer, 2008), and ample prior research has found support for planning mediating the intention-behavior relationship (Hamilton, Bonham et al., 2017; Hamilton et al., 2020; Hamilton, Cox, & White, 2012; Hattar, Pal, & Hagger, 2016; Reyes Fernández et al., 2016), with intervention studies also supporting complementary effects of action planning and coping planning (Kwasnicka et al., 2013; Zhou et al., 2015). Both action planning and coping planning are based on contingencies with anticipated situations. For example, an individual might state the following action plan: “I plan to run with my friend on Sunday at 10 a.m. for half an hour in the park without pausing.” The plan includes a number of situational cues and sufficient detail to qualify as a plan, going beyond a mere behavioral intention, such as “I intend to go jogging once a week.” Such an action plan is often called a when-where-how plan. The time and day of week and the presence of the friend constitute the cues that are proposed to trigger the behavior. Other cues can be stronger and more explicit, such as “If I arrive at home after work today before 5 p.m., then I will immediately go jogging in the park.” There is a long tradition of research on such action plans with if-then structures for health behaviors (see Chapters 6 and 39, this volume). The example items on physical activity presented include a level of uncertainty. This is because the conditions for performing the behavior might be unfavorable, such as bad weather, physical discomfort, a traffic jam, or a visiting friend, preventing the person from actually executing the plan. To account for such barriers, the concept of coping planning, initially known as barrier-related strategic planning, was developed (Sniehotta et al., 2005). Coping planning is proposed as a conceptually distinct construct from action planning; action plans are proposed to connect the individual with good opportunities to act through a task-facilitating strategy (i.e., making

93

plans that specify when, where, and how an intended behavior is to be performed), whereas coping plans are proposed to protect good intentions from anticipated obstacles via a distraction-inhibiting strategy (i.e., making plans that anticipate challenging situations that may obstruct behavioral enactment and mental representation of ways to overcome them). Empirical literature supports the conceptual distinction between the two planning constructs (Scholz et al., 2008; see meta-analyses by Carraro, & Gaudreau, 2013; Kwasnicka et al., 2013), although it should be noted that there is more convergent than discriminant validity of these two concepts but keeping them distinct is useful for the design of interventions (see Chapter 39, this volume). Coping plans serve a compensatory function in the HAPA. If one’s first-choice plan becomes unrealistic due to anticipated barriers or better options to attain one’s goal, coping plans contain several alternative responses identified beforehand that could be retrieved and adopted. An example is: “If I feel tired of studying, then I will go to the kitchen and prepare a coffee; however, if it is already after 6 p.m. and I don’t want to risk not falling asleep later, then I’d rather go for a refreshing walk outside to maximize goal attainment of studying for the exam.” Behavioral interventions in which individuals are prompted to produce a number of well-elaborated coping plans to make goal attainment more likely have been shown to be useful in promoting behavioral engagement (Hagger & Luszczynska, 2014). This pertains to individuals who are able to vividly imagine and forecast possible scenarios, as well as anticipate barriers and opportunities, and who are capable of understanding the contingencies. Action planning and coping planning are alterable variables. They can be easily communicated to individuals with self-regulatory deficits and, for this reason, have been frequently applied in interventions to change health behaviors. Further, it is suggested that the advantages of planning interventions include low cost and response burden

94

RALF SCHWARZER AND KYRA HAMILTON

(Hagger & Luszczynska, 2014). Numerous randomized controlled trials have documented the evidence in favor of such planning interventions (for a review, see Hagger & Luszczynska, 2014).

7.3.6 Action Control Action control is a self-regulatory strategy for promoting maintenance of an enacted behavior through the continual monitoring and evaluation of a behavior against a desired behavioral standard (Schwarzer, 2008). Unlike self-efficacy and planning strategies, which are generally elicited prior to a behavior being enacted, action control is performed retrospectively or concurrently with a behavior each time it is repeated, where the ongoing behavior is continuously evaluated with regard to a behavioral standard (Schwarzer, 2008). Action control comprises three facets: selfmonitoring (e.g., “I consistently monitor when, where, and how long I exercise”), awareness of standards (e.g., “I have always been aware of my prescribed training program”), and self-regulatory effort (e.g., “I took care to practice as much as I intended to”). Action control strategies can include keeping records of behaviors in the form of a diary or checkmarks on a calendar, which can make people more aware of their behavioral gains and deficits and, therefore, encourages continued action or alternative action if needed (Schwarzer, 2008). Action control is a useful behavior change technique that has been applied to a variety of health behaviors, including oral health behaviors (Hamilton et al., 2018; Schwarzer, Antoniuk, & Gholami, 2015; Zhou et al., 2015), dietary habits (Godinho, Alvarez, Lima, & Schwarzer, 2014), dust mask wearing (Zhou et al., 2016), or hand hygiene (Reyes Fernández et al., 2016).

7.4 Mechanisms: A Self-Regulation Framework Other social cognitive models such as the theory of planned behavior (see Chapter 2, this volume)

have been criticized for not addressing the intention-behavior “gap,” that is, the relatively modest link between intention and behavior (Sheeran & Webb, 2016; see also Chapter 6, this volume). As a comprehensive self-regulation model, the HAPA suggests a distinction between preintentional motivation processes that lead to a behavioral intention and post-intentional volition processes that lead to actual behavior (see Figure 7.1). In the following sections, the processes that underpin the HAPA are described and key issues such as goal setting and the goal striving phase of health self-regulation are further introduced (Schwarzer, 1992, 2008). Before changing behavior, individuals need to become motivated. This is seen as a process toward attaining an explicit goal or intended action (e.g., “I intend to quit smoking this week”). Three sets of social cognitive constructs are implicated as playing key roles in this intention formation process, namely outcome expectancies, action self-efficacy, and risk perceptions, and all three are proposed to have direct effects on intention. After an individual has become committed to their goal, they are proposed to move on to a volitional phase where they undergo necessary processes to prepare for action and, later, maintenance of the behavior, particularly in the face of barriers and setbacks. In this phase, the “good intention” has to be transformed into detailed instructions on how to perform the desired action and, once an action has been initiated, it has to be maintained. This involves self-regulatory beliefs, skills, and strategies such as coping self-efficacy and recovery self-efficacy, planning, and action control (self-monitoring) that help protect one’s goal pursuit from distracting or tempting situations. Coping and recovery self-efficacy are required to overcome obstacles that might derail the intended action, to overcome setbacks and recover from failed attempts to enact the target behavior, and to stimulate self-motivation repeatedly. Coping and recovery self-efficacy are

Changing Behavior Using the Health Action Process Approach

Coping Self-Efficacy

Action Self-Efficacy

Outcome Expectancies

Recovery Self-Efficacy

Action Planning Intention Coping Planning

Risk Perception

Preintenders

95

Target Behavior

Action Control

Intenders

Actors

Figure 7.1 The health action process approach (HAPA; Schwarzer 1992, 2008)

proposed to have direct effects on behavior and are also expected to be related to each other and to action self-efficacy. The forms of self-efficacy in the HAPA are, therefore, phase-specific, with action self-efficacy relevant to intention formation and coping and recovery self-efficacy implicated in the enactment and maintenance of behavior. Plans are required to specify the details to performance of the desired action and, when the critical situation arises, individuals take the initiative and invest in preparatory behaviors. Action and coping planning are thus proposed to mediate the intention-behavior relationship in the HAPA. Action control is required to help focus individuals’ attention on the task at hand, while avoiding focusing attention on distractors, resisting temptations, and managing negative emotions, and proposed to have direct effects on behavior. In sum, the volitional process can be viewed as a set of sequences that involve planning, initiation, maintenance, and relapse management. This process to health action is not achieved through an act of will but involves the development of self-regulatory skills and

strategies that influence an individual’s motivation and behaviors, such as the setting of attainable, proximal subgoals; creating incentives; drawing from an array of coping options; monitoring progress; and mobilizing social support. The purpose of the process model described and shown in Figure 7.1 is twofold: It allows a prediction of behavior and it explains the assumed causal mechanism of behavior change. Research that is based on HAPA, therefore, often employs path-analytic methods to test model predictions. There are a host of empirical studies that have applied the HAPA and confirmed its usefulness (for an overview, see Schwarzer & Luszczynska, 2015). However, it should be noted that there is not always a perfect match between the model and realworld applications. Owing to a variation in research questions and contextual constraints, there are often more parsimonious versions of the HAPA aiming at the examination of only certain aspects of the model. In some cases, for example, there has been no sufficient discriminant validity between action planning and coping planning and, thus, collapsing these two facets into one construct of planning has

96

RALF SCHWARZER AND KYRA HAMILTON

been preferred (Zhou et al., 2016). In other cases, there has been no sufficient discriminant validity between coping self-efficacy and recovery selfefficacy and, therefore, both were combined into an overall construct labeled volitional self-efficacy (Zhang et al., 2019).

7.5 How Has the Theory Been Used to Change Behavior? The HAPA proposes multiple targets for interventions and suggests that intervention components targeting key constructs in each phase may assist in intention formation and behavioral enactment and maintenance. The model has been used as a guide to develop intervention content aimed at changing individual constructs (e.g., Lippke et al., 2010; Payaprom et al., 2011) as well as multiple constructs from the HAPA (e.g., Duan et al., 2017; Keller et al., 2018; Lhakhang et al., 2015). As HAPA overlaps with other social cognitive models, and the use of more integrated models of behavior is becoming more popular (Brown et al., 2018; Hagger et al., 2016; Hagger et al., 2017; Hamilton, Kirkpatrick et al., 2017; see also Chapters 12 and 15, this volume), one needs to decide to what degree an intervention study is inspired, guided, or determined by a particular theory. Key elements of the HAPA are phase-specific self-efficacy (action, coping, and recovery selfefficacy) and one or two forms of specific planning (action and coping planning). The theory lends itself to longitudinal mediator research designs that typically also include more pervasive motivational constructs such as intention, outcome expectancies, and risk perception. Thus, one could argue that a study does not qualify as a HAPA study if these key elements (one type of self-efficacy, one type of planning) are missing (see Zhang et al., 2019). Accordingly, there are many studies that claim to be based on the HAPA but, in reality, are simply inspired by the HAPA.

In intervention studies that have targeted HAPA constructs, various behavior change methods or techniques (Kok et al., 2016; Michie et al., 2013; see also Chapter 20, this volume) have been applied to target participants’ risk perceptions, outcome expectancies, and intentions in the motivational phase, whereas the self-regulatory components of self-efficacy, action planning, coping planning, and action control (self-monitoring) constitute the behavioral support to initiate and maintain the desired health behaviors. Strategies that provide motivational support target intention formation as the primary outcome, whereas strategies that provide behavioral support target action as the primary outcome, in line with the motivation and volition phases of the model. Most intervention studies concentrate on the mechanisms and constructs that characterize the continuum layer of HAPA, with few intervention studies making use of the stage layer. For an example of a study protocol of a HAPA-based intervention, see O’Brien et al. (2018). A recent study tested the efficacy of an mHealth program based on the HAPA using the social media platform Telegram to promote good oral hygiene behavior and oral health outcomes among Iranian adolescents (Scheerman et al., 2019; see also Chapter 29, this volume). Results demonstrated that the oral health intervention compared to the control group resulted in significant improvements in toothbrushing behavior and clinical oral health indicators as well as more positive social cognitions (intention, outcomeexpectancies, risk perception, self-efficacy, action planning, coping planning, self-monitoring, and health-related quality of life) among Iranian adolescent students in the short and long term (for a detailed description of this study, see Sidebar 7.1). The focus of many intervention studies mainly lies on the volition phase, which makes HAPA distinct from other models, and most studies target directly the initiation and maintenance of health behaviors, employing behavior change techniques that are likely to serve this purpose. For example, interventions to improve self-efficacy may seek to

Changing Behavior Using the Health Action Process Approach

97

Sidebar 7.1 A health action process approach intervention in oral health

Based on the HAPA, Scheerman et al. (2019) tested the efficacy of a theory-based online program using Telegram to promote good oral hygiene behavior and oral health outcomes among Iranian adolescents. A three-arm randomized controlled trial design was used, consisting of an adolescent (A)–only intervention group (A group; n = 253), an adolescent (A) and mother (M) intervention group (A + M group; n = 260), and a control group (n = 278). The program targeted multiple behavior change techniques that mapped onto the constructs in the motivational and volitional phases of the HAPA. Specifically, intervention components targeted outcome expectancies (e.g., adolescents were encouraged to formulate their own potential pros and cons of regular toothbrushing), self-efficacy (e.g., instructions were given on effective behaviors as well as role modeling of effective behaviors), risk perceptions (e.g., feedback on oral health behavior change was provided), social influences (e.g., mothers were asked to encourage their adolescent to complete all intervention activities), action planning (e.g., adolescents were asked to make concrete plans on when, where, and after what activity they would brush their teeth in future using the if-then formulation), coping planning (e.g., adolescents were asked to identify barriers and possible solutions by making coping plans), and action control (e.g., adolescents were asked to monitor their oral hygiene behavior and oral health status) (see Appendix 7.2, supplemental materials). Findings showed increases in adolescent toothbrushing at the one- and six-month follow-ups in both intervention groups compared to the control group. Adolescents in the A + M group showed significant greater improvements in their toothbrushing behavior and in scores on two clinic-verified indicators of dental health, the visual plaque index and the community periodontal index, than adolescents in the A group. Improvements to toothbrushing social cognitions were also observed.

apply strategies that target change in action selfefficacy and coping (maintenance) self-efficacy. These can include providing opportunities to experience, reflecting on past success, developing skills, setting proximal goals, monitoring goal progress, providing encouraging feedback on progress, enhancing skills to manage setbacks, and presenting role models to provide vicarious experience of success with the behavior. Brief interventions of this kind provide participants with planning forms where they enter the “when,” “where,” and “how” of intended actions and concurrently generate several coping plans, including imagined barriers and ways to overcome them. In addition, daily diary forms or calendars are provided to allow for continuous self-monitoring (Gholami et al., 2013;

Keller et al., 2018; Lhakhang et al., 2014; Schwarzer et al., 2015). Moreover, role models can be introduced by displaying testimonials of others who have coped well (for self-efficacy improvement). A description of how the HAPA psychological constructs are operationalized in terms of techniques and concrete messages that constitute the treatment components is presented in Appendix 7.1 (supplemental materials).

7.6 Evidence Base for Use of the HAPA in Changing Behavior If an article reports a combination of one type of planning (action or coping planning) and one type of self-efficacy (task, coping/maintenance, or

98

RALF SCHWARZER AND KYRA HAMILTON

recovery self-efficacy), mediating between intervention and behavioral outcome, then it is considered a HAPA study, even if the model or its source is not mentioned. In other cases, where fewer of these constructs are investigated (e.g., planning and self-efficacy as mediators), it may not be clear whether the study has been derived from HAPA or not. As the HAPA is an openarchitecture framework, it tends to inspire research that is not necessarily in line with the original model and, in many cases, published reports address only a narrow aspect that is in line with specific research questions, but not providing a full account of the entire model that had informed the study. This is a problem endemic in the literature on behavior change and an issue that researchers are attempting to resolve through greater transparency in reporting and clearer specification of matching theory components with the specific techniques of the intervention that are purported to change them (see Chapter 20, this volume). Meta-analyses of HAPA-derived intervention studies have been conducted (Smith et al., 2013). The authors found that there are many interventions on a narrow number of chronic disease–related behaviors such as weight loss adopting the model and that the inconsistencies of terminology, definitions, and reporting make it difficult to identify to what degree studies are really based on HAPA. A more recent metaanalysis on the HAPA has been generated (Zhang et al., 2019). Studies were included if they measured intention and a health behavior and at least one type of self-efficacy (action/task, coping/maintenance, or recovery self-efficacy) and one type of planning (action or coping planning). Studies that met such key inclusion criteria have often also used risk perceptions, outcome expectancies, and, less frequently, action control, and they were likely to be inspired by the overall longitudinal mediator mechanism that is unique to HAPA. Results indicated positive associations among HAPA constructs across studies with

small-to-medium effect sizes. The majority of studies identified were correlational in design with few intervention studies targeting change in individual HAPA components. Action selfefficacy and coping self-efficacy, as well as outcome expectancies, had small-to-medium–sized effects on health behaviors. Effects of selfefficacy and outcome expectancies on health behaviors were mediated by intentions and planning. Effects of action self-efficacy on intentions and behavior were larger in physical activity studies compared to studies on dietary behaviors, whereas effects of coping and recovery self-efficacy on behavior were larger in studies on dietary behaviors. Findings highlight the importance of self-efficacy in predicting health behavior in motivational and volitional phases of behavior change. Importantly, risk perception was the only exception to the predictions of the model in the meta-analysis, with much smaller effects observed. This is not surprising, because risk perceptions generally do not have a pervasive influence on health-related behaviors unless they have a clear, explicit, and proximal link to reduced risk (e.g., taking prophylactic medication, safety behaviors, vaccination) or are pertinent to a specific population at risk such as patients in medical rehabilitation. Although there is an obvious potential of action control as an independent predictor of behavior, only ten of the studies in this meta-analysis provided data to compute effect sizes, which was insufficient for them to be included in the test of the model. What has not been identified to date is a meta-analysis on interventions that address the main constructs along with the stage approach. Such interventions segment the audience into pre-intenders, intenders, and actors and design matched treatments focusing on self-efficacy, planning, and action control. This is an avenue for future meta-analyses as the literature on HAPA interventions expands.

Changing Behavior Using the Health Action Process Approach

99

Sidebar 7.2 Using the health action process approach (HAPA) and technology to change behavior

Most eHealth or mHealth applications address individuals in the volition phase of behavior change because people generally do not sign up for such “apps” if they do not have some level of motivation to change. In such cases, users need behavioral support to overcome obstacles and maintain behaviors based on planning, action control, and maintenance self-efficacy. Participants are guided to set goals, monitor their behavior, receive tailored feedback, make action plans and coping plans, and increase their self-efficacy by vicarious experience via testimonials and by problemsolving tasks (Hekler et al., 2018). The advanced technology allows for immediate feedback loops that keep users online, monitoring lapses as well as progress and providing resource information that is individually tailored (Spruijt-Metz et al., 2015). Technology allows for more fine-grained process characteristics to be included in the implementation of the model (Nahum-Shani et al., 2018). More frequent diagnostics allow the reclassification of individuals into concurrent stages, tailored for individual needs. Combining HAPA treatment components such as self-efficacy, planning, and action control in the volitional phase of behavior change seems to be promising in the context of advanced technology-driven interventions.

7.7 Conclusion The HAPA distinguishes between a motivational and a volitional phase of behavior change. It can be seen as a hybrid model combining the features of stage models and continuum models (see Chapter 6, this volume). In the motivational phase, outcome expectancies, action self-efficacy, and risk perceptions are factors that make the intention formation (goal setting) more likely. If a behavioral intention is not yet formed, then the individual is not sufficiently motivated to change and, therefore, in need of motivational support. In the volition phase, coping self-efficacy along with action planning and coping planning is important, with behavior maintenance supported by recovery self-efficacy and continuous action control (such as self-monitoring). In this phase, a behavioral intention has already been formed and the individual is motivated to attain a target behavior. Yet such individuals are in need of behavioral support to overcome barriers that prevent them translating their intentions into action. Behavioral intention operates as a bridge between the motivational and

volitional constructs. Intervention design considers the position of the individual along the change continuum and targets the appropriate components within each phase that are most likely to move the individual further toward goal attainment. In general, current evidence provides support for both the motivational and the volitional components of the HAPA, particularly the stage-specific selfefficacy constructs, on health behavior (Zhang et al., 2019). Results corroborate research applying the model to a variety of different health behaviors (Schwarzer & Luszczynska, 2015). In moving the field forward, future HAPA interventions may benefit from drawing on advanced information and communication technologies as reflected by eHealth or mHealth applications (see Chapter 29, this volume). An example of an mHealth application is presented in Sidebar 7.2.

References Abraham, C., & Sheeran, P. (2000). Understanding and changing health behaviour: From health beliefs to self-regulation. In P. Norman, C. Abraham, &

100

RALF SCHWARZER AND KYRA HAMILTON

M. Conner (Eds.), Understanding and Changing Health Behaviour (pp. 3–24). Amsterdam: Harwood. Bandura, A. (1997). Self-Efficacy: The Exercise of Control. New York: Freeman. Brown, D., Hagger, M. S., Morrissey, S., & Hamilton, K. (2018). Predicting fruit and vegetable consumption in long-haul heavy goods vehicle drivers: Application of a multi-theory, dual-phase model and the contribution of past behavior. Appetite, 121, 326–336. https://doi.org /10.1016/j.appet.2017.11.106 Carraro, N., & Gaudreau, P. (2013). Spontaneous and experimentally induced action planning and coping planning for physical activity: A meta-analysis. Psychology of Sport and Exercise, 14, 228–248. https://doi.org/10.1016/j .psychsport.2012.10.004 Duan, Y. P., Wienert, J., Hu, C., Si, G. Y., & Lippke, S. (2017). Web-based intervention for physical activity and fruit and vegetable intake among Chinese university students: A randomized controlled trial. Journal of Medical Internet research, 19, 4. https://doi.org/10.2196/jmir.7152 Fishbein, M., & Ajzen, I. (1975). Belief, Attitude, Intention and Behavior: An Introduction to Theory and Research. Reading, MA: Addison-Wesley. Gholami, M., Lange, D., Luszczynska, A., Knoll, N., & Schwarzer, R. (2013). A dietary planning intervention increases fruit consumption in Iranian women. Appetite, 63, 1–6. https://doi.org /10.1016/j.appet.2012.12.005 Godinho, C. A., Alvarez, M. J., Lima, M. L., & Schwarzer, R. (2014). Will is not enough: Coping planning and action control as mediators in the prediction of fruit and vegetable intake. British Journal of Health Psychology, 19, 856–870. https://doi.org/10.1111/bjhp.12084 Hagger, M. S., & Luszczynska, A. (2014). Implementation intention and action planning interventions in health contexts: State of the research and proposals for the way forward. Applied Psychology: Health and Well-Being, 6, 1–47. https://doi.org/10.1111/aphw.12017 Hagger, M. S., Sultan, S., Hardcastle, S. et al. (2016). Applying the integrated trans-contextual model to mathematics activities in the classroom and homework behavior and attainment. Learning and

Individual Differences, 45, 166–175. https://doi .org/10.1016/j.lindif.2015.11.017 Hagger, M. S., Trost, N., Keech, J., Chan, D. K. C., & Hamilton, K. (2017). Predicting sugar consumption: application of an integrated dual-process, dual-phase model. Appetite, 116, 147–156. https://doi.org/10 .1016/j.appet.2017.04.032 Hamilton, K., Bonham, M., Bishara, J., Kroon, J., & Schwarzer, R. (2017). Translating dental flossing intentions into behavior: A longitudinal investigation of the mediating effect of planning and self-efficacy on young adults. International Journal of Behavioral Medicine, 24, 420–427. https://doi.org/10.1007/s12529-016-9605-4 Hamilton, K., Cornish, S., Kirkpatrick, A., Kroon, J., & Schwarzer, R. (2018). Parental supervision for their children’s tooth brushing: Mediating effects of planning, self-efficacy, and action control. British Journal of Health Psychology, 23, 387–406. https://doi.org/10.1111/bjhp.12294 Hamilton, K., Cox, S., & White, K. M., (2012). Testing a model of physical activity among mothers and fathers of young children: Integrating self-determined motivation, planning, and theory of planned behavior. Journal of Sport & Exercise Psychology, 34, 124–145. https://doi.org/10.1123 /jsep.34.1.124 Hamilton, K., Kirkpatrick, A., Rebar, A., & Hagger, M. S. (2017). Child sun safety: Application of an integrated behavior change model. Health Psychology, 36, 916–926. https://doi.org/10.1037 /hea0000533 Hamilton, K., Vayro, C., & Schwarzer, R. (2015). Social-cognitive antecedents of fruit and vegetable consumption in truck drivers: A sequential mediation analysis. Journal of Nutrition Education and Behavior, 47, 379–384. https://doi.org/10.1016/j.jneb.2015.04.325 Hamilton, K., Ng, H., Zhang, C-Q., Phipps, D., & Zhang, R. (2020). Social psychological predictors of sleep hygiene behaviors in Australian and Hong Kong university students. International Journal of Behavioral Medicine. https://doi.org/10.1007/ s12529-020-09859-8 Hattar, A., Pal, S., & Hagger, M. S. (2016). Predicting physical activity-related outcomes in overweight and obese adults: A health action process

Changing Behavior Using the Health Action Process Approach

approach. Applied Psychology: Health and WellBeing, 8, 127–151. https://doi.org/10.1111/aphw .12065 Hekler, E., Rivera, D., Martin, C. et al. (2018). Tutorial for using control systems engineering to optimize adaptive mobile health interventions. Journal of Medical Internet Research, 20, e214. https://doi .org/10.2196/jmir.8622 Keller, J., Motter, S., Motter, M., & Schwarzer, R. (2018). Augmenting fruit and vegetable consumption by an online intervention: Psychological mechanisms. Appetite, 120, 348–355. https://doi.org/10.1016/j .appet.2017.09.019 Kok, G., Gottlieb, N. H., Peters, G.-J. Y. et al. (2016). A taxonomy of behavior change methods: An intervention mapping approach. Health Psychology Review, 10, 297–312. https://doi.org /10.1080/17437199.2015.1077155 Kok, G., Peters, G. J. Y., Kessels, L. T., Ten Hoor, G. A., & Ruiter, R. A. (2018). Ignoring theory and misinterpreting evidence: The false belief in fear appeals. Health Psychology Review, 12, 111–125. https://doi.org/10.1080/17437199 .2017.1415767 Kwasnicka, D., Presseau, J., White, M., & Sniehotta, F. F. (2013). Does planning how to cope with anticipated barriers facilitate health-related behaviour change? A systematic review. Health Psychology Review, 7, 129–145. https://doi.org/10 .1080/17437199.2013.766832 Lewin, K. (1951). Intention, will, and need. In D. Rapaport (Ed.), Organization and Pathology of Thought (pp. 95–153). New York: Columbia University Press. Lhakhang, P., Godinho, C., Knoll, N., & Schwarzer, R. (2014). A self-regulatory intervention increases fruit and vegetable intake: A comparison of two intervention sequences. Appetite, 82, 103–110. https://doi.org/10.1016/j.appet.2014.07.014 Lhakhang, P., Hamilton, K., Sud, N. et al. (2016). Combining self-management cues with incentives to promote interdental cleaning among Indian periodontal disease outpatients. BMC Oral Health, 16. https://doi.org/10.1186/s12903-016-0164-5 Lhakhang, P., Lippke, S., Knoll, N., & Schwarzer, R. (2015). Evaluating brief motivational and self-regulatory hand hygiene interventions: A

101

cross-over longitudinal design. BMC Public Health, 15. https://doi.org/10.1186 /s12889-015-1453-7 Lippke, S., Schwarzer, R., Ziegelmann, J. P., Scholz, U., & Schüz, B. (2010). Testing stage-specific effects of a stage-matched intervention: A randomized controlled trial targeting physical exercise and its predictors. Health Education and Behavior, 37, 533–546. https://doi.org/10.1177/1090198109359386 Luszczynska, A., & Schwarzer, R. (2003). Planning and self-efficacy in the adoption and maintenance of breast self-examination: A longitudinal study on self-regulatory cognitions. Psychology and Health, 18, 93–108. https://doi.org/10.1080 /0887044021000019358 Michie, S., Richardson, M., Johnston, M. et al. (2013). The behavior change technique taxonomy (v1) of 93 hierarchically clustered techniques: Building an international consensus for the reporting of behavior change interventions. Annals of Behavioral Medicine, 46, 81–95. https://doi.org /10.1007/s12160-013-9486-6. Nahum-Shani, I., Smith, S. N., Spring, B. J. et al. (2018). Just-in-time adaptive interventions (JITAIS) in mobile health: Key components and design principles for ongoing health behavior support. Annals of Behavioral Medicine, 52, 446–462. https://doi.org/10.1007/s12160-016-9830-8 O’Brien, J., Hamilton, K., Williams, A. et al. (2018). Improving physical activity, pain and function in patients waiting for hip and knee arthroplasty by combining targeted exercise training with behaviour change counselling: Study protocol for a randomised controlled trial. Trials, 19, 425. https://doi.org/10.1186/s13063-018-2808-z. Ochsner, S., Scholz, U., & Hornung, R. (2013). Testing phase-specific self-efficacy beliefs in the context of dietary behaviour change. Applied Psychology: Health and Well-Being, 5, 99–117. https://doi.org /10.1111/j.1758-0854.2012.01079.x Payaprom, Y., Bennett, P., Alabaster, E., & Tantipong, H. (2011). Using the health action process approach and implementation intention to increase flu vaccination uptake in high risk Thai individuals: A controlled before-after trial. Health Psychology, 30, 492–500. https://doi.org/10.1037/a0023580

102

RALF SCHWARZER AND KYRA HAMILTON

Renner, B., & Schupp, H. (2011). The perception of health risks. In H. Friedman (Ed.). The Oxford Handbook of Health Psychology (pp. 639–666). New York: Oxford University Press. Reyes Fernández, B., Knoll, N., Hamilton, K., & Schwarzer, R. (2016). Social-cognitive antecedents of hand washing: Action control bridges the planning–behaviour gap. Psychology and Health, 31, 993–1004. https://doi.org/10.1080 /08870446.2016.1174236 Scheerman, J. F. M., Hamilton, K., Sharif, M. O., Lindmark, U., & Pakpour, A. H. (2019). A social media-based intervention for oral health promotion among Iranian adolescents: A cluster randomized controlled trial. Psychology and Health. Advance online publication. https://doi .org/10.1080/08870446.2019.1673895 Scholz, U., Schüz, B., Ziegelmann, J. P., Lippke, S., & Schwarzer, R. (2008). Beyond behavioural intentions: Planning mediates between intentions and physical activity. British Journal of Health Psychology, 13, 479–494. https://doi.org/10.1348 /135910707X216062 Scholz, U., Sniehotta, F. F., & Schwarzer, R. (2005). Predicting physical exercise in cardiac rehabilitation: The role of phase-specific self-efficacy beliefs. Journal of Sport and Exercise Psychology, 27, 135–151. https://doi.org /10.1123/jsep.27.2.135 Schwarzer, R. (1992). Self-efficacy in the adoption and maintenance of health behaviors: Theoretical approaches and a new model. In R. Schwarzer (Ed.), Self-Efficacy: Thought Control of Action (pp. 217–243). Washington, DC: Hemisphere. Schwarzer, R. (2008). Modeling health behavior change: How to predict and modify the adoption and maintenance of health behaviors. Applied Psychology: An International Review, 57, 1–29. https://doi.org/10.1111/j.1464-0597 .2007.00325.x Schwarzer, R. (2016). Coping planning as an intervention component: A commentary. Psychology and Health, 31, 903–906. https://doi .org/10.1080/08870446.2016.1158260 Schwarzer, R., Antoniuk, A., & Gholami, M. (2015). A brief intervention changing oral self-care, self-efficacy, and self-monitoring. British Journal

of Health Psychology, 20, 56–67. https://doi.org /10.1111/bjhp.12091 Schwarzer, R., & Luszczynska, A. (2015). Health action process approach. In M. Conner & P. Norman (Eds.), Predicting Health Behaviours (3rd ed., pp. 252–278). Maidenhead: McGrawHill Open University Press. Schwarzer, R., & Renner, B. (2000). Social-cognitive predictors of health behavior: Action self-efficacy and coping self-efficacy. Health Psychology, 19, 487–495. https://doi.org/10.1037/0278-6133 .19.5.487 Sheeran, P., & Webb, T. L. (2016). The intention– behavior gap. Social and Personality Psychology Compass, 10, 503–518. https://doi.org/10.1111 /spc3.12265 Smith, J., Blockley, K., Murray, N., Greaves, C., Abraham, C., & Hooper, L. (2013). A Systematic Review of Intervention Studies Using Health Action Process Approach (HAPA) Model Components to Target Behaviours for Preventing and Managing Chronic Diseases. Protocol 23/5/ 13. PROSPERO International Prospective Register of Systematic Reviews, York University. https://www.crd.york.ac.uk/PROSPEROFILES/ 3596_PROTOCOL_20130424.pdf Sniehotta, F. F., Schwarzer, R., Scholz, U., & Schüz, B. (2005). Action planning and coping planning for long-term lifestyle change: Theory and assessment. European Journal of Social Psychology, 35, 565–576. https://doi.org/10.1002/ejsp.258 Spruijt-Metz, D., Hekler, E., Saranummi, N. et al. (2015). Building new computational models to support health behavior change and maintenance: New opportunities in behavioral research. Translational Behavioral Medicine, 5, 335–346. https://doi.org/10.1007 /s13142-015-0324-1 Wicker, A. W. (1969). Attitudes versus actions: The relationship of verbal and overt behavioral responses to attitude objects. Journal of Social Issues, 25, 41–78. Zarski, A.-C., Berking, M., Reis, D. et al. (2018). Turning good intentions into actions by using the health action process approach to predict adherence to internet-based depression prevention: Secondary analysis of a randomized controlled

Changing Behavior Using the Health Action Process Approach

trial. Journal of Medical Internet Research, 20, e9. https://doi.org/10.2196/jmir.88149 Zhang, C. Q., Zhang, R., Schwarzer, R., & Hagger, M. S. (2019). A meta-analysis of the health action process approach. Health Psychology, 38, 623–637. https://doi/org/10.1037/hea0000728 Zhang, C. Q., Fang, R., Zhang, R., Hagger, M. S., & Hamilton, K. (2020). Predicting hand washing and sleep hygiene behaviors among college students: Test of an integrated social-cognition model. International Journal of Environmental Research and Public Health, 17, 1209. https://doi.org/ 0.3390/ijerph17041209 Zhang, C. Q., Wong, M. C.-Y., Zhang, R., Hamilton, K., & Hagger, M. S. (2019). Adolescent sugar-sweetened beverage consumption: An extended health action process approach. Appetite, 141, 104332. https://doi .org/10.1016/j.appet.2019.104332

103

Zhou, G., Gan, Y., Hamilton, K., Schwarzer, R. (2017). The role of social support and self-efficacy for planning fruit and vegetable intake. Journal of Nutrition Education and Behavior, 49, 100–107. https://doi.org/10.1016/j .jneb.2016.09.005 Zhou, G., Gan, Y., Ke, Q., Knoll, N., Lonsdale, C., & Schwarzer, R. (2016). Avoiding exposure to air pollution by using filtering facemask respirators: An application of the Health Action Process Approach. Health Psychology, 35, 141–147. https://doi.org/10.1037 /hea0000264 Zhou, G., Sun, C., Knoll, N., Hamilton, K., & Schwarzer, R. (2015). Self-efficacy, planning, and action control in an oral self-care intervention. Health Education Research, 30, 671–681. https:// doi.org/10.1093/her/cyv032

8

Changing Behavior Using Self-Determination Theory Martin S. Hagger, Nelli Hankonen, Nikos L. D. Chatzisarantis, and Richard M. Ryan

Practical Summary According to self-determination theory, it is the quality rather than the quantity of motivation that counts when it comes to behavior change. The theory distinguishes between two general classes of motivation: autonomous and controlled. People are autonomously motivated when they feel that they have freely chosen or endorsed their actions, while people are controlled motivated when they feel their actions have been dictated by others or determined by pressures or events over which they have little or no control. Beyond autonomy, the theory argues that behaviors will be better internalized and maintained when they allow satisfaction of basic psychological needs to feel competent and related to others. People who experience a behavior as autonomously motivated, and who feel competent to act, are predicted to persist with that behavior and to experience positive outcomes like interest, enjoyment, and life satisfaction as well as vitality. Studies have indicated that interventions that provide training for significant others (e.g., parents, family members, managers, teachers, coaches) to display actions and language that support the need for autonomy, as well as interventions that promote support for competence and relatedness needs, promote autonomous motivation and behavior change maintenance.

8.1 Introduction Self-determination theory is a broad meta-theory that adopts a needs-based, organismic approach to understanding human behavior and attempts to understand the underlying needs and conditions within the individual that give rise to motivated behavior. In contrast to many social cognition and motivational theories (for examples, see Chapters 2

Martin S. Hagger’s contribution was supported by a Finnish Distinguished Professor (FiDiPro) award from Business Finland (1801/31/2015). https://doi.org/10.1017/9781108677318.008

and 4, this volume), the theory offers a uniquely different approach by focusing on the quality, rather than just the quantity, of motivation as the key determinant of behavior. According to Deci and Ryan (1985b), the originators and proponents of the theory, “Cognitive theories begin their analysis with … a motive, which is a cognitive representation of some future desired state. What is missing, of course, is the consideration of the conditions of the organism that makes these future states desired” (p. 228). The focus on the sources of one’s motivation and the relations of behavior to basic psychological needs are among the key assumptions that

Changing Behavior Using Self-Determination Theory

make self-determination theory distinct from social cognition theories (Ryan & Deci, 2017). The present chapter provides an overview of the theory and research evidence on the application of self-determination theory to behavior change. The chapter begins by providing a brief overview of self-determination theory and the component mini-theories of which the broader meta-theory consists. Next, the specific ways that the theory has been used to change behavior is described, with a particular focus on autonomy support, followed by a summary of research that has applied the theory to change behavior, and tested the theory-based mechanisms responsible, in multiple behaviors, populations, and contexts. The chapter also outlines future recommendations for self-determination theory-based approaches to behavior change.

8.2 A Brief Overview of SelfDetermination Theory 8.2.1 Self-Determination Theory: Origins and Mini-Theories Self-determination theory (Deci & Ryan, 1985b, 2000; Ryan & Deci, 2017) is a general metatheory of motivation with origins in theories of intention, personal causation and effectance, and competence. The concept of basic psychological needs and the distinction between autonomous and controlled forms of motivation are unifying concepts central to the theory and its predictions on human motivation. The theory comprises six interconnected “mini-theories,” each focusing on identifying key constructs and mechanisms that relate to particular aspects of motivation and its origins. The current chapter focuses on three of these six mini-theories that are especially pertinent to sustained behavior and behavior change: cognitive evaluation theory, organismic integration theory, and basic psychological needs theory. Accordingly, the next section introduces these three mini-theories and

105

provides a brief overview of each, along with some key evidence supporting their predictions. The premises of these key mini-theories are also summarized in Figure 8.1 and the figure should serve as a reference guide as each mini-theory is introduced. It is important to note that there are three additional mini-theories: causality orientations theory, goal contents theory, and relationship motivation theory. Causality orientations theory outlines how individual differences in three generalized, dispositional motivational orientations, autonomous, control, and impersonal, determine the type of motivation generally experienced by individuals across multiple behavioral domains (Deci & Ryan, 1985a). Goal contents theory suggests that the pursuit of intrinsic and extrinsic long-term aspirations yields different effects on wellness (Vansteenkiste, Lens, & Deci, 2006). Finally, relationship motivation theory, the newest of the mini-theories, focuses on the role of basic need support in maintaining high-quality relationships with others (Ryan & Deci, 2017).

8.2.2 Cognitive Evaluation Theory Cognitive evaluation theory concerns the concept of intrinsic motivation, a fundamental construct in self-determination theory. Intrinsic motivation is engaging in tasks or behaviors for their inherent satisfaction, without reliance on external reward contingencies or reinforcement. When individuals are intrinsically motivated to perform tasks or behaviors, they feel a sense of choice and personal effectance and derive a sense of interest, engagement, competence, and enjoyment from them. Some tasks or behaviors are inherently intrinsically motivating, such as puzzles, games, hobbies, and pastimes, but people can also become intrinsically motivated for new activities under conditions described within cognitive evaluation theory. The extent to which individuals engage a task or behavior out of intrinsic motivation is determined,

106

MARTIN S. HAGGER, NELLI HANKONEN, NIKOS L. D. CHATZISARANTIS, AND RICHARD M. RYAN

Need for Competence

Need for Autonomy

Need for Relatedness

Basic Psychological Need Satisfaction Type of motivation Position on perceived locus of causality continuum Perceived locus of causality scale

Defining features and reward contingencies

Degree of internalization

Intrinsic motivation

Extrinsic motivation

Autonomous motives (high autonomy)

Controlled motives (low autonomy)

Intrinsic motivation

Integrated regulation

Identified regulation

Introjected regulation

External regulation

Acting for enjoyment, pleasure and fun; no discernible reinforcement or reward

Behaviors that are fully incorporated into the repertoire of behaviors that satisfy psychological needs

Acting for personallyheld values such as learning new skills; internallyreinforced contingency

Acting for avoiding external sources of disapproval or gaining externally referenced approval

Acting for external reinforcement such as gaining rewards or avoiding punishment

Integrated

High internalization

Amotivation

Amotivation

Acting for no clear reason or rationale; lack of intentionality and personal causation

Low internalization

Figure 8.1 Diagram summarizing three key mini-theories of self-determination theory: Cognitive evaluation theory, organismic integration theory, and basic needs theory

to some extent, by the context in which it is performed. A key tenet of self-determination theory is that the introduction of rewards in situations where people are already intrinsically motivated can shift the individual’s perception of the “cause” of their behavior from their inherent interest in the task to the external reward, undermining their intrinsic motivation (Deci, 1971). The shift in the perceived locus of causality has been studied under various types of reward contingencies, suggesting that perceptions of what is controlling one’s actions can affect motivation toward, and persistence with, tasks and behaviors (Deci, Koestner, & Ryan, 1999). Although early research focused on the effects of varied types of rewards, cognitive evaluation theory was extended based on research suggesting that the way communications, instructions, competence feedback, deadlines, and other

interpersonal events are presented similarly influences whether or not individuals’ intrinsic motivation will be undermined. Specifically, communications that convey external control or pressure tend to undermine autonomy, and thus diminish intrinsic motivation, whereas those that support autonomy and feelings of competence tend to enhance intrinsic motivation. For example, Ryan, Mims, and Koestner (1983) demonstrated that, when rewards were presented as “informational” on progress rather than contingent on behavioral performance, the undermining effect was not observed. External contingencies such as criticism or controlling praise also undermined intrinsic motivation, whereas fostering choice or providing competence-related feedback enhanced intrinsic motivation. Thus, consistent with cognitive evaluation theory, the key is whether such events shift the individual’s

Changing Behavior Using Self-Determination Theory

perceived locus of causality from internal to external or, oppositely (as in provision of choice), highlight personal causation, thereby enhancing autonomy. Research on cognitive evaluation theory paved the way for the elaborated predictions in self-determination theory and the development of other mini-theories.

8.2.3 Organismic Integration Theory Organismic integration theory extends the distinction between intrinsic motivation and extrinsic forms of motivation by broadening the perceived locus of causality and outlines the processes that determine the type of motivation individuals experience when performing particular tasks or behaviors. Within organismic integration theory, perceived locus of causality was conceptualized as a continuum along which varied types of extrinsic motivation, as well as intrinsic motivation, could be located. In other words, motivational types vary in their perceived locus of causality, with some being relatively autonomous and others relatively controlled (see the top line of the continuum in Figure 8.1). At the poles of the continuum are intrinsic motivation and external regulation, two types of behavioral regulation reflecting the prototypical forms of autonomous and controlled motivation. Under intrinsic motivation, individuals view their behavior as highly volitional or autonomous, whereas in external regulation persons see their behavior as driven by externally administered rewards or punishments. Identified regulation is an autonomous form of regulation located alongside intrinsic motivation on the continuum and represents engaging in tasks or behaviors because of their perceived value or importance. In contrast, introjected regulation is located adjacent to external regulation on the continuum and reflects performing tasks and behaviors to maintain selfesteem and feel self- or other-approval. An additional form of regulation, amotivation, has also been proposed. Amotivation reflects an absence

107

of either internal or external motives or reasons for acting and, strictly speaking, falls outside the continuum. Individuals experiencing tasks or behaviors as amotivated express listlessness, disinterest, and boredom. A person’s motivation toward any given behavior can be measured experimentally or through a family of validated questionnaires, many domain- or task-specific, first proposed by Ryan and Connell (1989, see also the scale in Figure 8.1). Although it is feasible that individuals could endorse more than one regulation type for a given behavior, correlations among the constructs suggest a characteristic pattern of correlations consistent with a continuum (Chatzisarantis et al., 2003; Howard, Gagné, & Bureau, 2017). Organismic integration theory also proposes the processes of internalization and integration, which explain how behaviors that are not inherently engaging or interesting can come to be personally valued and maintained. Because internalization describes the process by which individuals shift their perceived locus of causality for tasks and behaviors from an external locus to an internal one, it is particularly relevant in behavior change contexts. For example, behaviors perceived as being performed for controlled reasons can be “taken in” or assimilated, leading behavior to be performed for more autonomous reasons. Integration is a complete form of internalization such that the behavior is performed for reasons that are fully congruent and self-endorsed. Internalization can be influenced by the interpersonal context or by social agents operating in the interpersonal sphere, for example teachers in classrooms, health care staff in health consultations, leaders in organizations, or coaches in athletes’ training. Deci and colleagues (1994), for instance, identified three means to promote internalization: provision of choice, providing a rationale, and acknowledging conflict. These means promote internalization by highlighting personal origin, making personally endorsed reasons for performing the behavior salient, and demonstrating social support

108

MARTIN S. HAGGER, NELLI HANKONEN, NIKOS L. D. CHATZISARANTIS, AND RICHARD M. RYAN

by acknowledging possible challenges. These means form initial guidance on how social agents, over time, can promote internalization, and eventual integration, of behaviors,1 although there are numerous other techniques discussed within self-determination theory (Ryan & Deci, 2017). Organismic integration theory conceptualizes internalization as a “process model,” in which support for the basic needs of autonomy, competence, and relatedness by social agents fosters greater relative autonomy and thus improved behavioral persistence. Tests of the process model have demonstrated that provision of autonomy support by social agents (e.g., parents, teachers, physicians) promotes autonomous motivation and subsequent behavioral persistence, providing initial guidelines on how to change behavior based on the theory (Vasquez et al., 2015; Williams et al., 1998). A test of the self-determination theory process model is illustrated in Sidebar 8.1.

8.2.4 Basic Needs Theory Basic needs theory (Deci & Ryan, 2000) proposes that all humans have three basic or fundamental psychological needs, the fulfillment of which supports optimal functioning and wellness: needs for autonomy, competence, and relatedness. The need for autonomy reflects the need to feel that one is the “origin” of one’s actions, in the sense that the person engages in them willingly and feels a sense of ownership and choice in acting. The need for competence reflects the need to feel effectance, control, and mastery over tasks and behaviors. The need for relatedness reflects the need for noncontingent, unconditional support and connectedness with others. 1

To reflect this process further, the perceived locus of causality continuum has been augmented to include integrated regulation, which reflects full internalization of tasks or behaviors that were previously controlled motivated into those that are fully integrated and experienced as autonomous (see Figure 8.1). Measures of integrated regulation, however, have not always achieved discriminant validity with other forms of regulation (Howard et al., 2017).

Satisfaction of these basic psychological needs, in turn, predicts optimal psychological functioning, well-being, life satisfaction, and positive affect. When needs are thwarted or frustrated, individuals experience ill-being, dissatisfaction, and negative affect, among other signs of nonoptimal functioning. A principle of complementarity means that satisfaction of all three needs is important for optimal functioning. Persistent engagement in behaviors that are autonomously motivated is a pathway to need satisfaction. For example, behaviors that are intrinsically motivating (e.g., hobbies or pastimes), or fully internalized (e.g., pursuing tasks or behaviors that are deeply valued), are likely to fulfill all three psychological needs. For example, Weinstein and Ryan (2010) showed how volitionally engaging in helping behaviors is associated with feelings of autonomy, competence, and relatedness and explains why prosocial behaviors are so frequently associated with enhanced personal well-being. Individuals are likely to actively seek out and show greater persistence on behaviors that are need satisfying. Conversely, when individuals feel that behaviors are pressured or dictated by others (undermining autonomy), too difficult to master (undermining competence), or occur in contexts that are interpersonally unsupportive (undermining relatedness), needs are likely to be frustrated and behavior will less likely be maintained. Classic examples of needthwarting events are the imposition of unreasonable deadlines for tasks, micromanaging by leaders, or criticizing a person’s competency.

8.2.5 Putting It All Together Psychological need satisfaction is a unifying concept in self-determination theory and a principal mechanism that determines the type of motivation individuals experience when performing tasks or behaviors (Deci & Ryan, 2000). Need satisfaction is also the key concept that determines how interventionists, social agents, and other leaders in the interpersonal sphere can foster positive change in

Changing Behavior Using Self-Determination Theory

109

Sidebar 8.1 Williams et al.’s (1998) test of a process model of autonomy support, autonomous motivation, and medication adherence

Williams et al. (1998) conducted a longitudinal test of a self-determination theory– based process model in predicting patients’ medication adherence. Consistent with organismic integration theory and the internalization and integration processes, they proposed that patients whose physicians supported their autonomy when communicating treatment protocols would be more likely to express autonomous motivation for their treatment and report better adherence. The process model is illustrated in Figure 8.2(a). (a) Perceived Autonomy Support

a

Autonomous Motivation

b

Medication Adherence

c (b)

0.34

Perceived Autonomy Support

Autonomy Support Intervention

Intention

0.51

Physical activity

0.40

0.28 Autonomous Motivation

Figure 8.2 Processes in self-determination theory (a) self-determination theory process model; (b) path model showing effects of a teacher-delivered autonomy-support intervention on school children’s leisure-time physical activity mediated by autonomous motivation and intention

Note. Figures shown are standardized path coefficients. Patients required to take a prescription medication for at least one month attended an interview with a clinical psychologist to discuss their health, medical regimen, relationship with their physician, and adherence. They completed self-report measures of perceived autonomy support from their physician and autonomous motivation for taking their medication. They also completed a “pill count” of their medication. Participants were contacted two weeks later and asked to conduct a follow-up pill count to measure adherence. Perceived autonomy support had a small-to-medium–sized direct effect on autonomous motivation (path a, Figure 8.2(a)), and autonomous motivation had a direct large effect on medication adherence (path b, Figure 8.2(a)). Most important, the effect of perceived autonomy support on medication adherence was fully mediated by autonomous motivation (path c, Figure 8.2(a)). This research provided an illustration that perceived support for autonomy fosters autonomous motivation and behavioral persistence.

110

MARTIN S. HAGGER, NELLI HANKONEN, NIKOS L. D. CHATZISARANTIS, AND RICHARD M. RYAN

motivation and behavior. Individuals that perceive their behavior as need satisfying are likely to experience their actions as autonomous and are therefore likely to continue to engage in such behaviors as a means to satisfy their psychological needs, leading to behavioral persistence. Individuals that perceive their behavior as thwarting or frustrating of psychological needs are likely to experience their actions as controlled, often leading to desistence and avoidance, particularly in the absence of any continuous external contingencies or controls. If individuals persist with such behaviors, they will likely experience maladaptive outcomes such as ill-being and negative affect. Social agents, interventionists, and other leaders may promote behavior change and maintenance by promoting autonomous motivation through autonomy support. By using strategies that support autonomy and that emphasize other potentially need-satisfying components of the behavior such as personal value, mastery, and connections with others, social agents can enhance individuals’ willingness, performance, and persistence. Over time, individuals that consistently experience a behavior as need satisfying will likely internalize and integrate it into their repertoire of need-satisfying behaviors and thus experience more autonomous motivation. Generalized causality orientations, from causality orientations theory, can have a moderating effect on the extent to which communications from social agents promote internalization (Hagger & Chatzisarantis, 2011). The minitheories of self-determination theory, therefore, provide complementary explanations of the origins of motivated behavior and outline the processes and contingencies that determine motivation, action, and persistence over time.

8.3 How Has Self-Determination Theory Been Used to Change Behavior? Two key premises from self-determination theory, derived from its constituent mini-theories,

provide the central pillars of strategies for behavior change interventions: autonomous motivation and basic psychological need support. According to the theory, autonomous motivation and need satisfaction can be fostered using autonomy- and need-supportive strategies, typically communicated to the target population by social agents, as well as by the strategic structuring of feedback and contingencies. Autonomy- and need-supportive strategies highlight autonomous reasons for participating in the behavior of interest or value, provide choices where possible, make the need-satisfying features of the behavior more salient to the individual (e.g., by indicating choice or providing a personally relevant rationale), and provide opportunities for individuals to experience tasks and behaviors as consistent with their needs, values, and motives. A growing number of studies has tested the efficacy of interventions and programs adopting autonomy- and need-supportive strategies in changing behavior (see Ng et al., 2012). While interventions are typically delivered by social agents in face-toface situations, they have also been communicated by written messages such as online and smartphone-based interventions. In the next section the form and content of self-determination theory-based interventions are introduced, and research examining the efficacy of self-determination theory-based interventions in changing behavior summarized.

8.3.1 Need Support Autonomy-supportive strategies include not only provision of choice but also a meaningful rationale that can provide a basis for volitional engagement. In addition, supportive interventions scaffold tasks so that people can feel a sense of growth and mastery (supporting need for competence). Finally, interventions that are interpersonally supportive (supporting the need for relatedness) lead to more willingness to connect and internalize behavior change. Thus, acknowledging the person’s

Changing Behavior Using Self-Determination Theory

perspective and experiences can enhance behavioral maintenance. Research focused on promoting internalization has indicated that a combination of several strategies is most effective in moving individuals to being more autonomously motivated by highlighting to the individual that their actions are freely chosen, that they have a personal reason or rationale for doing the behavior, and that the social agent recognizes the challenge presented by the behavior and therefore communicates a sense of understanding and regard for the individual in performing the behavior (Ryan & Deci, 2017). These formulations have been supported by a substantial literature. For example, experimental studies have confirmed the role of choice as a fundamental component of autonomy-support interventions (Patall, Cooper, & Robinson, 2008). In addition, research in the field of education has identified the autonomy-supportive strategies that teachers adopt to promote autonomous motivation and behavior change among students (Deci et al., 1982; Reeve, Bolt, & Cai, 1999; Reeve & Jang, 2006). These findings are based on formative research demonstrating consistent links between autonomous forms of motivation and school students’ interest and engagement in class and academic attainment (Reeve, 2002). The research has focused on the kinds of behaviors teachers display in class (“what teachers do”) and the language content and style when they instruct students (“what teachers say”) that communicate and foster autonomous motivation among students (Reeve & Jang, 2006; Reeve et al., 2004). Research has also identified the controlling behaviors and language that undermine psychological needs and lead to maladaptive outcomes (Deci et al., 1982). Minimizing the use of controlling behaviors is also important to promote autonomous motivation (Tilga et al., 2019). Reeve and colleagues (2002, 2006) conducted an influential set of studies that identified the autonomy-supportive and controlling behaviors

111

that teachers typically adopt in classroom contexts (see also Chapter 35, this volume). The research is a preeminent example of structured means to identify, code, and assess these kinds of behaviors in teachers, which may inform the content of behavior change interventions based on self-determination theory. They developed a list of twenty-one autonomy-supportive and controlling teacher behaviors, based on previous research that had observed and manipulated teachers’ behaviors from a self-determination theory perspective (Deci et al., 1982; Reeve et al., 1999). They coded the instructional behaviors used by a sample of teachers during a ten-minute teaching interaction and measured the perceived autonomy, engagement, and performance of their students. Autonomy-supportive (e.g., time listening; time student talking; communicating perspectivetaking statements; time allowing student to work in own way; providing informational feedback; offering encouragement; offering hints; being responsive to students’ questions) and controlling (e.g., exhibiting solutions/answers; uttering solutions/answers; time holding/monopolizing learning materials; uttering directives/commands; making should/got to statements; asking controlling questions) behaviors were consistently associated with autonomous motivation and learning ratings for the students. This research identified the behaviors autonomy-supportive teachers would expect to display in order to promote students’ autonomous motivation and adaptive outcomes in class. It also has significant translational value by providing a template for the identification of behaviors that social agents in other contexts may use to support autonomy. The list of behaviors is presented in Appendix 8.1 (supplemental materials) and includes operational definitions of each. A leading approach to developing effective autonomy-support interventions in education contexts is autonomy-support training programs (Cheon & Reeve, 2013; Cheon, Reeve, & Moon, 2012; Reeve & Jang, 2006; see

112

MARTIN S. HAGGER, NELLI HANKONEN, NIKOS L. D. CHATZISARANTIS, AND RICHARD M. RYAN

Chapter 35, this volume). These programs recognize the importance of training teachers to display autonomy-supportive behaviors and use autonomy-supportive language when communicating with students. In essence, the approach recognizes the need to change the behavior of those delivering the intervention in order to change the behavior of the target populations (see also Chapter 21, this volume). Promoting autonomy support in educational settings requires teachers to adopt and consistently apply sets of autonomy-supportive behaviors in their everyday lessons. The training programs provide instructional materials and teaching plans to develop teachers’ competencies in the use of these behaviors. The programs include descriptions of the autonomy-supportive behaviors, teaching points and examples of the use of the behaviors, tips on how to develop the behaviors, and relevant practices that teachers can perform to build competency in their use. Meta-analyses have demonstrated that autonomy-support training programs are generally effective in changing teachers’ behavior (Su & Reeve, 2011). Interventions that promote competence and relatedness need support have also been explored. Research has demonstrated that the provision of mastery experiences promotes autonomous motivation by enhancing competence (e.g., Jang, Reeve, & Deci, 2010; Tessier, Sarrazin, & Ntoumanis, 2010). Interventions aimed at promoting competence may play a role in enhancing autonomous motivation. Similarly, research has also identified behaviors that promote relatedness need support and their effects on autonomous motivation and behavioral engagement (e.g., Sparks et al., 2017). However, providing support for competence and relatedness needs in the absence of autonomy support may not be sufficient to promote internalization of behaviors and autonomous motivation. For example, it is possible for individuals to feel competent in performing tasks but to not feel they have complete ownership over their actions because they view

their behavior as controlled by external forces. Individuals can, therefore, feel competent but not autonomous. Furthermore, there may be interplay between need-supportive behaviors. For example, support for autonomy has also been shown to foster competence (Williams, Lynch, & Glasgow, 2007). More research is needed to systematically examine the extent to which different types of need support also support autonomous motivation and internalization.

8.3.2 Motivation and Behavior Change Techniques Following research identifying need-supportive behaviors in education (Reeve, 2002; Reeve et al., 1999; Reeve & Jang, 2006), Teixeira and colleagues (2020) conducted a study to develop a comprehensive description of the strategies or techniques used to promote autonomous motivation and behavior change based on self-determination theory in health contexts. Psychological need support was used as a central organizing principle. The researchers identified a list of candidate need-supportive techniques from a comprehensive literature review. They then matched each technique with its “primary” psychological need and produced labels, definitions, and functional descriptions for each. Next, the list of candidate techniques was circulated to a group of self-determination theory experts, who rated each technique according to its uniqueness, redundancy, essentiality, and match with its primary psychological need. Using a consensus approach from the expert ratings, a classification of twenty-one techniques was developed. The techniques identified from the consensus study are presented in Appendix 8.2 (supplemental materials), classified according to its primary targeted need. Although some of the techniques are common to those identified in previous research (e.g., Reeve & Jang, 2006), the classification is the first attempt to comprehensively isolate the unique techniques that comprise self-determination

Changing Behavior Using Self-Determination Theory

theory interventions. The classification will function as a resource for researchers and practitioners to inform the content of behavior change interventions; provide guidance on how those techniques might “work” in promoting need satisfaction, autonomous motivation, and behavior change; and deliver a common set of terms and descriptions to help researchers accurately report the content of self-determination theory interventions. Although the classification has been developed through consensus, future research is needed to establish its validity by using it as a basis for rigorous tests of selfdetermination theory–based interventions.

8.3.3 Evidence for Self-Determination Theory Interventions There is a considerable body of research on the efficacy of interventions based on selfdetermination theory in promoting motivation and changing behavior across multiple disciplines, populations, and behaviors (e.g., Chatzisarantis & Hagger, 2009; Cheon & Reeve, 2013; Silva et al., 2010). Interventions typically adopt combinations of the autonomy-supportive strategies or seek to minimize or eliminate controlling strategies displayed by social agents to “create” an autonomysupportive environment to promote the behavior of interest to the target population. Some interventions have delivered autonomy-supportive interventions via print communication or media such as websites or mobile phone apps (e.g., Spring et al., 2013). Studies evaluating self-determination theory–based interventions have tended to use controlled designs, with some using fully randomized controlled, experimental, or longitudinal pre- and post-intervention designs (e.g., Chatzisarantis & Hagger, 2009; Cheon et al., 2012; Hankonen et al., 2016; Shah et al., 2016; Silva et al., 2010; Williams et al., 2006). Studies usually use trained facilitators as the “social agents” that deliver the intervention content or train existing social agents to use the techniques, using an autonomy-support training program or

113

similar. For example, Williams et al. (2006) trained counselors to support smokers in making autonomous decisions to quit, and Cheon et al. (2012) trained teachers to display autonomy-supportive behaviors using their autonomy-support training program. Interventions are usually evaluated using relevant outcome measures of the behavior of interest at single or multiple follow-up occasions after the initiation of the intervention (e.g., Williams et al., 2016). Importantly, researchers have also included measures of key theoretical constructs expected to change as a consequence of the intervention. These measures include autonomous and controlled forms of motivation toward the behavior of interest, psychological need satisfaction, interest, enjoyment, competence, life satisfaction, and vitality. Measures such as autonomous motivation and need satisfaction are implicated in the process by which the intervention changes behavior consistent with the theory. These are considered mediators of the effects of the interventions, such that they explain “how” the content of the intervention results in behaviors change. Measures such as interest, enjoyment, and vitality are expected outcomes of engaging in a behavior for autonomous reasons and when needs are satisfied and are therefore considered secondary outcomes. Consistent with the designs of studies evaluating self-determination theory interventions, measures of behavioral and motivational outcomes are typically measured at multiple time points in conjunction with behavioral measures to establish the effects of intervention in changing outcomes over time. Such evaluations involve testing the extent to which measures of the self-determination theory constructs, such as perceived satisfaction of psychological needs and autonomous motivation, mediate the effect of the intervention on the behavioral outcome measured at post-intervention follow-up (e.g., Chatzisarantis & Hagger, 2009; Silva et al., 2010; Williams et al., 2006). An illustrative example is presented in Sidebar 8.2.

114

MARTIN S. HAGGER, NELLI HANKONEN, NIKOS L. D. CHATZISARANTIS, AND RICHARD M. RYAN

Sidebar 8.2 Chatzisarantis and Hagger’s (2009) process evaluation of an autonomy-support intervention

Chatzisarantis and Hagger developed an intervention to promote physical activity among high school students outside of school in their “leisure time.” Their intervention adopted an autonomy-support training program (see Cheon & Reeve, 2013; Cheon et al., 2012; Reeve et al., 2004) to promote teachers’ use of autonomysupportive behaviors in their lessons to promote autonomous motivation toward the target behavior in their students. Teachers assigned to an autonomy-support intervention condition received training on autonomy-supportive behaviors (e.g., providing positive feedback, giving a rationale, acknowledging difficulties, enhancing choice). Teachers assigned to a control condition were provided training that did not include the autonomy-supportive components. After training, teachers implemented their training in their regular lessons for five weeks. Students’ autonomous motivation, perceived autonomy support, leisure-time physical activity intentions, and leisure-time physical activity participation were measured pre- and post-intervention. Path analysis tested the effects of the intervention on follow-up measures and leisure-time physical activity participation. Results revealed small-tomedium–size direct effects of the intervention on students’ autonomous motivation and perceived autonomy support, a medium-size direct effect of autonomous motivation on intentions, a medium-size direct effect of intentions on physical activity participation, and a small-size indirect effect of the intervention through autonomous motivation and intentions. Findings are summarized in Figure 8.2(b). These illustrate the efficacy of the intervention in changing behavior, as well as the processes by which the intervention is presumed to affect behavior change. These effects are a specific example of the more generalized process model of how autonomy- and need-supportive interventions impact behavior change and associated outcomes through changes in need satisfaction and autonomous motivation (see Fortier et al., 2012 and the figure presented in Appendix 8.3, supplemental materials).

A growing body of research demonstrates the efficacy of autonomy- and need-support interventions in promoting motivation and behavior change and adaptive outcomes in health (e.g., Gillison et al., 2018; Ng et al., 2012; Teixeira, Palmeira, & Vansteenkiste, 2012), occupational (e.g., Deci, Connell, & Ryan, 1989), and educational (e.g., Cheon et al., 2012; Reeve et al., 2004) contexts. When evaluating the efficacy of these interventions, it is important to consider that most interventions aim to promote changes in autonomous motivation and behavior of the target

population (e.g., students, employees), by promoting autonomy- and need-supportive behaviors in appropriate social agents (e.g., teachers, managers). In many contexts (e.g., schools, workplaces), this means changing the behavior of the social agents themselves. Some interventions, therefore, have the end goal of changing the behavior of the teachers themselves (Cheon & Reeve, 2013; Su & Reeve, 2011). For example, a meta-analysis in educational contexts has shown autonomy-supportive interventions to be effective in producing change in teachers’ use of

Changing Behavior Using Self-Determination Theory

autonomy-supportive training styles, with longerterm interventions more effective (Su & Reeve, 2011). However, most aim to achieve change in the target population or both. It is also important to account for the goal of the intervention when assessing its efficacy, often illustrated by change in the dependent variables. A key question, therefore, is whether the intervention focuses on changing the autonomy-supportive behaviors of the social agents alone or both the social agents and the target population. Systematic reviews and meta-analyses have illustrated consistent effects of autonomy- and need-support interventions on individuals’ behavior in particular domains. For example, Teixeira and colleagues (2012) demonstrated that the majority of studies adopting autonomysupportive interventions found effects on physical activity participation and theoretical constructs, including perceived autonomy support, need satisfaction, and autonomous motivation. Similarly, Ng et al.’s (2012) meta-analysis revealed small-to-medium–size overall averaged effects of autonomous motivation on health behavior participation among studies adopting experimental and intervention designs, and these effects were stronger than effects in studies adopting nonexperimental designs. Overall, the evidence for autonomy-support interventions seems to be consistent, although, to date, there has been no quantitative research synthesis of autonomy- and need-support interventions on motivational and behavioral outcomes across multiple behavioral domains. The majority of autonomy-support interventions have evaluated change over a relatively brief period, with follow-up measures of behavior change being only a few weeks post-intervention (Cheon & Reeve, 2013; Ng et al., 2012). However, there is some evidence of long-term intervention effects over a year or more post-intervention (e.g., Cheon & Reeve, 2013; Silva et al., 2010). For example, Cheon and Reeve (2013) demonstrated substantive effects of their autonomy-supportive

115

intervention on children’s autonomous motivation, academic attainment, and adaptive outcomes in school physical education over a one-year period. Similarly, Silva et al. (2010) demonstrated maintenance of the effects of a thirty-session facilitatorled autonomy-support intervention on physical activity participation, weight-loss outcomes, and self-determination theory one and two years postintervention. Although the research evidence is relatively sparse and confined to the health domain, autonomy-support interventions demonstrate considerable promise in fostering long-term motivation and behavior change. Autonomy-support interventions vary in their duration and intensiveness, or “dose,” usually determined by intervention duration and the “contact time” that participants spend with the social agents delivering the intervention. While some brief interventions have been effective (e.g., Chatzisarantis & Hagger, 2009; Tessier et al., 2010), there is evidence that more intensive interventions involving long-term and frequent exposures to autonomy-supportive strategies tend to be more effective. Teixeira et al. (2012) noted considerable variability in the duration of self-determination theory–based interventions used to promote physical activity and noted that most were less than three months in duration and involved only a brief amount of contact time. They also noted considerable variability in the numbers of strategies used and the extent to which the interventions were based on the theory. However, they did not note whether this variability in content coincided with variability in efficacy. To date, there is no study that has systematically varied the duration and dose of self-determination theory interventions and assessed their effects on motivation and behavior change, and this remains an important avenue for future research. Many self-determination theory–based interventions comprise multiple techniques in a single intervention and test intervention effects on behavior change relative to a no-intervention control

116

MARTIN S. HAGGER, NELLI HANKONEN, NIKOS L. D. CHATZISARANTIS, AND RICHARD M. RYAN

group. In contrast, few studies have tested effects of individual autonomy-support intervention techniques on behavior change using factorial designs. The development of the classification of motivational and behavior change techniques provides a useful starting point for testing the main, additive, and interactive effects of individual techniques on motivation and behavior change (Teixeira et al., 2020). Research that systematically evaluates the effects of different techniques on motivation mediators (e.g., autonomous motivation, need satisfaction) and behavior change will progress knowledge by identifying the most effective techniques, as well as those that are less effective or redundant. It will also assist in identifying combinations of intervention techniques that lead to greater behavior change than each of the component techniques alone.

8.4 Conclusion and Future Directions Self-determination theory is a generalized theory of human motivation and wellness, with the quality of motivation, psychological need satisfaction, and environmental supports for motivation as key components. The theory proposes a fundamental distinction between autonomous motivation, characterized by reasons for acting that are selfendorsed and experienced as volitional, and controlled motivation, defined as acting out of externally-referenced reasons and experienced as determined by events or pressures outside the self. Autonomous forms of motivation have been consistently linked to better persistence with behaviors and more adaptive outcomes. Individuals’ motivational quality is influenced by the extent to which basic psychological needs for autonomy, competence, and relatedness are supported and satisfied. Behavior change interventions based on selfdetermination theory have focused on providing support for psychological needs to promote autonomous motivation and behavioral persistence. Most interventions focus on training social

agents (e.g., teachers, managers) to provide support for the autonomy of the target population. Research has demonstrated the efficacy of interventions based on the theory in changing behavior and theory-based constructs, including autonomous motivation, perceived autonomy support, and adaptive outcomes. Process evaluations of interventions have also demonstrated the role of autonomous motivation in mediating intervention effects on behavior change. Recent research has developed a classification of the techniques that comprise need-supportive interventions (Teixeira et al., 2020). Future research may seek to examine the unique and interactive effects of specific techniques on behavior change and provide further evidence of longterm effectiveness of interventions based on the theory. Extending this idea, a system to characterize motivating and demotivating style along the dimensions of provision of need support and level of directiveness has recently been developed (Aelterman et al., 2019). Four “styles” were identified, each defined by a quadrant bounded by high and low levels of the need-supportive and directiveness dimensions: autonomy-supportive, structuring, controlling, and chaotic. Future research should seek to establish the motivational and behavioral consequences of social agents adopting each style. Finally, research has also examined whether individuals can apply autonomy-supportive strategies to motivate their own behavior. Such self-enactable strategies may include modifying the behavior or task to be more enjoyable, reminding oneself of personally important reasons to engage in the behavior, aligning one’s identity with the target behavior, modifying ways of doing the behavior to allow for participation in other valued behaviors or pursuit of valued outcomes, reflecting on autonomous goals, and reminding oneself of past success in the behavior (Knittle et al., 2020). Research is needed to explore the effectiveness of these strategies in enabling individuals to change their own behavior.

Changing Behavior Using Self-Determination Theory

References Aelterman, N., Vansteenkiste, M., Haerens, L., Soenens, B., Fontaine, J. R. J., & Reeve, J. (2019). Toward an integrative and fine-grained insight in motivating and demotivating teaching styles: The merits of a circumplex approach. Journal of Educational Psychology, 111, 497–521. https:// doi.org/10.1037/edu0000293 Chatzisarantis, N. L. D., & Hagger, M. S. (2009). Effects of an intervention based on selfdetermination theory on self-reported leisure-time physical activity participation. Psychology and Health, 24, 29–48. https://doi.org/10.1080/ 08870440701809533 Chatzisarantis, N. L. D., Hagger, M. S., Biddle, S. J. H., Smith, B., & Wang, C. K. J. (2003). A meta-analysis of perceived locus of causality in exercise, sport, and physical education contexts. Journal of Sport and Exercise Psychology, 25, 284–306. https://doi.org/ 10.1123/jsep.25.3.284 Cheon, S. H., & Reeve, J. (2013). Do the benefits from autonomy-supportive PE teacher training programs endure? A one-year follow-up investigation. Psychology of Sport and Exercise, 14, 508–518. https://doi.org/10.1016/ j.psychsport.2013.02.002 Cheon, S. H., Reeve, J., & Moon, I. (2012). Experimentally based, longitudinally designed, teacher-focused intervention to help physical education teachers be more autonomy supportive toward their students. Journal of Sport and Exercise Psychology, 34, 365–396. https://doi.org/ 10.1123/jsep.34.3.365 Deci, E. L. (1971). Effects of externally mediated rewards on intrinsic motivation. Journal of Personality and Social Psychology, 18, 105–115. Deci, E. L., Connell, J. P., & Ryan, R. M. (1989). Selfdetermination in a work organization. Journal of Applied Psychology, 74, 580–590. https://doi.org/ 10.1037/0021-TS (global): Change en-dashes in URLs for DOIs to hyphens throughout to ensure links work correctly.9010.74.4.580 Deci, E. L., Eghrari, H., Patrick, B. C., & Leone, D. R. (1994). Facilitating internalization: The selfdetermination theory perspective. Journal of

117

Personality, 62, 119–142. https://doi.org/10.1111/ j.1467-6494.1994.tb00797.x Deci, E. L., Koestner, R., & Ryan, R. M. (1999). A meta-analytic review of experiments examining the effects of extrinsic rewards on intrinsic motivation. Psychological Bulletin, 125, 627– 668. Deci, E. L., & Ryan, R. M. (1985a). The general causality orientations scale: Self-determination in personality. Journal of Research in Personality, 19, 109–134. https://doi.org/10.1016/0092-6566 (85)90023-6 Deci, E. L., & Ryan, R. M. (1985b). Intrinsic Motivation and Self-Determination in Human Behavior. New York: Plenum Press. Deci, E. L., & Ryan, R. M. (2000). The “what” and “why” of goal pursuits: Human needs and the selfdetermination of behavior. Psychological Inquiry, 11, 227–268. https://doi.org/10.1207/ S15327965PLI1104_01 Deci, E. L., Spiegel, N. H., Ryan, R. M., Koestner, R., & Kauffman, M. (1982). Effects of performance standards on teaching styles: Behavior of controlling teachers. Journal of Educational Psychology, 74, 852–859. https://doi.org/10.1037/ 0022-0663.74.6.852 Fortier, M. S., Duda, J. L., Guerin, E., & Teixeira, P. J. (2012). Promoting physical activity: Development and testing of self-determination theory-based interventions. International Journal of Behavioral Nutrition and Physical Activity, 9, 20. https://doi .org/10.1186/1479-5868-9-20 Gillison, F. B., Rouse, P., Standage, M., Sebire, S., & Ryan, R. M. (2018). A meta-analysis of techniques to promote motivation for health behaviour change from a self-determination theory perspective. Health Psychology Review, 13, 110–130. https://doi.org/10.1080/ 17437199.2018.1534071 Hagger, M. S., & Chatzisarantis, N. L. D. (2011). Causality orientations moderate the undermining effect of rewards on intrinsic motivation. Journal of Experimental Social Psychology, 47, 485–489. https://doi.org/10.1016/j.jesp.2010.10.010 Hankonen, N., Heino, M. T. J., Araujo-Soares, V. et al. (2016). “Let’s Move It”: A school-based multilevel intervention to increase physical

118

MARTIN S. HAGGER, NELLI HANKONEN, NIKOS L. D. CHATZISARANTIS, AND RICHARD M. RYAN

activity and reduce sedentary behaviour among older adolescents in vocational secondary schools: A study protocol for a cluster-randomised trial. BMC Public Health, 16, 451. https://doi.org/ 10.1186/s12889-016-3094-x Howard, J. L., Gagné, M., & Bureau, J. S. (2017). Testing a continuum structure of self-determined motivation: A meta-analysis. Psychological Bulletin, 143, 1346–1377. https://doi.org/ 10.1037/bul0000125 Jang, H., Reeve, J., & Deci, E. L. (2010). Engaging students in learning activities: It is not autonomy support or structure but autonomy support and structure. Journal of Educational Psychology, 102, 588–600. https://doi.org/10.1037/a0019682 Knittle, K., Heino, M. T. J., Marques, M. M., Stenius, M., Beattie, M., Ehbrecht, F., Hagger, M. S., Hardeman, W., & Hankonen, N. (2020). The compendium of self-enactable techniques to change and self-manage motivation and behaviour v1. 0. Nature Human Behavior, 4, 215–223. https://doi.org/10.1038/s41562-019-0798-9. Ng, J. Y. Y., Ntoumanis, N., Thøgersen-Ntoumani, C. et al. (2012). Self-determination theory applied to health contexts. Perspectives on Psychological Science, 7, 325–340. https://doi.org/10.1177/ 1745691612447309 Patall, E. A., Cooper, H., & Robinson, J. C. (2008). The effects of choice on intrinsic motivation and related outcomes: A meta-analysis of research findings. Psychological Bulletin, 134, 270–300. https://doi.org/10.1037/0033-2909.134.2.270 Reeve, J. (2002). Self-determination theory applied to educational settings. In E. L. Deci & R. M. Ryan (Eds.), Handbook of Self-Determination Research (pp. 183–203). Rochester, NY: University of Rochester Press. Reeve, J., Bolt, E., & Cai, Y. (1999). Autonomysupportive teachers: How they teach and motivate students. Journal of Educational Psychology, 91, 537–548. https://doi.org/10.1037/00220663.91.3.537 Reeve, J., & Jang, H. (2006). What teachers say and do to support students’ autonomy during a learning activity. Journal of Educational Psychology, 98, 209–218. https://doi.org/10.1037/00220663.98.1.209

Reeve, J., Jang, H., Carrell, D., Jeon, S. Y., & Barch, J. (2004). Enhancing students’ engagement by increasing teachers’ autonomy support. Motivation and Emotion, 28, 147–169. https://doi .org/10.1023/B:MOEM.0000032312.95499.6 f Reeve, J., Jang, H., Hardre, P., & Omura, M. (2002). Providing a rationale in an autonomy-supportive way as a strategy to motivate others during an uninteresting activity. Motivation and Emotion, 26, 183–207. https://doi.org/10.1023/ a:1021711629417 Ryan, R. M., & Connell, J. P. (1989). Perceived locus of causality and internalization: Examining reasons for acting in two domains. Journal of Personality and Social Psychology, 57, 749–761. https://doi .org/10.1037/0022-3514.57.5.749 Ryan, R. M., & Deci, E. L. (2017). Self-Determination Theory: Basic Psychological Needs in Motivation, Development and Wellness. New York: Guilford Press. Ryan, R. M., Mims, V., & Koestner, R. (1983). Relation of reward contingency and interpersonal context to extrinsic motivation: A review and test using cognitive evaluation theory. Journal of Personality and Social Psychology, 45, 736–750. Shah, K. N., Majeed, Z., Yoruk, Y. B. et al. (2016). Enhancing physical function in HIV-infected older adults: A randomized controlled clinical trial. Health Psychology, 35, 563–573. https://doi .org/10.1037/hea0000311 Silva, M. N., Vieira, P. N., Coutinho, S. R. et al. (2010). Using self-determination theory to promote physical activity and weight control: A randomized controlled trial in women. Journal of Behavioral Medicine, 33, 110–122. https://doi .org/10.1007/s10865-009-9239-y Sparks, C., Lonsdale, C., Dimmock, J. A., & Jackson, B. (2017). An intervention to improve teachers’ interpersonally involving instructional practices in high school physical Education: Implications for student relatedness support and in-class experiences. Journal of Sport and Exercise Psychology, 39, 120–133. https://doi.org/10.1123/ jsep.2016-0198 Spring, B., Gotsis, M., Paiva, A., & Spruijt-Metz, D. (2013). Healthy apps: Mobile devices for continuous monitoring and intervention. IEEE

Changing Behavior Using Self-Determination Theory

Pulse, 4, 34–40. https://doi.org/10.1109/ MPUL.2013.2279620 Su, Y. L., & Reeve, J. (2011). A meta-analysis of the effectiveness of intervention programs designed to support autonomy. Educational Psychology Review, 23, 159–188. https://doi.org/10.1007/ s10648-010-9142-7 Teixeira, P. J., Palmeira, A., & Vansteenkiste, M. (2012). The role of self-determination theory and motivational interviewing in behavioral nutrition, physical activity, and health: An introduction to the IJBNPA special series. International Journal of Behavioral Nutrition and Physical Activity, 9, 17. https://doi.org/10.1186/1479-5868-9-17 Teixeira, P. J., Marques, M. M., Silva, M. N. et al. (2020). Classification of techniques used in selfdetermination theory-based interventions in health contexts: An expert consensus study. Motivation Science. https://doi.org/10.1037/mot0000172 Tessier, D., Sarrazin, P., & Ntoumanis, N. (2010). The effect of an intervention to improve newly qualified teachers’ interpersonal style, students motivation and psychological need satisfaction in sport-based physical education. Contemporary Educational Psychology, 35, 242–253. https://doi .org/10.1016/j.cedpsych.2010.05.005 Tilga, H., Hein, V., Koka, A., Hamilton, K., & Hagger, M. S. (2019). The role of teachers’ controlling behaviour in physical education on adolescents’ health-related quality of life: Test of a conditional process model. Educational Psychology, 39, 862– 880. https://doi.org/10.1080/ 01443410.2018.1546830 Vansteenkiste, M., Lens, W., & Deci, E. L. (2006). Intrinsic versus extrinsic goal contents in selfdetermination theory: Another look at the quality of academic motivation. Educational

119

Psychologist, 41, 19–31. https://doi.org/10.1207/ s15326985ep4101_4 Vasquez, A. C., Patall, E. A., Fong, C. J., Corrigan, A. S., & Pine, L. (2015). Parent autonomy support, academic achievement, and psychosocial functioning: A meta-analysis of research. Educational Psychology Review, 28, 605–644. https://doi.org/10.1007/s10648-015-9329-z Weinstein, N., & Ryan, R. M. (2010). When helping helps: Autonomous motivation for prosocial behavior and its influence on well-being for the helper and recipient. Journal of Personality and Social Psychology, 98, 222–244. https://doi.org/ 10.1037/a0016984 Williams, G. C., Deci, E. L., Niemiec, C. P., Ryan, R. M., & Patrick, H. (2016). Outcomes of the smoker’s health project: A pragmatic comparative effectiveness trial of tobacco-dependence interventions based on self-determination theory. Health Education Research, 31, 749–759. https:// doi.org/10.1093/her/cyw046 Williams, G. C., Lynch, M., & Glasgow, R. E. (2007). Computer-assisted intervention improves patientcentered diabetes care by increasing autonomy support. Health Psychology, 26, 728–734. https:// doi.org/10.1037/0278-6133.26.6.728 Williams, G. C., McGregor, H. A., Sharp, D. et al. (2006). Testing a self-determination theory intervention for motivating tobacco cessation: Supporting autonomy and competence in a clinical trial. Health Psychology, 25, 91–101. https://doi.org/10.1037/0278-6133.25.1.91 Williams, G. C., Rodin, G. C., Ryan, R. M., Grolnick, W. S., & Deci, E. L. (1998). Autonomous regulation and long-term medication adherence in adult outpatients. Health Psychology, 17, 269– 276.

9

Changing Behavior Using Control Theory Warren Mansell

Practical Summary Have you ever heard someone say that they do not feel in control of their life? People tend to feel stressed or distressed when they cannot control things they feel are important to them, like feelings, a sense of self, or adherence to personal values. Control theory states that control is fundamental to well-being and to the healthy functioning of any person, group of people, or organization. The theory also states that changing behavior allows people to counteract disturbances (obstacles, challenges) that can threaten their ability to stay in control of their lives. So behavior change is at the very heart of well-being, at every moment of the day. This chapter focuses on how practitioners can draw on control theory to assist in making sustained behavior change that is beneficial to their well-being, such as helping parents manage their infant’s development, helping recovery from mental health problems, organizing education and discipline within schools, and managing organizations.

9.1 Introduction The application of control theory in psychology has its basis in control engineering. It is also closely related to the field of cybernetics that emerged in the 1940s and 1950s. This chapter focuses on the form of control theory described by Powers (1973/2005, 2008). Powers’s theory has provided the basis of an important strand of self-regulation theory within social and clinical psychology (Carver & Scheier, 1982), as well as, more recently, informing psychotherapy, neuroscience, child development, education, sociology, and organizational psychology (for a historical view on Powers’s work, see Sidebar 9.1). The theory, also known as perceptual control theory, has also provided an integrative framework for theories of

behavior change to inform interventions for addiction (Webb, Sniehotta, & Michie, 2010). The theory proposes that behavior changes continuously in order to control one’s perception of aspects of the self and the environment. Control theory proposes that life requires “control” (Powers, 1973/ 2005), that well-being and mental health require control (Carey, 2006), and that good health also requires control (Carey, 2017). The purpose of this chapter is to review control theory and its evidence, to propose how it accounts for behavior change, and to describe examples of the interventions informed by the theory and evidence for their effectiveness. https://doi.org/10.1017/9781108677318.009

Changing Behavior Using Control Theory

121

Sidebar 9.1 A history lesson: What happened to Powers’s control theory?

William T. Powers (1926–2013) based his version of control theory on his knowledge of the artificial control systems he encountered as a medical physicist and engineer. He realized that even control systems machines control their inputs, not their outputs. Take the humble thermostat. It controls its perception of the current temperature, keeping it as close as possible to the value set by its human user. Control theory ideas had been applied to human behavior before, within the cybernetic movement (Ashby, 1952; Wiener, 1948), but Powers showed that these models made the “input blunder,” assuming that control occurred on the output side of the model (Powers, 1978). Powers had hoped that his first papers, his well-received first book (Powers, 1973/2005), and his landmark Psychological Review article (Powers, 1978) would change the course of psychological research. He hoped that scientists of behavior would begin to develop tests of what perceived aspects of the self or the environment a person was controlling. Yet, even now, the predominant research methodology involves studying the linear relationship between stimuli, cognitive processes, and responses (Mansell & Huddy, 2018).

9.2 Overview of Control Theory and Evidence The basic “unit” of control theory is the negative feedback loop (see Figure 9.1).1 The loop is a functional subsystem – or a working model – that can be used to simulate human (or animal) behavior and even to operate robotic systems (Vancouver, 2005; Young, 2017). These basic subsystems are organized in a branching hierarchy to model behavior in detail. In essence, people achieve long-term goals (e.g., to be healthy) by setting lower-level subgoals (e.g., to exercise), which in turn set subgoals at a lower level (e.g., to run), which, in turn, set subgoals (e.g., legs moving fast). These layers continue downward, at each layer specifying increasingly concrete, sensory inputs. Every level is being controlled simultaneously to flexibly achieve the long-term goals. The purpose of the negative feedback loop is to control some perceived aspect of the self or the world. Consider a concrete real-world example – walking on a tightrope. Some tightrope walkers report that they focus on an object ahead of them. The location of this object in their field of vision

would be the controlled variable. The walker has an internally specified goal, standard, or reference value for the location of this object – at the center of their vision. The current state of the variable is compared continuously with the reference value using a comparator. This generates a discrepancy, or error, which is amplified by a gain value, to drive a change in behavior that acts against disturbances in the environment, to keep the variable at, or as close as possible to, the reference value. Disturbances come from wind and from the movements of the walker’s own body and the rope. The actions of the walker are channeled through the environment to have their effects. The properties of the body and environment that allow this are called its feedback function. Because the loop is working continuously, the disturbances are continuously

1

The term “negative” feedback should not imply criticism or a negative emotional tone. It means that the actions of the system are subtracted from the disturbances in the environment in order to reduce error and meet a goal. In contrast, a “positive” feedback loop adds to the disturbance to increase error, which makes it inherently unstable and often regulated, in turn, by a negative feedback loop to restore control.

122

WARREN MANSELL

Figure 9.1 The feedback control unit Note. Adapted with permission from Forssell, D. (2008). Management and Leadership: Insight for Effective Practice. Hayward, CA: Living Control Systems Publications from an original diagram by William T. Powers.

counteracted as soon as they occur, maintaining stability. A clear implication of this model, so far, is that people control their input (perception) by varying their outputs (actions) in an ongoing manner. This occurs automatically and does not require conscious awareness – in fact, as described in Section 9.3.5, awareness is often focused only on a subset of the perceptions being controlled. Powers proposed that this unit of control is fundamental to behavior. It applies at more abstract

levels – such as trying to “make a good impression” – and even to behaviors that appear to be “triggered” as fixed, planned actions (see Figure 9.1). For example, the deliberate behavior of “being a good listener” requires that a person can perceive how much they are listening (the perceptual signal) and compare this to a reference value of how good at listening they want to be (reference value and comparator) to generate a discrepancy or error signal and engage in any actions (output quantities)

Changing Behavior Using Control Theory

within the environment (feedback function) necessary to act against any disturbances (e.g., to ask the person to speak louder when there is loud background noise) to keep the perceptual signal close to the reference value. Often when it appears that a “stimulus” has triggered a response, this happens because the disturbance is occurring in a sudden rather than smooth manner. For example, the classic “startle” response to a loud noise occurs because the startled person has a reference value for their desired loudness level of surrounding noise. The sudden onset of a loud noise disturbs this variable very quickly, creating large error and therefore a large response. However, as the onset is lengthened in time (e.g., when the volume increases steadily), the momentary error level is lower, giving more time for other varied actions to be utilized (e.g., saying “turn that noise down!”). A series of reviews have summarized the evidence for the control theory tenet that behavior is the control of perceptual input (Mansell & Huddy, 2018; Marken & Mansell, 2013). Around twenty studies have used evidence from manual tracking in which the user moves a computer mouse or joystick to control some aspect of the visual display despite disturbances to this variable that are provided by the computer. The studies show that control theory can be used to infer what variable the participant is controlling, such as the position of a moving cursor (Powers, 1978). In each of these studies, behavior is changing continuously, yet a perceptual variable is being kept at a desired value. Furthermore, when comparative tests have been carried out, the perceptual control theory models outperform alternative models in their match with behavioral data. These include the domains of tracking performance (Bourbon & Powers, 1999), robotics (Rabinovich & Jennings, 2010), experimental psychology (Marken, 2013), and organizational psychology (Vancouver & Scherbaum, 2008). Controversially, a series of studies have challenged the assumption that increased self-efficacy, a belief in one’s own mastery of a skill (see Chapters 3 and 32, this volume),

123

improves performance (e.g., Halper, Vancouver, & Bayes, 2018; Vancouver et al., 2014). Rather, based on a control theory model, these studies show that reductions in self-efficacy can be associated with improvements in performance because individuals use the discrepancy between their desired and actual performance to increase their effort and improve performance. Within the clinical realm, a recent series of studies presented participants with an image of a spider within a virtual corridor that moved toward and away from the participant. Thus, at any moment, the participant’s actions needed to vary to control the perceived distance from the spider. The spider-fearful participants kept their distance at a value greater than people with low levels of spider fear (Oliver & Mansell, 2018), even when they were required to learn new actions on each trial of the task. Taken together, the “control of input” model of behavior appears well supported by empirical evidence, despite the fact that most studies of behavior assume a “control of output” model, namely that behavior is controlled through planning, learning, or reinforcement. The inherent nature of behavior as dynamically changing has major implications for any model of “behavior change.” Powers (1992) suggests that “We think that our lives are full of repetitive and familiar actions, but to say that we ‘do the same thing’ every day is only a way of speaking . . . Controlling means producing repeatable consequences by variable actions” (p. 5; emphasis added). Yet the converse conclusion, that actions can be determined precisely as the way to counteract a current disturbance – given our adherence to a specific goal – is also true. Control theory entails a paradoxical relationship between pursuing a goal and the degree of control one has over one’s actions in doing so. People potentially surrender control of their actions to external forces in order to achieve a goal. The often complex concrete actions that individuals use to achieve a goal need to change dynamically in order to achieve that goal. For example, for a smoker to be able to

124

WARREN MANSELL

draw cigarette smoke into his lungs, he needs to find a shop that sells cigarettes, ask for a packet, hand over the correct money, take the packet, open it, take a cigarette out, light it, put it to his lips, and inhale. These behaviors vary as necessary depending on the location of the shop, the price of the cigarettes, the size of the packet, the location of the lighter, and so on. So when a researcher says they are attempting to make

smokers “change their behavior,” this cannot be an accurate description of the aim. A more successful aim would be to help people to not to want to perceive inhaling cigarette smoke and its consequential effects on the mind and body (see Figure 9.2). However, by explaining the aim in this way, it seems like a goal to attain “mind control,” which is maybe why it is avoided, because it both feels unethical and seems

Maintain self-ideal

Other principles, e.g., to be helpful, loyal

Be healthy

Other plans, e.g., to exercise, diet

Relieve stress

CONFLICT

Not to ‘be smoking’

Other methods, e.g., meditation Experience effects of nicotine

Inhale cigarette smoke

Nicotine patch on arm

Lit cigarette in mouth

Lit cigarette

Not to inhale cigarette smoke

Not to have lit cigarette in mouth

Uncovered nicotine patch

Not to have lit a cigarette

Lover levels that specify the perceptual outcomes of actions that achieve the goal above

Figure 9.2 An example of a control hierarchy for smoking behavior

Note. This figure illustrates the multiple levels that may be involved in smoking behavior. Starting from the most abstract level – the self-ideal, which approximates to the system concept (Powers, 1973), each level down specifies a branching array of increasingly more concrete perceptual experiences that are controlled in order to maintain the highest level of perception. At each level, only one or two branches are shown but examples of other, parallel perceptual goals are described at the left of the figure. This diagram illustrates the goal conflict involved within one hypothetical person who regards smoking as a way to reduce stress and stay healthy because they also recognize that stopping smoking will improve their health. The diagram shows two alternative behaviors – stopping smoking or using nicotine patches. The latter alternative may provide a way to relieve stress while ceasing smoking.

Changing Behavior Using Control Theory

impossible. It is not impossible to the degree that the manipulated person adheres to a specific goal (e.g., to be liked by the manipulator), but it would not be a helpful way to organize a successful intervention that ultimately improved health and well-being. Instead the environmental conditions need to be created that help people make the intrinsic changes that work for them.

9.3 Mechanisms of Behavior Change In addition to the workings of a control loop, a range of other principles form the architecture of control theory and allow the implications for behavior change to be elaborated. These principles are hierarchical levels, conflict, and reorganization. In the next section, these principles are introduced within a control theory account of the possible reasons for any behavior change. They are summarized as a table in Appendix 9.1 (supplemental materials).

9.3.1 Unlearned Mechanisms The wider implication of the control theory model for behavior change is that behavior is changing most of the time to control perceptual input, and often these changes are not signs of any new learning but just the normal operation of a control system in the face of disturbances or changing feedback functions (see Figure 9.1). These changes typically happen automatically and unconsciously. One could call these unlearned reasons for behavior change, and they include incidents in which the person is trying to control their inputs but is not able to sufficiently. This happens, for example, to the person who has their house flooded and can no longer use all of the feedback functions built into their home to keep them safe, warm, and in touch with other people. Changes that undermine control can be subtle too. Consider, for example, the wellknown “nudge” effects (Thaler & Sunstein,

125

2008), such as when a supermarket places confectionery at the checkout, making it harder for shoppers to avoid buying unhealthy food choices (see Chapter 14, this volume). Even though the immediate changes in behavior that emerge from altering disturbances and feedback functions are not learned, they can still be important because they either help or hinder individuals to achieve their goals. Yet, if the goal is important to the individual, they will adapt to meet it, as described in the next section.

9.3.2 Learned Mechanisms So when does a change in behavior actually indicate a more enduring form of learning? There are many mechanisms available but none of them can be targeted directly by psychological interventions because they change within the nervous system of the individual. Yet they will change spontaneously through learning as one attempts to optimize one’s control during development and as the individual accesses new, and changing, environments. The elements of the negative feedback loop indicate where learning can occur and they provide some examples of learned reasons for behavior change (see Figure 9.1). First, the perceptual function that allows the person to perceive the state of the controlled variable can change. For example, an experienced firefighter improves in her ability to sense the location of a fire, and a man who regularly experiences panic attacks gets more accurate at detecting his own heart rate. Second, the reference value can change. For example, a woman with anorexia may set lower and lower values for the feeling of fullness she is willing to tolerate. According to control theory, the principal purpose of memory is to store reference values. For example, a child stores the memory of the sights, sounds, and smells of her mother and father in order to get close to them when feeling hungry or unsafe. Third, the gain can change. This is often reflected in the amount of

126

WARREN MANSELL

effort a person goes to in order to meet their standard. For example, a person addicted to heroin may go to huge lengths to obtain the drug. Finally, the output function can change; people can learn new means to achieve their goals, such as the gambler who finds new ways to get money to afford the next session at the casino. This last element sounds like a new behavior has been learned, but this is not exactly the case; outputs are also control perceptions. Within control theory, the outputs of a control unit specify the reference values for another control unit – the subgoal at the next level down in the hierarchy; the outputs do not specify actual muscle movements.

9.3.3 Conflict The third indirect reason for behavior change that emerges from control theory is where the individual has lost control due to the effects of conflict. Conflict occurs when two or more control systems attempt to control the same variable at different reference values. It can occur between, or within, one or more individuals or groups. For example, international group conflict often occurs over disputed territory – the same variable, that is, the area of land – that is subject to two opposing standards: to be owned by country A versus to be owned by country B. An example of conflict between individuals is the sixteen-yearold girl who wants to go out with her friends but her mother wants her to stay at home. The same teenage girl may have internal conflict – whether to do what her mother says versus to disobey her. Behavior during conflict can take unusual forms. It can oscillate back and forth as both sides attempt to keep control. When this occurs within an individual, it appears as indecision, uncertainty, and confusion. Conflict can also entail one side “getting its way” while the other side maintains its contradictory standard and yet does not have the resources to win. In this context, a change in situation can “tip the balance” and

allow the suppressed goal to flourish. One example of this is how the suppression of child abuse experiences can be maintained until the victim feels strong and supported enough to speak out and seek justice, with the #MeToo movement a recent example. One further complex side effect of conflict is that two goals in conflict can increase their gain so much that the person overcompensates and their own attempts to counteract disturbances create more disturbances than they can counteract, oscillating the system out of control. It is feasible that such a process occurs in acute episodes of stress, panic, mania, and psychosis. Analogues of these features of conflict have been demonstrated using computer simulations based on control theory (Carey, 2008; McClelland, 2004). A more recent review found that goal conflict is associated with reduced wellbeing and mental health, and control theory has been used to integrate a range of well-known theoretical accounts of its effects (Kelly, Mansell, & Wood, 2015).

9.3.4 Reorganization The fourth indirect reason for behavior change relates to what occurs when a person attempts to regain control through a process known as reorganization. This is a trial-and-error process through which the elements of a control system (e.g., functions, gains, etc.) change in order to help regain control. As a person loses control of what matters to them, and error builds up, this ramps up the rate of random changes in the control system itself through new neural connections, resulting in, for example, new ways of perceiving (perceptual functions) and new ways of acting (output functions). Thus, apparently random, or unhelpful, behaviors can sometimes be observed in people who are attempting to try to regain control. The process of reorganization has been examined in computer simulations (Powers, 2008). Further studies have inferred that the process of reorganization is occurring when a period

Changing Behavior Using Control Theory

of poor performance precedes a sudden “step change” in ability during skill acquisition in infants (Sadurní, Burriel, & Plooij, 2010) and in experimental tasks (Robertson & Glines, 1985).

9.3.5 Awareness The fifth and final indirect reason for observing behavior change is during changes in the focus and duration of awareness. Because conflict is the most pernicious factor that undermines control and cooperation between individuals, it is vital that human beings have a way to resolve it. Powers proposed that reorganization needs to be directed to the control system that is superordinate to the conflicting goals in order for the conflict to be resolved by trial-and-error changes in the parameters of that system. Thus, at any one point in time, the spotlight of awareness is directed to only a tiny subset of the control systems within an individual, and it occurs during the imagination mode, a specific configuration of control systems that permits mental simulation (Powers, Clark, & McFarland, 1960). As a person’s awareness shifts, this is likely to be accompanied by changes in behavior – for example, focusing on controlling visual imagery of one’s goals rather than following a group conversation (see Chapter 33, this volume).

9.3.6 Integration and Critique of Behavior Change Interventions The control theory framework described here provides a rubric for describing the content and target of existing behavior change interventions. In particular, it reveals that any intervention may be associated with an array of behavior changes, only some of which represent the kind of enduring changes that are desired. Interventions that aim to disturb people’s attempt to control what is currently important to them are likely to have short-term benefits if they can effectively block people’s actions, but they are likely to have very limited effectiveness in the long term because human control systems are designed

127

to counteract disturbances to meet their goals; people are purposeful and will raise their efforts, sometimes to the extreme of violence, to achieve them. Intervention schemes that utilize threat, withholding of rewards, and restriction of availability would be of this kind. Intervention schemes that provide resources and opportunities (feedback functions) for people to meet their goals and needs have the potential for long-term benefits, partly through helping people meet their goals and partly through making it less likely that people will be drawn into conflict over limited resources. However, even in a plentiful environment, the most fundamental conflicts (e.g., between being honest vs. kind; being adventurous vs. safe) persist. The most pernicious conflicts are present within the individual and may be outside their immediate awareness. Therefore, the paramount aim is to provide opportunities for people to resolve conflict. Control theory suggests that people require environments in which they feel safe enough to experience uncertainty, error, and risk. Further, they require environments in which they can allow the trialand-error changes observable in their thoughts, feelings, and behavior to persist long enough to identify reorganizations of the control systems driving the conflicting goals that enable them to restore control. Interventions of this kind are harder to pin down but they would include campaigns, groups, and therapies that embrace uncertainty, risk, making mistakes, individuality, creativity, play, freedom of expression, and exploration. The following section introduces interventions informed by control theory that incorporate some of these key elements.

9.4 Evidence Base for the Use of Control Theory in Changing Behavior 9.4.1 Child Development and Parenting Possibly one of the most successful sets of resources based on control theory has been the

128

WARREN MANSELL

Wonder Weeks program (van de Rijt-Plooij & Plooij, 1993). The Wonder Weeks program, initially published as a book and later launched as a mobile phone app, is aimed at new parents and describes to them the early stages of infant development in a highly accessible way. It explains how the infant’s “fussy” periods are transitory; are preceded by, or co-occur with, the emergence of a new type of perception; and are followed by a “mental leap” shown in new skills and abilities. Within control theory terms, each mental leap involves the construction of an input function that extracts an increasingly abstract perceptual signal that can now be controlled via the already developed lower-level systems through feedback control. By helping parents to understand this process in lay terms, the authors aim to help parents work through their baby’s challenging developmental stages. Based on control theory, Plooij’s research has indicated that each developmental stage represents a new class of perception that the baby is learning to control (van de RijtPlooij & Plooij, 1993). Skill development in infancy is consistent with our earlier description of reorganization; until the child masters a new skill through trial-and-error reorganization, they can experience the distress of losing control temporarily and show unusual changes in behavior. Based on the Wonder Weeks program, van de Rijt-Plooij, van der Stelt, and Plooij (1996) developed a Dutch parental support and education program called “Hordenlopen” (“Leaping Hurdles”) for a group of single mothers who were at risk of abusing their infants. The course made parents aware of the regression periods, explained that their babies could not help being difficult, and showed them how they could comfort their babies in these difficult periods and facilitate the new type of perception and consequent learning. In an evaluation study, an experimental group receiving the Leaping Hurdles program was compared with a control group following an alternative, current program. In terms of behavior change, the study found that parents in the Leaping

Hurdles program reported more control over their parenting, their infants scored higher on the measures of skill performance, and their children were socially more accepting and open toward strangers. Notably, no research appears to have been conducted on the changes that might occur within parents after the program. One might predict that their reference values for what would constitute a problem with their child might increase following the program, such that the parent is able to manage the child’s distress during the period of regression rather than increasing their efforts in a way that overcompensates for the child’s distress and exacerbates it.

9.4.2 Education Control theory has informed a variety of interventions within the education system and a number of manuals have been published (Carey, 2012; Ford, 1997; Good, Grumley, & Roy, 2003). They each advocate a whole-school approach based on perceptual control theory. The Responsible Thinking Process was developed by Ford (1997) as a way to help staff and students address infringements of the school rules in a mutually respectful manner. It uses specific questioning to help students weigh up for themselves the internal conflicts between their own goals and those of the other pupils, teachers, and the school as whole. There are no published evaluations of the program, although a qualitative case study has reported on the features of effective implementation of the approach (Rynberg, 2016). The Connected School (Good et al., 2003) takes a wider perspective to include all of the relationships within the school, and it takes a more direct approach toward explaining the principles of control theory to school administrators and teachers, utilizing role play and experiential methods. It involves three 4-day workshops and 30 hours of supervision. Good (2010) summarizes the reports from schools across Canada and the United States that describe the impact of the program on

Changing Behavior Using Control Theory

increasing school attendance, reducing discipline incidents, reduced teacher sick leave, and improvements in students’ grades. Finally, a wider perspective is taken in Carey’s (2012) Control in the Classroom book, illustrating how control theory informs education as well as relationships and discipline. He builds on earlier empirical work showing how staff and students can be locked into counter-control relationships when punitive or coercive methods are attempted by schoolteachers (Carey & Bourbon, 2006). He also illustrates how any apparently disruptive behavior can be understood as an attempt at control. Teachers can utilize empirical principles to “test the controlled variable” (see Sidebar 9.2), and they can use questioning to help pupils to explore and understand their own goals and goal conflict.

129

9.4.3 Mental Health Control theory has informed mental health interventions for a number of decades. For a while, it formed the basis of reality therapy (Glasser, 1981) and paved the way for the development of Grawe’s (2004, 2007) comprehensive integration of psychological therapies. In more recent years, it has formed the basis of an individual psychotherapy known as method of levels (MOL; Carey, 2006) therapy, following the initial, unpublished work of Powers during the 1950s and 1960s. The MOL therapist allows the client to lead the focus of the session and helps the client to shift and sustain their awareness to the source of the conflict that underlies their difficulties (see Chapter 45, this volume). The MOL therapist manages this through simple questioning about

Sidebar 9.2 Control theory: The behavioral illusion and testing the controlled variable

Since the rise of behaviorism during the early twentieth century, experimental psychology became the study of the prediction and control of behavior (Mansell & Carey, 2008). Despite the “cognitive revolution” in the 1950s, research designs persist in attempting to manipulate “stimuli” and measure responses in order to infer the nature of the intervening cognitive processes (Marken, 2002). Yet observers can be mistaken in judging the nature of another person’s behavior. One series of studies shows that a volunteer can be given a simple instruction (to keep a knot of a rubber band over a dot on a page while holding a pen) and yet over 90 percent of observers of this activity fail to correctly identify that this was the volunteer’s instruction (Willett et al., 2017). This behavioral illusion occurs because observers notice the eyecatching features of a behavior, in this case drawing a shape on the page, and assume that what they have observed was intended or controlled. Yet, according to Powers, and evident in replicated studies, what is controlled is the perception of an aspect of the self or environment. Powers concludes that the science of behavior should therefore require a test for the controlled variable (TCV). The TCV uses the deductive scientific method. The researcher hypothesizes what aspect of the self or environment the participant is controlling. Next, they apply a variety of disturbances to this putative variable and, if the variable is not controlled by the participant, it should change accordingly. If the variable is, however, kept protected from these disturbances, then it is the controlled variable. The TCV is essentially a way to infer a person’s goals, or intentions, to “mind read” without requiring self-report (Marken, 1982).

130

WARREN MANSELL

present-moment experience, either to help the client describe the problem (Goal One) or to notice disruptions that may indicate “background thoughts” about the problem (Goal Two). Background thoughts may include verbal thoughts, images, feelings, and memories and they allow the client to go “up a level” to their more fundamental goals. Ideally, clients control their own access to MOL, so that they get it when they want it, as frequently as they want it, for as many sessions as they want. Thus, the principle of control runs through both the delivery and the practice of MOL. To be specific, the therapist’s questions act as feedback functions for the client, facilitating the client’s goal of trying to put into words, explore, and ultimately resolve a problem. This can often involve helping the client keep difficult experiences in their awareness – i.e., perceptual signals that are far from their desired reference value – because this aids reorganization to resolve goal conflict. When the therapist asks about disruptions, this aids the expression of the client’s higher-level goals that are at the edge of awareness. By bringing these to the forefront of awareness, they can be explored; and again this permits reorganization of the conflict that they may be creating. Within Figure 9.2, the client may begin with their awareness at the level of “experiencing the effects of smoking/nicotine” and during MOL become aware that this serves to “reduce stress” which, in turn, at a higher level, serves to “stay healthy.” It is only from this perspective that the conflict is clear – the paradox of carrying out an unhealthy activity to stay healthy. An MOL therapist would help a client to get to this realization, stay there, and explore it on their own terms. Appendix 9.2 provides more details on how to carry out MOL therapy. Most studies on MOL therapy have involved pragmatic samples of patients with wide-ranging mental health problems and they have shown large pre-post effect sizes on measures of distress (Carey & Mullan, 2008; Carey et al., 2009;

Cocklin et al., 2017). One study within a secondary care mental health service showed larger effect sizes than benchmarked studies and a greater efficiency ratio – the reduction in distress per session (Carey, Tai, & Stiles, 2013). No randomized trials of MOL therapy have been published but two other mental health interventions based on control theory have been evaluated. One of these, MYLO (Manage Your Life Online), is a computerized agent who engages in a text-based conversation with the user about a current problem (see Chapter 29, this volume). Two studies of students have shown reductions in distress about a problem relative to a comparison computer agent (ELIZA) but no differences in the reduction of anxiety and depression (Bird et al., 2018; Gaffney et al., 2014). The second intervention is a six-session program known as the Take Control Course (TCC), which provides experiential learning on the principles of control theory to primary care mental health patients with a variety of diagnoses (Morris, Mansell, & McEvoy, 2016). The TCC has been shown to be noninferior at six months follow-up compared to the “low intensity” individual therapy that is typically offered in primary care (Morris et al., 2019). Currently, the TCC is being evaluated in high schools for adolescents with diverse mental health problems. Taken together, mental health interventions derived from control theory show clear promise, and they are being practiced widely. In addition, the computerized exposure paradigm mentioned in Section 9.2 found specific effects on behavior (Oliver & Mansell, 2018). Spider-fearful participants were exposed to moving images of spiders in a “virtual corridor” (Healey, Mansell, & Tai, 2017). They were randomly allocated to one of two yoked conditions – a high control condition in which they could control their perceived distance from the spider image on the screen and a low control condition who received the distances from a yoked high control participant but had very little control. The high control group approached closer to a real spider in a tank after

Changing Behavior Using Control Theory

the task and they reported avoiding spiders less than the low control group two weeks later.

9.4.4 Organizations Research applying control theory in the field of organizational psychology has focused on building and testing computer models of psychological processes within organizations and work environments (e.g., Vancouver & Purl, 2017). However, authors have developed control theory interventions for marketing and businesses (Forssell, 2008; Madden, 2014; Twijnstra & Plooij, 2011). The fundamental contribution of control theory to organizations is the realization that it is not productive to instruct other people’s behavior or coerce people into behaving in a particular way. Using control theory, teamwork and leadership are more effectively managed by talking openly about the organization’s goals and soliciting feedback from all members on their view of these goals, their ability to contribute to them, and any conflicting goals that may get in the way. Members are given their own responsibility to do things that contribute to achieving and maintaining the organization’s goals. Control theory ideas can also be used by individuals to help professional development in a way that is genuinely transformative (Twijnstra & Plooij, 2011). A number of studies have tested and confirmed the validity of the control theory model of the hierarchical organization of goals (e.g., Paulssen & Bagozzi, 2006; Pieters, Baumgartner, & Allen, 1995). Soldani (1989) reported a study of a control theory teamwork program within a manufacturing group that was struggling to meet production schedules. The intervention was based on the assumption, described in Section 9.3.3, that conflict undermines control and effective performance. In turn, control is reestablished, by all parties, when given the opportunity through open dialogue to sustain awareness on the sources of conflict and allow reorganization within the individuals involved. The interventions involved

131

agreeing on a focal goal for the organization – to meet 95 percent of the production schedule on time from a baseline level of 23 percent. The intervention required meetings between all individuals whose direct or indirect work might support the goal, and opinions and feedback were sought regularly from all members. Each member had a one-to-one meeting with a facilitator (acting as a feedback function) to discuss and raise awareness of any obstacles (disturbances) to the goal. In particular, this provided the opportunity to identify when different people’s goals in the organization were not aligned. For example, one manager admitted that she was trying to control her reputation, and she tried to achieve this by scolding employees who she perceived to be challenging it, which entailed that many of the employees controlled for avoiding being criticized, making team meetings conflicted and unproductive. The facilitator was able to use simple questioning to each member to help them reflect on their goals and reprioritize them, given the negative impacts they were having on one another. Each member was given the choice of whether they wanted to pursue the organizational goal or give up their place for another member. Soldani (1989) reported that the group met its goal after the program with no loss of quality or increase on overtime.

9.5 Future Research and Interventions Thus far, the existing evidence for control theory interventions across a wide range of domains has been summarized. However, there is emerging research and practice within a much wider arena of behavior change. This includes patient-perspective health care (Carey, 2017), skills training (Brown-Ojeda & Mansell, 2018), medication prescribing behaviors (Ferguson, Keyworth, & Tully, 2018), and dementia (McEvoy et al., 2016). The breadth of applications underlies a fundamental feature of control theory – it is a

132

WARREN MANSELL

theory that specifies the same principles for all realms of behavior, across societies and culture. It illustrates the generalizability and universality of its principles to human behavior across populations, contexts, and behaviors. Dividing people up into groups based on their roles ignores the fact that individual behaviors involve the same control processes; and what is involved in a person’s behavior, and what they learn from that process, in one domain, can apply to another. Based on current theory and research on control theory, an important translational message for policymakers and funders is that investing in generalized approaches to behavior change based on control theory would be vastly more efficient than funding projects for specific “groups.” One related advantage of the unifying focus of control theory is that it can bring together diverse theories under a “common language” (Marken & Mansell, 2013). Examples include the integration of various theories of addiction using control theory (Webb et al., 2010); the integration of self-determination theory (see Chapter 8, this volume) with theories of goal conflict, ambivalence, and self-discrepancy (Kelly et al., 2015); and the analysis of cognitive and behavioral techniques and implementation intentions (see Chapter 39, this volume) acting at different levels of a control hierarchy (Mansell, 2012).

9.6 Summary and Conclusion Powers’s control theory provides a unique perspective on behavior change. It assumes, unlike other theories, that behavior is the means to control perceptual input. It also provides an architecture of components that can be used to model people’s behavior and its validity can be established through the fit of the model to the behavioral data. It implies a number of different routes to behavior change, each varying in their likely success and longevity. In particular, it points to the detrimental role of conflict between and

within individuals and the benefits of ways to help people bring conflicts to awareness and explore them in a way that promotes spontaneous solutions – through a process known as reorganization. The breadth of its application is evident, but clearly further empirical work is necessary to evaluate its enduring benefits and compare them with interventions derived from other established theories of behavior change.

References Ashby, W. R. (1952). Design for a Brain. New York: Wiley. Bird, T., Mansell, W., Wright, J., Gaffney, H., & Tai, S. (2018). Manage Your Life Online: A web-based randomized controlled trial evaluating the effectiveness of a problem-solving intervention in a student sample. Behavioural and Cognitive Psychotherapy, 46, 570–582. https://doi .org10.1017/S1352465817000820 Bourbon, W. T., & Powers, W. T. (1999). Models and their worlds. International Journal of HumanComputer Studies, 50, 445–461. https://doi.org/ 10.1006/ijhc.1998.0263 Brown-Ojeda, C., & Mansell, W. (2018). Do perceptual instructions lead to enhanced performance relative to behavioral instructions? Journal of Motor Behavior, 50, 312–320. Carey, T. A. (2006). The Method of Levels: How to Do Psychotherapy without Getting in the Way. Hayward, CA: Living Control Systems Publishing. Carey, T. A. (2008). Conflict, as the Achilles heel of perceptual control, offers a unifying approach to the formulation of psychological problems. Counselling Psychology Review, 23, 5–16. Carey, T. A. (2012). Control in the Classroom: An Adventure in Learning and Achievement. Hayward, CA: Living Control Systems Publishing. Carey, T. A. (2017). Patient-Perspective Care: A New Paradigm for Health Systems and Services. London: Routledge. Carey, T. A., & Bourbon, W. T. (2006). Is countercontrol the key to understanding chronic

Changing Behavior Using Control Theory

behavior problems? Intervention in School and Clinic, 42, 5–13. https://doi.org/10.1177/ 10534512060420010201 Carey, T. A., Carey, M., Mullan, R. J., Spratt, C. G., & Spratt, M. B. (2009). Assessing the statistical and personal significance of the method of levels. Behavioural and Cognitive Psychotherapy, 37, 311–324. https://doi.org/10.1017/ S1352465809005232 Carey, T. A., & Mullan, R. J. (2008). Evaluating the method of levels. Counselling Psychology Quarterly, 21, 247–256. https://doi.org/10.1080/ 09515070802396012 Carey, T.A., Tai, S.J. & Stiles, W.M. (2013). Effective and efficient: Using patient-led appointment scheduling in routine mental health practice in remote Australia. Professional Psychology: Research and Practice, 44, 405–414. https://doi .org10.1037/a0035038 Carver, C. S., & Scheier, M. F. (1982). Control theory: A useful conceptual framework for personalitysocial, clinical, and health psychology. Psychological Bulletin, 92, 111–135. https://doi .org/10.1037/0033-2909.92.1.111 Cocklin, A. A., Mansell, W., Emsley, R. et al. (2017). Client perceptions of helpfulness in therapy: A novel video-rating methodology for examining process variables at brief intervals during a single session. Behavioural and Cognitive Psychotherapy, 45, 647–660. https://doi .org10.1017/S1352465817000273 Ferguson, J., Keyworth, C., & Tully, M. P. (2018). “If no-one stops me, I’ll make the mistake again”: Changing prescribing behaviors through feedback; A Perceptual Control Theory perspective. Research in Social and Administrative Pharmacy, 14, 241–247. https:// doi.org/10.1016/j.sapharm.2017.03.001 Ford, E. E. (1997). Discipline for Home and School. Scottsdale, AZ: Brandt Publications. Forssell, D. (2008). Management and Leadership: Insight for Effective Practice. Hayward, CA: Living Control Systems Publications. Gaffney, H., Mansell, W., Edwards, R., & Wright, J. (2014). Manage Your Life Online (MYLO): A pilot trial of a conversational computer-based intervention for problem solving in a student

133

sample. Behavioural and Cognitive Psychotherapy, 42, 731–746. https://doi .org10.1017/S135246581300060X Glasser, W. (1981). Stations of the Mind: New Directions for Reality Therapy. New York, USA: Harper Collins. Good, F. (2010). A Connected School: Improving Schools and Communities from the Inside Out. www.pctweb.org/Connected.pdf Good, E. P., Grumley, J., & Roy, S. (2003). A Connected School. Chapel Hill, NC: New View Publications. Grawe, K. (2004). Psychological Therapy. Cambridge, MA: Hogrefe Publishing. Grawe, K. (2007). Neuropsychotherapy: How the Neurosciences Inform Effective Psychotherapy. Mahwah, NJ: Lawrence Erlbaum Associates. Halper, L. R., Vancouver, J. B., & Bayes, K. A. (2018). Self-efficacy does not appear to mediate training’s effect on performance based on the moderationof-process design. Human Performance, 31, 216– 237. https://doi.org/10.1080/ 08959285.2018.1509343 Healey, A., Mansell, W., & Tai, S. (2017). An experimental test of the role of control in spider fear. Journal of Anxiety Disorders, 49, 12–20. http://dx.doi.org/10.1016/j.janxdis.2017.03.005 Kelly, R. E., Mansell, W., & Wood, A. M. (2015). Goal conflict and well-being: A review and hierarchical model of goal conflict, ambivalence, selfdiscrepancy and self-concordance. Personality and Individual Differences, 85, 212–229. https:// doi.org/10.1016/j.paid.2015.05.011 Madden, B. J. (2014). Reconstructing Your Worldview. Naperville, IL: LearningWhatWorks. Mansell, W. (2012). The transdiagnostic approach: The importance of formulating interpersonal and intrapersonal control. In W. Dryden (Ed.), Cognitive Behavioural Therapies (pp. 261–283). London: SAGE. Mansell, W., & Huddy, V. (2018). The assessment and modeling of perceptual control: A transformation in research methodology to address the replication crisis. Review of General Psychology, 22, 305– 320. https://doi.org/10.1037/gpr0000147 Marken, R. (1982). Intentional and accidental behavior: A control theory analysis. Psychological Reports,

134

WARREN MANSELL

50, 647–650. https://doi.org/10.2466/ pr0.1982.50.2.647 Marken, R. (2002). Looking at behavior through control theory glasses. Review of General Psychology, 6, 260–270. Marken, R. S. (2013). Taking purpose into account in experimental psychology: Testing for controlled variables. Psychological Reports, 112, 184–201. https://doi.org/10.2466/03.49.PR0.112.1.184201 Marken, R. S., & Mansell, W. (2013). Perceptual control as a unifying concept in psychology. Review of General Psychology, 17, 190–195. HTTPS://DOI.ORG10.1037/a0032933 McClelland, K. (2004). The collective control of perceptions: Constructing order from conflict. International Journal of Human-Computer Studies, 60, 65–99. https://doi.org/10.1016/ j.ijhcs.2003.08.003 McEvoy, P., Eden, J., Morris, L., & Mansell, W. (2016). Dementia: Towards a perceptual control theory perspective. Quality in Ageing and Older Adults, 17, 229–238. https://doi.org/10.1108/QAOA-032015-0013 Morris, L., Lovell, K., McEvoy, P. et al. (2019). A brief transdiagnostic group (the Take Control Course) compared to individual low-intensity CBT for depression and anxiety: a randomized noninferiority trial. Unpublished manuscript, University of Manchester. Morris, L., Mansell, W., & McEvoy, P. (2016). The Take Control Course: Conceptual rationale for the development of a transdiagnostic group for common mental health problems. Frontiers in Psychology, 7, 99. https://doi.org/10.3389/ fpsyg.2016.00099 Oliver, K., & Mansell, W. (2018). What is avoidance and when is it a problem? A control theory approach to approach-avoidance conflict in spider fears using a force-feedback joystick paradigm. Unpublished manuscript, University of Manchester. https://doi.org/10.13140/ RG.2.2.19390.38724 Paulssen, M., & Bagozzi, R. P. (2006). Goal hierarchies as antecedents of market structure. Psychology and Marketing, 23, 689–709. https://doi.org/ 10.1002/mar.20124

Pieters, R., Baumgartner, H., & Allen, D. (1995). A means-end chain approach to consumer goal structures. International Journal of Research in Marketing, 12, 227–244. https://doi.org/10.1016/ 0167-8116(95)00023-U Powers, W. T. (1973/2005). Behavior: The Control of Perception. New York: Hawthorne. Powers, W. T. (1978). Quantitative analysis of purposive systems: Some spadework at the foundations of scientific psychology. Psychological Review, 85, 417–435. http://dx.doi .org/10.1037/0033-295X.85.5.417 Powers, W. T. (1992). Living Control Systems II: Selected Papers of William T. Powers. Gravel Switch, KY: The Control Systems Group. Powers, W. T. (2008). Living Control Systems III: The Fact of Control. New York: Benchmark Publications. Powers, W. T., Clark, R. K., & McFarland, R. L. (1960). A general feedback theory of human behavior: Part I. Perceptual and Motor Skills, 11, 71–88. Rabinovich, Z., & Jennings, N. R. (2010). A hybrid controller based on the egocentric perceptual principle. Robotics and Autonomous Systems, 58, 1039–1048. https://doi.org/10.1016/j .robot.2010.05.002 Robertson, R. J., & Glines, L. A. (1985). The phantom plateau returns. Perceptual and Motor Skills, 61, 55–64. https://doi.org/10.2466/pms.1985.61.1.55 Rynberg, A. D. (2016). Using RTP (responsible thinking process) as a lever for improving school culture: A case study of an alternative secondary school’s implementation of RTP. Unpublished PhD dissertation, Western Michigan University. Sadurní, M., Burriel, M. P., & Plooij, F. X. (2010). The temporal relation between regression and transition periods in early infancy. The Spanish Journal of Psychology, 13, 112–126. https://doi .org/10.1017/S1138741600003711 Soldani, J. (1989). Effective personnel management: An application of perceptual control theory. Advances in Psychology, 62, 515–529. http:// ww.livingcontrolsystems.com/intro_papers/per sonnel_mgmt.pdf Thaler, R. H., & Sunstein, C. R. (2008). Nudge: Improving Decisions about Health, Wealth, and Happiness. New Haven, CT: Yale University Press.

Changing Behavior Using Control Theory

Twijnstra, M. H., & Plooij, F. X. (2011). Oei, ik groei! voor managers [Oops, I’m Growing! for Managers]. Utrecht: Kosmos Uitgevers. Vancouver, J. B. (2005). The depth of history and explanation as benefit and bane for psychological control theories. Journal of Applied Psychology, 90, 38–52. https://doi.org/10.1037/00219010.90.1.38 Vancouver, J. B., Gullekson, N. L., Morse, B. J., & Warren, M. A. (2014). Finding a between-person negative effect of self-efficacy on performance: Not just a within-person effect anymore. Human Performance, 27, 243–261. https://doi.org/ 10.1080/08959285.2014.913593 Vancouver, J. B., & Purl, J. D. (2017). A computational model of self-efficacy’s various effects on performance: Moving the debate forward. Journal of Applied Psychology, 102, 599–616. http://dx .doi.org/10.1037/apl0000177 Vancouver, J. B., & Scherbaum, C. A. (2008). Do we self-regulate actions or perceptions? A test of two computational models. Computational and Mathematical Organization Theory, 14, 1–22. https://doi.org10.1007/s10588-008-9021-7 van de Rijt-Plooij, H., & Plooij, F. X. (1993). Distinct periods of mother-infant conflict in normal

135

development: Sources of progress and germs of pathology. Journal of Child Psychology and Psychiatry, 34, 229–245. https://doi.org/10.1111/ j.1469-7610.1993.tb00981.x van de Rijt-Plooij, H., van der Stelt, J. & Plooij, F. (1996). Hordenlopen: Een preventieve oudercursus voor de eerste anderhalf jaar [Hurdling: A Preventive Parent Course for the First Year and a Half]. Lisse: Swets & Zeitlinger. Webb, T. L., Sniehotta, F. F., & Michie, S. (2010). Using theories of behaviour change to inform interventions for addictive behaviours. Addiction, 105, 1879–1892. https://doi.org/10.1111/j.13600443.2010.03028.x Wiener, N. (1948). Cybernetics and Control and Communication in the Animal and the Machine. Cambridge, MA: Technology Press. Willett, A., Marken, R. S., Parker, M., & Mansell, W. (2017). Control blindness: Why people can make incorrect inferences about the intentions of others. Attention, Perception, and Psychophysics, 79, 841–849. https://doi.org/10.3758/s13414-0161268-3 Young, R. (2017). A general architecture for robotics systems: A perception-based approach to artificial life. Artificial Life, 23, 236–286.

10 Changing Behavior Using the Transtheoretical Model Carlo C. DiClemente and Meagan M. Graydon

Practical Summary The transtheoretical model (TTM) was developed in the 1980s and offered a novel way to view intentional change whether that is quitting smoking, increasing physical activity, or managing depression and anxiety. Rather than considering people as “not motivated,” the TTM classifies people who need to start, stop, or modify a behavior into five stages that mark the journey to successfully sustained change. These stages identify critical tasks for people who are not yet considering change (precontemplation), those who are thinking about and possibly are ambivalent about change (contemplation), those preparing and planning to make a change (preparation), and those making and maintaining changes (action and maintenance, respectively). The TTM processes of change consider how people move through these stages and identify common personal activities that facilitate behavior change and movement through the stages. The TTM is used to tailor various therapeutic and selfhelp interventions by personalizing strategies to specific stages of change, which often leads to more successful change outcomes.

10.1 Introduction The transtheoretical model (TTM) emerged from the complexity and confusion about psychotherapy approaches among researchers and practitioners in the 1970s. Therapy theories multiplied, each claiming superiority in predicting and explaining what people need to do to make change happen, despite research suggesting no difference between treatment approaches (Sloane et al., 1975). A systematic review of these major theories indicated that there may be some common processes of change derived from different theories that could become the foundation of an integrated eclectic framework (Goldfried, 1980; Prochaska, 1979). Moving the field forward, research examining processes of change with smokers clarified

that the processes or principles of change experienced by those attempting to quit smoking were embedded in larger, more extensive processes of behavior change (DiClemente, 1978). This research and theorizing led to the development of a model and several initial publications (Prochaska & DiClemente, 1982, 1983, 1984). These publications outlined the stages and processes of change as the basis for understanding behavior change and creating an integrative perspective for therapy. Over the past thirty-five years, the stages and processes of change and their interaction, which are at the core of the model, have been studied with a wide variety of medical, mental health, and addictive behaviors. It also has been used with https://doi.org/10.1017/9781108677318.0010

Changing Behavior Using the Transtheoretical Model

health promotion and protection behaviors, including exercise, diet, and cancer screening. The focus of the model is the process of intentional behavior change that leads to successful sustained change, although it should be noted that there are other types of behavior change not addressed in the model, namely developmental and imposed behavior change (DiClemente, 2018). This chapter provides an overview of the TTM and its components. In addition, it outlines how this dynamic process describes the change journey of an individual and how it fits into treatment or prevention and can be used to guide interventions for initiation, modification, and cessation patterns of behavior change. Finally, the chapter examines some of the research that explores and supports various elements of the TTM and offers some implications and directions for future research.

10.2 Brief Overview of the Theory and Evidence There are four important elements of the TTM: stages, processes, context, and markers of change. The model describes a process that is multidimensional with interactions among the key constructs. The first two elements, stages and processes of change, describe the client tasks and mechanisms that drive the behavior change process. The stages represent a series of critical tasks that describe the motivational and temporal dimensions of behavior change. The processes detail important coping activities and mechanisms that enable the completion of stage tasks and can promote movement through the stages. The interaction of the stages and the processes of change is the essential seminal insight into the behavior change process. Context and markers also interact with process activity and stage tasks. Stages of change highlight where one is in the process of changing a specific behavior. The five stages identified in the TTM are precontemplation, contemplation, preparation, action, and

137

maintenance. These stages represent states, not traits, and indicate where one is regarding the motivational and behavior change tasks. In multiple studies, stage status, or readiness of individuals, identified substantive differences in attitudes, intentions, decisional considerations, and behavioral activities among individuals in different stages and predicted behavior change outcomes (DiClemente & Hughes, 1990; DiClemente et al., 1991; Project MATCH Research Group, 1997). Table 10.1 offers a view of each of these stages and their relevant client and provider tasks. There have been several measures and approaches that have been used to evaluate stage status and examine the relationships with other constructs as well as the effect of stage status or readiness on outcomes. The most clinically useful way to assess stages is either with a one-item measure, like the readiness ruler (see Sidebar 10.1), or to use a sensitive interview to assess stage tasks (Connors et al., 2013).1 Processes of change are the mechanisms and coping activities that enable an individual to accomplish tasks and move toward change. They were derived from change principles identified by different systems of psychotherapy, justifying the term transtheoretical (Prochaska & DiClemente, 1984; Prochaska, DiClemente, & Norcross, 1992). Conceptual and psychometric analyses have identified ten specific processes that fall into two larger categories of five cognitive/experiential and five behavioral processes (Prochaska et al., 1988). Multi-item subscales have been used to assess the ten processes. The processes of change are dynamically related to the client engagement in tasks across the stages (Prochaska & DiClemente, 1986; DiClemente et al., 1991) and movement through the stages (Carbonari & DiClemente, 2000). These 1

Measures of stages, processes, and markers of change can be found at the following websites: www.uri.edu/ cprc and www.umbc.edu/psyc/habits.

138

CARLO C. DICLEMENTE AND MEAGAN M. GRAYDON

Table 10.1 Stage of change: example client and provider tasks for behavior change intervention Stage of Change

Client Tasks

Provider Tasks

Precontemplation • Increase awareness, concern, hope, and • Engage and accept client confidence • Build rapport • Increase client’s perception of risk and problems • Normalize ambivalence Contemplation • Conduct risk-reward analysis of pros • Evoke reasons for change, risks of not and cons of change changing • Tip decisional balance • Help tip decisional balance • Solidify decision to change • Strengthen client’s self-efficacy • Offer a menu of options Preparation • Commit to change • Help client determine the best course of • Create an effective and appropriate action change plan • Develop a plan, considering barriers for quitting and social support • Help client implement the plan Action • Implement change plan • Help client identify and develop skills to • Problem solve and revise plan, as cope with change necessary • Help client problem solve • Help client identify strengths and Maintenance • Integrate new behavior into lifestyle strategies to prevent relapse • Develop strategies for preventing • Resolve relational issues and associated relapse problems • Engage with social support • Provide support Relapse • Revise change plan • Determine triggers and develop • Reimplement new plan prevention plan • Help client recycle through stages again Note. Relapse is not a stage of change but an event that triggers a recycling through stages in order for the individual to accomplish the tasks adequately to support maintenance. Sidebar 10.1 Using a readiness ruler to assess motivation and stage of change status

On a scale of 1 to 10 with 1 being “not ready at all” and 10 being “very ready,” how ready are you to do this … or make this change …? 1 2 3 Low Readiness

4 5 6 Moderate Readiness

7

8 9 10 High Readiness

Lower numbers usually indicate that the individual is in precontemplation; and, as they move to moderate readiness, they move through contemplation and into preparation stage as they approach high readiness. If the individual says they are a 10, often they are already taking steps and may be in action. The ruler can start a conversation that allows the therapist to explore the various stage tasks and discuss activities related to the experiential and behavioral processes of change.

Changing Behavior Using the Transtheoretical Model

139

Table 10.2 Defining and recognizing client processes of change Client Process of Change Experiential Processes Consciousness Raising Emotional Arousal Self-Reevaluation Environmental Reevaluation Social Liberation Behavioral Processes Reinforcement Management Helping Relationship Counterconditioning Stimulus Control Self-Liberation

Definition/Activity Becoming more aware, gaining knowledge about problem or solution Affective experiences related to the current behavior or change Recognizing value conflicts and realigning values supporting change. Changing view of self Understanding how current behavior and the change of that behavior might affect significant others or activities Becoming aware of the social mores, policies, and opportunities relating to current behavior or future change Finding behaviors and activities that are reinforcing for change and disengaging from reinforcers that support status quo Finding a person(s) who can be trusted and with whom you can share successes and struggles in the journey of change Becoming aware of cues that trigger behaviors and using alternative responses to the cues Becoming aware of cues that trigger behaviors and avoiding or managing/ changing the cues Taking responsibility for one’s behavior and choosing to make change or engage in change process

measures typically include statements about frequency of client’s activities related to the definition of the processes in Table 10.2. Measurement of the processes has been found to be highly reliable and valid across problem areas (Rosen, 2000). There is an interaction between the stages and processes of change. The cognitive/experiential processes appear more relevant and important in completing tasks in earlier stages (i.e., before the action stage) that focus on developing interest and concern and decision-making. In contrast, the behavior processes are more important as the individual moves into the preparation, action, and maintenance stages and needs to employ active coping activities and behavioral strategies. Health behavior change and recovery from mental health and addiction problems have multiple types of behavior change targets that represent three patterns of change: initiating new behaviors, stopping problematic behaviors, or

modifying dysregulated and risk behaviors. The same model of stages and processes has been found to be relevant for all patterns of change. The processes of change interact with stage status across a variety of behaviors, from initiating physical activity (Marcus et al., 1992) to quitting smoking and drinking (Carbonari & DiClemente, 2000; DiClemente et al., 2001; Prochaska et al., 1988) and modifying diet (DiClemente et al., 2015). In fact, the interaction between stages and processes has been used to describe and document the journey into and out of addictions (DiClemente, 2018). Since the model seems to be applicable to different patterns of change and a range of target behaviors, it can be used to guide both prevention and treatment. Prevention activities include interfering with the development of problematic behaviors, initiation of health protective behaviors, and modification of problematic risk behaviors. Treatment activity targets include initiation and

140

CARLO C. DICLEMENTE AND MEAGAN M. GRAYDON

engagement in treatment and treatment strategies, medication adherence, elimination of healththreatening behaviors, modification of coping skills, and readiness to achieve specific personal behavior change goals. However, it is important to remember that the TTM describes the client’s process of change as the focus of either prevention or treatment activities. In addition, change can be accomplished without engagement or contact with formal prevention or treatment. Natural change, self-directed change, and self-change appear to use the same process of change (DiClemente, 2006). Often this was called “spontaneous remission.” However, a closer look indicated that many of the smokers who quit on their own without treatment engage in processes of change and traverse stages on their journey (Prochaska & DiClemente, 1983). Even when treatment did not predict drinking outcomes, process of change variables at the end of treatment discriminated posttreatment drinking outcomes (Carbonari & DiClemente, 2000). Thus, the TTM perspective is that all intentional behavior change is essentially self-change that includes accomplishment of stage tasks and use of processes. Prevention and treatment should always be viewed in the service of promoting client selfchange. Context of change is an additional dimension of the model that recognizes that change always happens within the life context of the individual. What is happening in the life context of the individual is important to consider, particularly when these factors may complicate or confound successful change of the target behavior. Other problems, risks, or resources in this life context influence positively or negatively completion of stage tasks and engagement in experiential and behavioral change processes. Contextual factors can enhance or interfere with change. The areas that have been outlined in the TTM include life situations and symptoms, belief systems, dyadic interpersonal relationships, systems issues (family, work, and social network), and more

sustained characterological or identity issues. For example, financial and marital problems often interfere with motivation and ability to accomplish stage tasks for individuals with a severe alcohol use disorder. Crises in one or more of these areas often disrupt decision-making and promote procrastination and ambivalence. On the other hand, a supportive spouse and family or an understanding and cooperative employer can facilitate an individual’s journey of change. Markers of change have been included in the model to measure stage task completion and serve as important indicators of progress. Decisional considerations (e.g., an important value that is being compromised, a serious consequence that promotes consideration of change) and the balance between pros and cons of change (e.g., considerations about the pros and cons of the current status quo and the pros and cons of making a change) were included because of their clear role in contemplation tasks and managing ambivalence. The conceptual perspective on decision-making came from Janis and Mann’s (1977) work on decision-making (see also Chapter 4, this volume). Shifts in these decisional considerations have often tracked or predicted movement through the stages of change (Prochaska, DiClemente, & Velicer, 1985). Across twelve different health behaviors the ratio of pros for change compared to the cons predicted stage status and differentiated those in precontemplation from those in contemplation and preparation (Prochaska, Velicer et al., 1994). The other marker came from research using Bandura’s social learning perspective. Self-efficacy expectations differed from outcome expectations and could influence planning and effort as well as successful and sustained action (Bandura, 1997; DiClemente et al., 1994; see also Chapter 3, this volume). Measures of these markers in studies of the model indicated that they were related to stage status and successful action and maintenance (Prochaska, DiClemente, & Velicer; 1985; Prochaska, Velicer et al., 1994). Smoking

Changing Behavior Using the Transtheoretical Model

abstinence self-efficacy increased during the preparation and action stages and peaked in maintenance as one would expect from Bandura’s assertion that past behavior is one of the key influences on efficacy. In another study, the end of alcohol treatment abstinence self-efficacy and its relation to temptation to drink across multiple situations has been a solid predictor of relapse (Shaw & DiClemente, 2016).

10.3 How Has the Theory Been Used to Change Behavior? As mentioned in Section 10.2, all intentional behavior change is essentially self-change. Although systems of psychotherapy and research on treatment have focused primarily on therapy techniques or therapist relationships, the critical mechanisms of change are under the control of the client. Therefore, the perspective of the TTM is that treatment should be tailored to the client processes of change and functions as a moderator for the client change process (see Figure 10.1). This is the only way to explain the confusing results of many of the research studies that compare philosophically and technically different types of treatment and find essentially equivalent

141

results. However, when different types of treatment do not differentially predict outcomes, client process activity, motivation, and confidence are related to outcomes (Carbonari & DiClemente, 2000). If all change is essentially self-change, what is the role of the therapist? The key to creating a TTM-designed treatment involves the therapist connecting treatment strategies to client tasks and processes (Brogan, Prochaska, & Prochaska, 1999). What the therapist does is create experiences, offer strategies, and design learning opportunities that engage client processes of change and promote the journey through the stages of change. Treatment activities should be focused on engaging appropriate processes of change. Velasquez and colleagues (2015) have created a group treatment program that exemplifies this perspective. Each group session is designed explicitly to engage specific processes of change. For example, a session on “Expressions of Concern” is an early stage activity designed to promote environmental reevaluation and shift the decisional balance of participants. A session for individuals in action on “Managing Cravings and Urges” targets the stimulus control process and supports successful

Figure 10.1 Understanding treatment as promoting and assisting self-change

142

CARLO C. DICLEMENTE AND MEAGAN M. GRAYDON

implementation of the action plan. The activities are not very different from other group exercises. It is the intentional targeting of client processes and stage tasks that makes it a TTM-designed intervention (Miller, 1999; Prochaska, Norcross, & DiClemente, 1994). From this perspective, treatment effects should be measured as to whether treatment or the activities of the therapist have activated appropriate client processes and promoted completion of stage tasks toward successful, sustained behavior change. Figure 10.1 illustrates how treatment moderates client coping activities leading to behavior change outcomes. In this figure, the client’s change process is the driver and the therapy provides the assistance that may be needed to help the client along their journey of change. There is an essential difference between the client activities and tasks and the therapist strategies (see Table 10.1). This leads to a clientcentered approach. Therapist activities need to be subordinate, yet supportive, to the client. There is an important distinction between a “treatment” plan and a “change” plan. The first is what the therapist offers; the second is what the client needs to do to make a change happen. The treatment plan should always be in the service of the change plan, though this is often not the case. Simple attendance at treatment or adhering to the treatment plan will not by itself produce change (DiClemente, 2018). In fact, many individuals make the desired behavior changes on their own, even after they drop out of treatment (DiClemente, 2006).

10.3.1 Using the TTM to Inform Treatment If the client is responsible for change activities and controls the mechanisms of change, it is critical for the therapist to have a broad range of techniques and an eclectic perspective. The TTM “borrows” from many different therapy systems to identify critical processes of change

(Prochaska & Norcross, 2016; see Table 10.3 for some suggested therapist activities that can promote different processes of change). For clients in early stages, therapists should have strategies to increase knowledge of a problem and potential solutions and offer a compelling rationale and cognitive strategies that resonate with the client, promoting self and environmental reevaluation. Therapists need ways to engage emotional arousal and stimulate social liberation. There are many motivational interviewing, cognitive therapy, and gestalt strategies very compatible with promoting these processes (Prochaska & Norcross, 2016). When clients move forward into later stages, the therapist needs strategies that promote selfliberation, which involves taking responsibility for the change and committing to taking action, as well as some strategies from both behavior therapy and behavior modification to help the client create effective plans and sustainable solutions. Therapists need a broad repertoire of strategies and approaches to be able to tailor their techniques and strategies to client stages and processes of change. One way for therapists to assess their ability to apply the TTM to their work with clients is to ask themselves what strategies and approaches could be used to address the following client stage tasks as illustrated in Table 10.4.

10.3.2 Using the TTM to Address Change in Different Settings and with Varied Problems The TTM provides a framework that offers a unique perspective for applying appropriate strategies and creating the most effective approach for addressing the client change process. In efforts to address motivation and behavior change, many treatment systems have incorporated dimensions of the model into assessment and treatment protocols. For example, an initial visit for someone seeking professional help will often include assessment of stage status or

Changing Behavior Using the Transtheoretical Model

143

Table 10.3 Therapist activities to promote experiential and behavioral processes of change Process of Change Experiential Processes Consciousness Raising Emotional Arousal Self-Reevaluation Environmental Reevaluation Social Liberation Behavioral Processes Reinforcement Management Helping Relationship Counterconditioning Stimulus Control Self-Liberation

Strategies to Activate Feedback, offering information, bibliotherapy Grief work, gestalt exercises, evoking important consequences Values clarification, assessing how current behavior fits with view of self, view of future Conducting life inventory of who has been harmed, how behavior affects family and friends Providing and advocating for policies or opportunities that support change, e.g. smoke-free environment Building a reward schedule for small and big steps, finding positives in change, functional analysis Identifying or being a support person who can help with the change process for this behavior Teaching mindfulness or promoting alternative responses to cues, e.g., physical exercise instead of eating junk food Teach how to restrict access to stimuli and cues Emphasize choice and responsibility, help with developing commitment and change plan

Table 10.4 Therapist inventory of intervention strategies What do you do to help the client complete the following tasks? a) Increasing interest and concern b) Decision-making and ambivalence c) Commitment and planning d) Taking action and implementing the plan e) Sustaining change f) Managing relapse and recycling

readiness to change and attempt to tailor efforts to the client’s stage. Application of the American Society of Addiction Medicine criteria for treatment placement includes an assessment of both readiness to change the addictive behavior and readiness for the most appropriate treatment option (Mee-Lee, 2013). Treatments often teach the client the stages and allow them to self-assess

their stage status as part of their treatment (Connors et al., 2013). A unique aspect of the implementation of the TTM is that, since it is an approach that describes relevant concepts and dimensions of a process that is focused on the client, it can be used in settings of differing intensity and length. The model, and particularly the stages construct, has been applied in most screening and brief interventions (SBIRT) protocols (Substance Abuse and Mental Health Services Administration, 2013), allowing providers to assess readiness to craft messages and strategies tailored to stage status. Smoking cessation interventions using telephone quit lines and self-help materials most often assess stage and tailor messages (Prochaska et al., 1993). One important aspect of many of these interventions is that they offer feedback to the client on their readiness and use motivational communication approaches to promote stage movement and activate processes of change.

144

CARLO C. DICLEMENTE AND MEAGAN M. GRAYDON

In summary, the model is being used as a template for improving assessment, tailoring treatment strategies, developing appropriate motivational and behavioral approaches and messages, and guiding development of treatment programs. Thus, it is better to talk about TTM-informed treatments and strategies rather than a single uniform or manualized transtheoretical therapy. There are multiple examples that illustrate this approach, including guidebooks (e.g., Connors et al., 2013) and manuals (e.g., Velasquez et al., 2015). There is also a recently updated description of how these model constructs can promote understanding of addictions and guide interventions for initiation and recovery (DiClemente, 2018).

10.4 Evidence Base for Use of Theory to Change Behavior The TTM is one of the most widely studied models of behavior change in the field of health behavior. More than three decades of research on this model has supported its utility in effective behavior change across problem areas.

10.4.1 Model Predicts Clinical Outcomes Some of the strongest support for the utility of the TTM to inform treatment has been in the field of addiction research. Studies examining the stages of change and other TTM constructs have found that individuals entering treatment vary in their profiles of these variables (DiClemente & Hughes, 1990). While these profiles differ at the beginning of treatment, they are often unaffected by the type of treatment (DiClemente et al., 2001) and yet their initial and subsequent profile scores have significant impact on treatment outcomes. For Project MATCH, there were robust findings for the role of initial client motivation on treatment outcomes (DiClemente et al., 2001, 2002). Baseline readiness to change was a strong predictor of working alliance and processes of change

posttreatment and throughout follow-up. Most notably, motivational readiness measures were the strongest predictor of later drinking behavior during the three-year follow-up period (DiClemente et al., 2001). Additional research has found that those who demonstrate forward stage transition (i.e., from pre-action to action stages) show significantly greater improvements in drinking outcomes than those who did not transition (Heather, Hönekopp, & Smailes, 2009). Motivation and stage status have consistently emerged in the literature as important predictors of substance use treatment outcomes (e.g., DiClemente et al., 1991; Penberthy et al., 2011) and therapeutic alliance (Connors et al., 2000). Recent meta-analyses support these findings while also expanding the utility of examining readiness for change in predicting treatment outcomes for eating disorders and mood disorder symptoms (Krebs et al., 2018; Norcross, Krebs, & Prochaska, 2010). As discussed in Section 10.2, baseline characteristics of patients entering treatment appear to play a large role in treatment success. However, shifts in attitudes and behaviors over the course of treatment are also instrumental in clinical outcomes (see Chapters 2 and 31, this volume). There is an increase in the use of change processes, particularly behavioral processes, across the treatment episode that is associated with treatment success (Carbonari & DiClemente, 2000; Prochaska, Norcross et al., 1992). Finally, endorsement of confidence in abstaining that outweighs temptation to drink is associated with abstinence. The impact of self-efficacy and temptation has been consistently reported in the literature (e.g., Penberthy et al., 2011; Shaw & DiClemente, 2016) and appears to change as a function of stage status (DiClemente et al., 2001). There is also research that indicates stage and processes of change are related to early termination of treatment and successful completion of treatment (Brogan, Prochaska, & Prochaska, 1999). In fact, change processes used at the

Changing Behavior Using the Transtheoretical Model

beginning of treatment were the strongest predictor of treatment attendance (Prochaska, Norcross et al., 1992). Importantly, these TTM constructs appear to play a role across the change process, including engaging in treatment, initial effect of treatment, and long-term success.

10.4.2 Model as a Mediator or Moderator While there is undoubtedly a direct impact of the transtheoretical constructs on treatment outcomes, there is also support for more indirect routes of influence. One important finding that has emerged is the relationship between motivation and therapeutic alliance in predicting treatment success. The impact of client motivation on alcohol use may be moderated by the therapist’s impression of their working alliance (Ilgen, McKellar et al., 2006). Therefore, a strong therapeutic alliance may be particularly important in achieving positive treatment outcomes for patients with low motivation, independent of their compliance with treatment. Similarly, client self-efficacy to change their behavior has been established as both a mediator (Hartzler et al., 2011) and a moderator (Ilgen, Tiet et al., 2006) of therapists’ perception of the working alliance and clinical outcomes. Additional research suggests that a strong therapeutic relationship may be particularly important for achieving treatment success among individuals with low self-efficacy at baseline (Ilgen, Tiet et al., 2006). In fact, patients who endorsed low self-efficacy but developed a strong therapeutic alliance had better one-year drinking outcomes compared with patients with low self-efficacy and poor working alliance and patients with high self-efficacy alone.

10.4.3 Stage-Based Interventions Many commonly used psychotherapies are actionoriented treatments. Yet most people in the behavior change process are not in the action stage.

145

Therefore, using a stage paradigm to inform treatment approaches can improve treatment outcomes. There have been numerous studies that have implemented interventions from a transtheoretical perspective that match treatment approaches to client stage of change. Studies have supported the use of these individualized treatment approaches that match based on stage to improve treatment outcomes (e.g., DiClemente et al., 2001; Dijkstra, Conijn, & De Vries, 2006; Prochaska et al., 1993). Stage-matched interventions have been found to be effective in making behavior changes and preventing relapse for various health behaviors, including smoking, healthy eating, cancer screening, and chronic disease management (Prochaska et al., 2005). A meta-analysis of eighty-eight studies on computer-tailored interventions found that tailored interventions were more effective that nontailored treatment (Krebs, Prochaska, & Rossi, 2010). Additional benefit was found when dynamically tailoring treatment over time rather than based on a single, initial assessment. There have also been studies that have failed to identify improvements when using stage-matched interventions (e.g., Griffin-Blake & DeJoy, 2006; Quinlan & McCaul, 2000) and there are critics of the model. These critiques center around whether the stages are discrete and sequential or continuous, adequacy of measures of stages, and whether there is adequate support for stage transitions and process use over time (Sutton, 2001; West, 2005). Although there have been additional studies (Heather et al., 2009; Krebs et al., 2018) and commentaries to counter these criticisms, continued research and additional meta-analyses are needed. Moreover, studies need to use standardized measures and approaches to allow for comparison across studies. The TTM constructs can be measured and employed in different ways and it is often difficult to determine how interventions are modified to match stages of change and whether they have included use of processes of change to guide the interventions.

146

CARLO C. DICLEMENTE AND MEAGAN M. GRAYDON

10.5 Summary and Conclusions Intentional behavior change represents a multidimensional set of tasks driven by experiential and behavioral activities labeled stages and processes of change. It is the individual’s change journey that is the focus of the TTM. The stages and processes offer a descriptive framework for understanding how people change. This perspective empowers individuals and providers trying to promote this change process to avoid negative and stigmatizing labels for what are normal and expected elements of the personal change process. The ambivalence of decision-making is a common experience and should lead to developing ways to overcome it rather than considering it procrastination (see also Chapter 45, this volume). Understanding relapse as the failure of learning and inadequate completion of stage tasks is far better than describing relapse as a personal or program failure. The TTM is not simply a clinical model that addresses psychopathology. The process of intentional behavior change is a universal human phenomenon. Thus, the model can be used to promote self-change. The TTM can also be used by public health professionals in their efforts to prevent initiation or promote modification of a problematic behavior and to foster initiation of health protective behaviors. In medical settings, the model has been used to promote screening and brief motivational interventions. Moreover, intervention efforts should be tailored to the client’s goals and change process and should follow the different parts of the journey. Treatment should not simply be more of the same approach but focus on the specific processes needed to complete tasks of the various stages. The TTM is essentially eclectic, meaning that there is no single set of strategies that can be manualized to promote change. There are many roads that lead to sustained behavior change as is demonstrated by the head-to-head comparisons among different systems of treatment. However,

if the strategies and approaches do not activate the appropriate processes for a particular client and help the client to complete change tasks, interventions will not promote change. The important message of the model is for clinicians to stop looking solely at themselves and what they do in interventions and instead focus on what their clients are doing. The model advocates increased focus on clients’ process of change and how to find creative ways to prevent or promote that process. How, then, can professionals learn how to use the model in their work? Here are some suggestions based on training doctoral-level clinicians. Professionals should learn the basic principles of the model from the literature and their personal experiences of moving through this change process. They should understand that there are different types of client mechanisms that need to be activated in order to build a multidimensional set of strategies that can address both pre-action and action stage tasks. Motivational communication strategies seem well suited to early stage tasks. It is critical for the professional to know ways to promote decision-making, commitment and planning techniques to sustain successful change, and strategies to promote maintained change and prevent relapse. There are many different types of therapy or treatments that contain one or more of these elements. The professional should make sure the focus is on the process, not simply the person or the problem, and become collaborators in the client’s journey of change. The TTM model and its application should also shift the focus of behavior change research from outcome research to mechanism research. The recent focus on mechanisms of change is very much in line with the focus of the model – just looking at the end of the client’s behavior change journey will not be informative about what it took to arrive at the destination. Therefore, research needs to explore mechanisms of change in the client instead of focusing solely on the therapist, their techniques, or their relationship with the client.

Changing Behavior Using the Transtheoretical Model

Instead of simply staying with the “siloed” approach of trying to develop treatments for each different mental and emotional problem or isolated disease, researchers and practitioners should be looking for opportunities to create a synergistic approach that facilitates the common elements needed to address recovery across disorders while continuing to deepen understanding of the unique aspects needed to address each. It was always the hope that the TTM could offer an eclectic and synergistic perspective that could cross boundaries of therapy systems and facilitate the behavior changes needed to live more productive, healthy, and meaningful lives.

References Bandura, A. (1997). Self-Efficacy: The Exercise of Control. New York: Freeman. https://doi.org/ 10.1891/0889-8391.13.2.158 Brogan, M. M., Prochaska, J. O., & Prochaska, J. M. (1999). Predicting termination and continuation status in psychotherapy by using the transtheoretical model. Psychotherapy, 36, 105– 113. https://doi.org/10.1037/h0087773 Carbonari, J. P., & DiClemente, C. C. (2000). Using transtheoretical model profiles to differentiate levels of alcohol abstinence success. Journal of Consulting and Clinical Psychology, 68, 810–817. https://doi.org/10.1037/0022-006X.68.5.810 Connors, G. J., DiClemente, C. C., Dermen, K. H., Kadden, R., Carroll, K. M., & Frone, M. R. (2000). Predicting the therapeutic alliance in alcoholism treatment. Journal of Studies on Alcohol, 61, 139–149. https://doi.org/10.15288/ jsa.2000.61.139 Connors, G. J., DiClemente, C. C., Velasquez, M., & Donovan, D. (2013). Substance Abuse Treatment and the Stages of Change (2nd ed.). New York: Guilford Press. DiClemente, C. C. (1978). Perceived processes of change on the cessation of smoking and the maintenance of that change. Unpublished PhD dissertation, University of Rhode Island. DiClemente, C. C. (2006). Natural change and the troublesome use of substances. In W. R. Miller &

147

K. M. Carroll (Eds.), Rethinking Substance Abuse: What the Science Shows and What We Should Do About It (pp. 81–96). New York: Guilford Press. DiClemente, C. C. (2018). Addiction and Change: How Addictions Develop and Addicted People Recover. New York: Guilford Press. DiClemente, C. C., Carbonari, J. P., Montgomery, R. P., & Hughes, S. O. (1994). The alcohol abstinence self-efficacy scale. Journal of Studies on Alcohol, 55, 141–148. https://doi.org/10.15288/ jsa.1994.55.141 DiClemente, C. C., Carbonari, J., Zweben, A., Morrel, T., & Lee, R. E. (2001). Motivation hypothesis causal chain analysis. In R. Longabaugh & P. W. Wirtz (Eds.), Project MATCH : A Priori Matching hypotheses, Results, and Mediating Mechanisms (National Institute on Alcohol Abuse and Alcoholism Project MATCH Monograph Series) Vol. 8 (pp. 206–222). Rockville, MD: National Institute on Alcohol Abuse and Alcoholism. DiClemente, C. C., Carroll, K. M., Miller, W. R., Connors, G. J., & Donovan, D. M. (2002). A look inside treatment: Therapist effects, the therapeutic alliance, and the process of intentional behavior change. In T. F. Babor & F. K. DelBoca (Eds.), Treatment Matching in Alcoholism (pp. 166–183). London: Cambridge University Press. DiClemente, C. C., Delahanty, J. C., Havas, S. W., & Van Orden, O. R. (2015). Understanding selfreported staging of dietary behavior in lowincome women. Journal of Health Psychology, 20, 741–753. https://doi.org/10.1177/ 1359105315580213 DiClemente, C. C., & Hughes, S. O. (1990). Stages of change profiles in outpatient alcoholism treatment. Journal of Substance Abuse, 2, 217– 235. https://doi.org/10.1016/S0899-3289(05) 80057-4 DiClemente, C. C., Prochaska, J. O., Fairhurst, S. K., Velicer, W. F., Velasquez, M. M., & Rossi, J. S. (1991). The process of smoking cessation: An analysis of precontemplation, contemplation, and preparation stages of change. Journal of Consulting and Clinical Psychology, 59, 295–304. https://doi.org/10.1037/0022-006X.59.2.295 Dijkstra, A., Conijn, B., & De Vries, H. (2006). A match–mismatch test of a stage model of

148

CARLO C. DICLEMENTE AND MEAGAN M. GRAYDON

behaviour change in tobacco smoking. Addiction, 101, 1035–1043. https://doi.org/10.1111/j.13600443.2006.01419.x Goldfried, M. (1980). Toward the delineation of therapeutic change principles. American Psychologist, 35, 931–950. https://doi.org/ 10.1037/0003-066X.35.11.991 Griffin-Blake, C. S., & DeJoy, D. M. (2006). Evaluation of social-cognitive versus stagematched, self-help physical activity interventions at the workplace. American Journal of Health Promotion, 20, 200–209. https://doi.org/10.4278/ 0890-1171-20.3.200 Hartzler, B., Witkiewitz, K., Villarroel, N., & Donovan, D. (2011). Self-efficacy change as a mediator of associations between therapeutic bond and one-year outcomes in treatments for alcohol dependence. Psychology of Addictive Behaviors, 25, 269. https://doi.org/10.1037/ a0022869 Heather, N., Hönekopp, J., & Smailes, D. (2009). Progressive stage transition does mean getting better: A further test of the transtheoretical model in recovery from alcohol problems. Addiction, 104, 949–958. https://doi.org/10.1111/j.13600443.2009.02578.x Ilgen, M. A., McKellar, J., Moos, R., & Finney, J. W. (2006). Therapeutic alliance and the relationship between motivation and treatment outcomes in patients with alcohol use disorder. Journal of Substance Abuse Treatment, 31, 157–162. https:// doi.org/10.1016/j.jsat.2006.04.001 Ilgen, M., Tiet, Q., Finney, J., & Moos, R. H. (2006). Self-efficacy, therapeutic alliance, and alcohol-use disorder treatment outcomes. Journal of Studies on Alcohol, 67, 465–472. https://doi.org/ 10.15288/jsa.2006.67.465 Janis, I. L., & Mann, L. (1977). Decision Making. New York: Free Press. Krebs, P., Norcross, J. C., Nicholson, J. M., & Prochaska, J. O. (2018). Stages of change and psychotherapy outcomes: A review and metaanalysis. Journal of Clinical Psychology, 74, 1964–1979. https://doi.org/10.1002/jclp.22683 Marcus, B. H., Rossi, J. S., Selby, V. C., Niaura, R. S., & Abrams, D. B. (1992). The stages and processes of exercise adoption and maintenance in a

worksite sample. Health Psychology, 11, 386–395. https://doi.org/10.1037//0278-6133.11.6.386 Mee-Lee, D. (Ed.). (2013). The ASAM Criteria: Treatment Criteria for Addictive, SubstanceRelated, and Co-occurring Conditions. Bethesda, MD: American Society of Addiction Medicine. Miller, W. R. (1999). Enhancing Motivation to Change in Substance Abuse Treatment. Treatment Improvement Protocol Series, Vol. 35. Rockville, MD: US Department of Health and Human Services. Norcross, J. C., Krebs, P. M., & Prochaska, J. O. (2010). Stages of change. Journal of Clinical Psychology: In Session, 67, 143–154. https://doi.org/10.1002/ jclp.20758 Penberthy, J. K., Hook, J. N., Vaughan, M. D. et al. (2011). Impact of motivational changes on drinking outcomes in pharmacobehavioral treatment for alcohol dependence. Alcoholism: Clinical and Experimental Research, 35, 1694– 1704. https://doi.org/10.1111/j.15300277.2011.01516.x Prochaska, J. O. (1979). Systems of Psychotherapy: A Transtheoretical Analysis. Homewood, IL: Dorsey Press. Prochaska, J. O., & DiClemente, C. C. (1982). Transtheoretical therapy: Toward a more integrative model of change. Psychotherapy: Theory, Research and Practice, 19, 276–288. https://doi.org/10.1037/h0088437 Prochaska, J. O., & DiClemente, C. C. (1983). Stages and processes of self-change of smoking: Toward an integrative model of change. Journal of Consulting and Clinical Psychology, 51, 390–395. https://doi.org/10.1037/0022-006X.51.3.390 Prochaska, J. O., & DiClemente, C. C. (1984). The Transtheoretical Approach: Crossing the Traditional Boundaries of Therapy. Malabar, FL: Krieger. Prochaska, J. O., & DiClemente, C. C. (1986). Toward a comprehensive model of change. In W. R. Miller & N. Heather (Eds.), Treating Addictive Behaviors (pp. 3–27). Boston: Springer. Prochaska, J. O., DiClemente, C. C., & Norcross, J. C. (1992). In search of how people change: Applications to the addictive behaviors. American Psychologist, 47, 1102–1114. https://doi.org/ 10.1037/0003-066X.47.9.1102

Changing Behavior Using the Transtheoretical Model

Prochaska, J. O., DiClemente, C. C., & Velicer, W. F. (1985). Predicting change in smoking status for self-changers. Addictive Behaviors, 10, 395–406. https://doi.org/10.1016/0306-4603(85)90036-X Prochaska, J. O., DiClemente, C. C., Velicer, W. F., & Rossi, J. S. (1993). Standardized, individualized, interactive and personalized self-help programs for smoking cessation. Health Psychology, 12, 399–405. https://doi.org/10.1037//02786133.12.5.399 Prochaska, J. O. & Norcross, J. (2016). Systems of Psychotherapy: A transtheoretical analysis (8th ed.). New York: Brooks Cole. Prochaska, J. O., Norcross, J. C., & DiClemente, C. C. (1994). Changing for Good. New York: Morrow. Prochaska, J. O., Norcross, J. C., Fowler, J. L., Follick, M. J., & Abrams, D. B. (1992). Attendance and outcome in a work site weight control program: Processes and stages of change as process and predictor variables. Addictive Behaviors, 17, 35–45. https://doi.org/10.1016/0306-4603 (92)90051-V Prochaska, J. O., Velicer, W. F., DiClemente, C. C., & Fava, J. L. (1988). Measuring the processes of change: Applications to the cessation of smoking. Journal of Consulting and Clinical Psychology, 56, 520–528. https://doi.org/0022-006 X/88/ S00.75 Prochaska, J. O., Velicer, W. F., Redding, C. et al. (2005). Stage-based expert systems to guide a population of primary care patients to quit smoking, eat healthier, prevent skin cancer, and receive regular mammograms. Preventive Medicine, 41, 406–416. https://doi.org/10.1016/j .ypmed.2004.09.050 Prochaska, J. O., Velicer, W. F., Rossi, J. S. et al. (1994). Stages of change and decisional balance for 12 problem behaviors. Health Psychology, 13, 39– 46. https://doi.org/10.1037/0278-6133.13.1.39 Project MATCH Research Group. (1997). Matching alcoholism treatments to client heterogeneity:

149

Project MATCH post-treatment drinking outcomes. Journal of Studies on Alcohol, 58, 7–29. https://doi.org/10.15288/jsa.1997.58.7 Quinlan, K. B., & McCaul, K. D. (2000). Matched and mismatched interventions with young adult smokers: Testing a stage theory. Health Psychology, 19, 165–171. https://doi.org/10.1037/ 0278-6133.19.2.165 Rosen, C. S. (2000). Is the sequencing of change processes by stage consistent across health problems? A meta-analysis. Health Psychology, 19, 593–604. https://doi.org/ 10.1037/02786133.19.6.593 Shaw, M. A. & DiClemente, C. C. (2016). Relapse vulnerability measure of the alcohol abstinence self-efficacy scale predicting time to first drink and amount of drinking. Journal of Studies on Alcohol and Drugs. 77, 521–525. https://doi.org/ 10.15288/jsad.2016.77.521 Sloane, R. B., Staples, F. R., Cristol, A. H., Yorkston, N. J., & Whipple, K. (1975). Psychotherapy Versus Behavior Therapy. Cambridge, MA: Harvard University Press. Substance Abuse and Mental Health Services Administration (2013). Systems-Level Implementation of Screening, Brief Intervention, and Referral to Treatment. Technical Assistance Publication (TAP) Series 33, HHS Publication No. (SMA) 13–4741. Rockville, MD: USDHHS Center for Substance Abuse Treatment. Sutton, S. (2001). Back to the drawing board? A review of applications of the transtheoretical model to substance use. Addiction, 96, 175–186. https://doi .org/10.1080/09652140020017049 Velasquez, M., Crouch, C., Stephens, N., & DiClemente, C. C. (2015). Group Treatment for Substance Abuse: Stages of Change Therapy Manual (2nd ed.). New York: Guilford Press. West, R. (2005). Time for a change: Putting the transtheoretical (stages of change) model to rest. Addiction, 100, 1036–1039.

11 Changing Behavior Using Integrative Self-Control Theory Wilhelm Hofmann, Simone Dohle, and Kathi Diel

Practical Summary An integrative self-control theory is used to describe how many behavioral interventions in the self-control domain can be organized around the following conceptual hubs: desire regulation, boosting goal commitment, conflict identification, self-control motivation, self-control ability (training), and behavior enactment barriers. Links between the proposed theory and techniques used to change behavior are made based on survey of the literature on behavioral interventions of self-control. The theory is also situated within other frameworks, such as Michie et al.’s (2014) commitment-opportunity-motivation behavior model. Taken together, theory and evidence indicate that a “one-size-fits-all” approach to combating self-control problems may be vastly inferior to an approach that specifically targets the “weak spots” in the self-control system in individuals, problem populations, or given applied contexts. For these reasons, those interventions that build on an attempt to diagnose these weak spots first and those interventions that offer a combination of most appropriate treatments are likely to be most effective.

11.1 What Is Self-Control? To eat a tempting dessert or rather stick to a diet of carrots and celery sticks? To keep investing effort toward a work-related deadline or go for a “quick surf” on Facebook? To save money toward one’s long-term goals or go on a rewarding shopping splurge? Everyday life is full of difficult but consequential choices such as these. Because the repeated enactment of the tempting alternative can have substantial long-term costs (e.g., health problems, financial problems, punishment, etc.), understanding the processes affecting decisions and behavior in multi-motive settings has enormous implications for individual and societal well-being (Mischel, Shoda, & Peake, 1988;

Moffitt et al., 2011). Some public health figures suggest that 40 percent of deaths every year are associated with behaviors that are at least partially attributable to the way people control problematic temptations such as those for unhealthy foods, tobacco, alcohol, unprotected sex, aggressive urges, and illicit drugs (Schroeder, 2007). One only needs to look at the rising rates of obesity, for instance, to realize that the promotion of successful self-control through instruction, training, and environmental design is among the most beneficial applied downstream consequences that can stem from a deeper psychological understanding of self-control success and failure. https://doi.org/10.1017/9781108677318.011

Changing Behavior Using Integrative Self-Control Theory

151

Sidebar 11.1 Origins of self-control research

Experimental research on self-control began more than four decades ago, with Walter Mischel and colleagues’ classic study of delay of gratification in children (e.g., Mischel, Ebbesen, & Zeiss, 1972). In this famous “marshmallow test,” children had to choose between a small, immediate reward and a larger but delayed reward for which they had to wait approximately fifteen minutes. Subsequent longitudinal research found over-time correlates of children’s ability to delay gratification, measured at ages four or five, with important subsequent life outcomes such as academic achievement and social competence (Mischel et al., 1988).

Traditional self-control research and associated public health research have approached the issue of successful self-control primarily from the vantage point of a deficit in knowledge and/or ability. Regarding knowledge, traditional approaches aim to better educate people about the negative long-term consequences and risks of certain behaviors via educational campaigns (e.g., Ajzen & Albarracín, 2007; Godin & Kok, 1996; Janz & Becker, 1984; Strecher et al., 1995; Johnson, Wolf, & Maio, 2017; see Sidebar 11.1 on the origins self-control research). Regarding ability, self-control has often been defined as “the ability to override or change one’s inner responses, as well as to interrupt undesired behavioral tendencies (such as impulses) and refrain from acting on them” (Tangney, Baumeister, & Boone, 2004, p. 247). Accordingly, the concept most in the focus of traditional self-control research is the idea of inhibition, along with a focus on the dispositional and situational risk factors that diminish people’s ability to control themselves (Baumeister et al., 1998; Fujita & Han, 2009; Hofmann, Schmeichel, & Baddeley, 2012). Even though both of these foci have provided numerous valuable insights and applications for behavior change, modern approaches take a much broader, integrative perspective (e.g., Kotabe & Hofmann, 2015). Such integrative models of selfcontrol attempt to better combine knowledge and cognitive ability factors with a wider range of

internal and external elements of self-control. Most centrally, there is a notable shift toward motivational factors of self-control failure such as commitment devices and balancing decisions (De Witt Huberts, Evers, & de Ridder, 2014; Inzlicht, Schmeichel, & Macrae, 2014; Kotabe & Hofmann, 2015), a closer consideration of the costs and benefits of various self-control strategies (Duckworth, Gendler, & Gross, 2016; Fishbach & Hofmann, 2015; Hofmann & Kotabe, 2012), as well as a closer consideration of how the individual decision-maker is embedded in a choice context in space and time (Sobal & Wansink, 2007; Thaler & Sunstein, 2009; Thaler, Sunstein, & Balz, 2010). The present chapter aims to address the issue of behavior change from such an integrative perspective of self-control, specifically the recently developed integrative self-control theory (Kotabe & Hofmann, 2015). In the next section, the theory is introduced, outlining the various conceptual roots on which it draws. Then, the utility of the theory to identify and classify various behavior change techniques in the self-control domain is outlined. Finally, differences and similarities of the theory with a broader approach to behavior change that is not specific to self-control, the commitment-opportunity-motivation behavior (COM-B) model together with the behavior change wheel (Michie, Atkins, & West, 2014) are outlined, along with suggestions of open questions for future research.

152

WILHELM HOFMANN, SIMONE DOHLE, AND KATHI DIEL

11.2 A Brief Overview of Integrative Self-Control Theory This section provides a brief overview of integrative self-control theory (Kotabe & Hofmann, 2015). Integrative self-control theory posits that the behavioral outcome of a prototypical selfcontrol episode is determined by the interplay of seven core psychological components: (1) Desire: An “affectively-charged cognitive event” (Kavanagh, Andrade, & May, 2005) with a lower-order goal component that motivates appetitive behavior. (2) Self-control goal: A higher-order goal associated with mental representations of an endorsed end state. (3) Self-control conflict: A disharmonious state caused by the coactivation of a given desire

(4)

(5)

(6)

(7)

and an at least partially incompatible selfcontrol goal. With self-control conflict, a desire turns into a “temptation” that threatens self-control goal attainment. Self-control motivation: Reflects the degree to which a person aspires to control temptation. Self-control capacity: Refers to all the mental resources a person possesses to override temptation with a competing self-control goal. Self-control effort: Refers to the actual amount of self-control capacity used at a given point in time. Its upper limit is determined by self-control motivation and selfcontrol capacity. Enactment constraints: Anything in the environment that hinders the enactment of the desire or the self-control goal.

Sidebar 11.2 Desires and goals in integrative self-control theory

The term “desire” can be understood quite narrowly (in the sense of passionate sexual desire) as well as very broadly, including all sorts of ideals and wishes. In this chapter, it concerns appetitive desires – desires rooted in physiological need states (such as for food, alcohol, sex, rest) or acquired through a history of reinforcement learning as in the case of drugs, media addiction, spending urges, and so on. Desires propel individuals to approach certain stimuli in their environment and engage in activities with them that provide us with a relative gain in immediate pleasure (including relief from discomfort). At a phenomenological level, desire is experienced as a want for an object or activity that is usually associated with pleasure (Berridge, Robinson, & Aldridge, 2009). The goal concept encompasses all those responses that are intentionally performed to bring about a desired state and response and those intentionally performed to control or prevent an undesired state (Moskowitz & Grant, 2009). Even though the desire construct shares some overlap with the goal construct, as indicated, these terms refer to different “prototypes” in the landscape of human motivation (see also Hofmann & Nordgren, 2015). That is, the prototypical desire may be higher on aspects such as intrinsic reward and affective “hotness” when compared to the prototypical goal, whereas the prototypical goal may be experienced as being higher in intentionality, personal control, temporal scope, and meaning when compared to the prototypical desire. Moreover, the desire concept may be useful in preventing an overextension of the goal concept to those more extreme cases of impulsive behavior that are hard to reconcile with the idea of goals as intentional and meaning-providing responses (Moskowitz & Grant, 2009).

Changing Behavior Using Integrative Self-Control Theory

Figure 11.1 presents an illustration of how these seven components can be thought of as working together according to the theory (for a more in-depth and formalized treatment, see Kotabe & Hofmann, 2015). Integrative self-control theory posits that the coactivation of desire and an at least partially incompatible self-control goal induces self-control conflict, which activates self-control exertion via self-control motivation. Specifically, a desire (e.g., for relaxation) by itself is unproblematic unless the person endorses a higher-order goal (e.g., to meet a tight deadline) that is at least partially incompatible with the desire. In this case, the desire becomes a temptation and the higher-order goal becomes a selfcontrol goal. The extent of self-control conflict experienced is a function of the strength of the desire, the strength of the higher-order goal, and the degree to which they are incompatible (see Sidebar 11.2 on desires). Self-control motivation and self-control capacity jointly determine the upper limit of self-control effort a person is prepared to invest in pursuing the desired action or behavior at a given point in time. In other words, the resultant self-control effort is the net “force” (Lewin, 1951) that opposes the net force of the focal desire (see Figure 11.1). The prevailing force,

153

except when it is too weak to overcome enactment constraints, drives behavior. That is, if the motivational force of self-control effort prevails over the motivational force of desire, then selfcontrol will succeed – provided that self-control goal enactment constraints do not prevent selfcontrol goal enactment (see Figure 11.1). If desire prevails over self-control effort, then self-control will fail provided that desire enactment constraints do not prevent desire enactment. At a broader level, the seven components can be assigned to three parts of the self-control process: activation, exertion, and enactment constraints. The first three components, desire, selfcontrol goal, and self-control conflict, comprise the self-control activation part. Self-control conflict, the output of this activation cluster, is assumed to trigger the self-control exertion part, comprised of the following three components: self-control motivation, self-control capacity, and self-control effort – the latter of which is the output of the exertion cluster. Integrative selfcontrol theory posits that the fourth and fifth components – self-control motivation and selfcontrol capacity – are major determinants of self-control effort. Together, these three components form the exertion cluster. Higher selfcontrol motivation and self-control capacity

Sidebar 11.3 Enactment constraints in integrative self-control theory

Enactment constraints are environmental factors not under the person’s immediate control that constrain the range of available behavioral options in a given situation (see Kotabe & Hofmann, 2015). Such factors include finite resources such as time and money as well as physical and social barriers that can keep a person from enacting a given desire or goal even when there are no “inner constraints” operating. For example, a dieter might not be able to restrain the urge to eat a slice of cake at a wedding but, if a stranger were to take that last piece of wedding cake, the behavioral options for temptation enactment would be substantially constrained. From a choice architecture perspective (Thaler & Sunstein, 2009), introducing desireenactment constraints and reducing self-control goal–enactment constraints is an effective recipe for nudging people toward better self-control (see Chapter 14, this volume).

154

WILHELM HOFMANN, SIMONE DOHLE, AND KATHI DIEL

yield higher potential self-control effort – the maximum amount of self-control effort that could be spent toward combating temptation. In terms of basic mechanisms, integrative self-control theory posits that people invest actual self-control effort to match the challenge provided by desire strength but only up to this maximum (for further details, see Kotabe & Hofmann, 2015). The last part of the process concerns possible constraints (and opportunities) in people’s environments (enactment constraints), thus reflecting the often critical role of the physical and social environment. If self-control effort prevails over temptation, then self-control will succeed provided that self-control goal enactment constraints (e.g., availability of healthy alternatives) do not prevent success. If temptation prevails over selfcontrol effort, then self-control will fail provided that temptation enactment constraints (e.g., availability of tempting alternatives; social control) do not prevent failure (see Sidebar 11.3 on enactment constraints).

11.3 Links to Earlier Research and Theory The components of integrative self-control theory are informed by extensive, yet largely separate, bodies of research. With respect to desires, Kavanagh et al. (2005) theorized specifically about desire, drawing attention to questions about the nature of desire as a distinct construct, conditions that trigger and fuel desire, and ways desire impacts cognition and action. As for selfcontrol goals, goal systems theory (Kruglanski et al., 2002) has provided important theoretical predictions on the structural properties and functions of goal systems, which is pertinent to understanding how self-control goals activate and operate and when and where they might fail. Specifically, concerning self-control goal and desire violations, hierarchical cybernetic models propose hypotheses relating to the differential consequences of violating goals at various levels

of a theoretical goal hierarchy (Carver & Scheier, 1982, see Chapter 9, this volume). With regard to self-control conflict, the topic of monitoring for, and detecting, response conflicts has received considerable attention from cognitive neuroscientific perspectives (Botvinick et al., 2001). The interplay of self-control motivation and self-control capacity in determining effort allocation was inspired by cognitive energetics theory (Kruglanski et al., 2012) as well as motivational intensity theory (Brehm & Self, 1989). Finally, regarding enactment constraints, insights from research on choice architecture – the proactive designing of environments to facilitate better decision-making (e.g., Thaler & Sunstein, 2009; see also Chapter 14, this volume) – were integrated into the model.

11.4 Applications in Self-Control Research and Interventions Integrative self-control theory, or more compressed versions thereof, has been used as a guiding framework in carrying out or reviewing empirical research on self-control, both across behavior domains (Bernecker, Job, & Hofmann, 2018; Delaney & Lades, 2017; Hofmann et al., 2012; Krönke et al., 2018; Ozaki et al., 2017) and with a specific domain in focus such as eating behavior (Hofmann et al., 2014; Lopez et al., 2014), media use (Du, van Koningsbruggen, & Kerkhof, 2018; Hofmann et al., 2017), environmental behavior (Nielsen, 2017), and self-control in organizational/work settings (Lian et al., 2017). In terms of behavioral interventions, integrative self-control theory can be usefully applied as a conceptual map for first identifying and distinguishing among various “soft spots” of self-control failure and, consequently, tailored recommendations for targeted interventions (ideally in that order). For instance, if the primary “soft spot” in a given target sample is weak commitment to the self-control goal, traditional risk

Changing Behavior Using Integrative Self-Control Theory

155

Table 11.1 Overview of some intervention approaches mapped onto components of integrative self-control theory Component

Intervention Approach

Example References

Desire Strength

• Preventive approaches (situation and stimulus control; early distraction)

Mahoney & Thoresen (1972); Florsheim et al. (2008); Van Dillen, Papies, & Hofmann (2013) Houben, Havermans, & Wiers (2010); Houben, Schoenmakers, & Wiers (2010); Wiers et al. (2011); Attwood et al. (2008) Alberts et al. (2010); Forman et al. (2007)

• Shaping of desire (evaluative conditioning; avoidance training; attentional bias modification training) Effective down-regulation of desire (mindfulness/meditation training; acceptance-based therapy) Self-Control Goal Facilitating goal setting and boosting goal Strength commitment through risk education programs Self-Control Conflict Self-monitoring of unwanted behaviors Self-Control • Motivational interviewing Motivation • Increasing the stakes (e.g., penalties) • Increasing the rewards Self-Control Training of executive functions (working Capacity memory capacity; behavioral inhibition) Enactment • Choice architecture Constraints • Situation and stimulus control

education approaches may be indicated (see also Chapters 31 and 38, this volume). If overpowering desire is part of the problem, such as is often the case in addictions, then craving reduction or mindfulness-based techniques (e.g., Alberts et al., 2010), as well as enactment constraints that render desire-eliciting stimuli less visible and available (see also Chapter 14, this volume), may be more promising avenues. Table 11.1 provides examples of a wide spectrum of intervention approaches to improving self-control, which can easily be mapped onto the components of integrative self-control theory (see also Chapter 40, this volume). Regarding interventions targeting the activation cluster, there are desire-based interventions that fall into two broad classes of techniques: (1) those that prevent desire from occurring and (2) those that

Ajzen & Albarracín (2007); Godin & Kok (1996); Janz & Becker (1984); Strecher et al. (1995) McFall (1970); Yon et al. (2007) Miller & Rollnick (2012); Thush et al. (2009); Schelling (1984); Trope & Fishbach (2000) Milkman et al. (2013) Houben, Wiers, & Jansen (2011); Houben, Nederkoorn et al. (2011) Fishbach & Zhang (2008); Hanks et al. (2012); Thorndike et al. (2012); Mahoney & Thoresen (1972)

reduce desire strength after desire is operative. The first class comprises situation and stimulus control techniques (Mahoney & Thoresen, 1972), which can effectively eliminate or reduce the emergence of desire. The second class comprises relatively new techniques that either reduce the initial hedonic appeal of tempting stimuli (shaping approaches) – for example, through evaluative conditioning (Houben, Havermans, & Wiers, 2010; Houben, Schoenmakers, & Wiers, 2010; Van Gucht et al., 2010) or avoidance training (Wiers et al., 2011) – or approaches that teach people how to down- regulate desires and cravings (desire-regulation approaches) – for example, through distraction (Florsheim et al., 2008; Metcalfe & Mischel, 1999; Van Dillen, Papies, & Hofmann, 2013), mindfulness, or acceptancebased interventions (Alberts et al., 2010;

156

WILHELM HOFMANN, SIMONE DOHLE, AND KATHI DIEL

Forman et al., 2007). As for interventions targeting the self-control goal, many examples come from health psychology. These approaches aim to get people to set goal intentions and boost commitment to goals, for example via educational campaigns that inform about health-behavior links and potential risks and long-term consequences of impulsive behavior (Ajzen & Albarracín, 2007; Godin & Kok, 1996; Janz & Becker, 1984; Strecher et al., 1995; see also Chapters 2, 3, and 31, this volume). Regarding self-control conflict, beyond targeting self-monitoring (e.g., Webb, Miles, & Sheeran, 2012; see also Chapter 37, this volume), interventions generally aimed at increasing mindfulness (for a review, see Baer, 2003) of inner states may be useful in that they may help people identify selfcontrol conflict and respond to it effectively. Regarding the exertion cluster, one prominent approach targets self-control motivation through motivational interviewing. This technique involves various counseling tactics that elicit self-motivational statements (see Hardcastle et al., 2017; Miller & Rollnick, 2012; Thush et al., 2009; see also Chapter 45, this volume), or those techniques that increase the stakes (e.g., penalties) associated with temptation enactment (e.g., Trope & Fishbach, 2000), or the rewards associated with self-control goal enactment (e.g., Milkman, Minson, & Volpp, 2013; see Chapter 36, this volume). More recently, self-control capacity has become a promising intervention focus (see also Chapter 40, this volume). This trend has been inspired by experimental research in cognitive psychology, suggesting that executive functions can be trained (e.g., see Jaeggi et al., 2008; Olesen, Westerberg, & Klingberg, 2003; but, for a critical perspective, see Shipstead, Redick, & Engle, 2012). For example, researchers have developed programs that aim to increase healthy behaviors by boosting executive functions such as response inhibition (for a metaanalysis, see Allom, Mullan, & Hagger, 2016; Houben, Nederkoorn et al., 2011) and working

memory capacity (Houben, Wiers, & Jansen, 2011). Finally, interventions can also target enactment constraints by taking advantage of the principles of nudging. Nudging is all about proactively altering environments or choice architecture in ways that facilitate better/smarter decisions (Thaler & Sunstein, 2009; Thaler, Sunstein, & Balz, 2010; see also Chapter 14, this volume). An example of a nudge intervention in the domain of pro-environmental behavior comes from Pichert and Katsikopoulos (2008), who argued that the default choice option for electricity provider selection in countries such as Germany is so-called gray (i.e., polluting, nonrenewable) electricity whereas “green” electricity is (more or less implicitly) the alternative, nondefault choice option. Across natural and laboratory settings, they could show that simply changing the default option from gray to green electricity has a huge impact on people’s behavioral choice for green electricity (for similar results in the domain of organ donation, see Johnson & Goldstein, 2003). Moreover, classical techniques of stimulus and situation control (e.g., Mahoney & Thoresen, 1972) would also fall under the category of enactment constraints to the extent that they increase desire enactment constraints and/or decrease self-control goal enactment constraints.

11.5 Recommendations for Intervention from the Perspective of Integrative Self-Control Theory One central insight from integrative self-control theory is that self-control is not a unitary phenomenon or single essence or resource but rather an interplay of several components (see also Chapter 40, this volume). The multicomponent nature of self-control featured here implies that there are many routes to poor self-control and, consequently, multiple “soft spots” for behavior change

Changing Behavior Using Integrative Self-Control Theory

intervention. Therefore, a “one-size-fits-all” approach to combating self-control failures is unlikely to be the most effective route of intervention. Rather, which treatment works and which does not is likely to differ across individuals, problem populations, and contexts, depending on where the “weak spots” in the self-control system are located. For these reasons, those interventions that build on an attempt to diagnose these weak spots first through appropriate formative research in the target population and for the behavior of interest and those interventions that offer a combination of most appropriate treatments are likely to be more effective than those focused on one standard intervention. To date, however, there is little research that has compared the effectiveness of the typical “blanket solutions” versus customized, adaptive, and combined interventions.

11.5.1 Similarities and Differences with Existing Taxonomies Integrative self-control theory shares some overlap with frameworks of behavior change, such as the COM-B model by Michie and colleagues and their associated behavior change wheel, which interlinks the COM-B model with public policy interventions (Michie, Atkins, & West, 2014; Michie, Van Stralen, & West, 2011). First, like the COM-B model, integrative self-control theory distinguishes between motivational-based (i.e., desire, self-control goal commitment, selfcontrol motivation), capacity-based (self-control capacity), and opportunity-related (i.e., enactment constraints) sources of behavior. However, integrative self-control theory, due to its specific focus on the self-control process, adds further granularity by distinguishing between various subcomponents of motivation, including desire strength, goal commitment, and motivational conflict. In addition, integrative self-control theory deals with how these components may interact with each other as part of a psychological

157

process. Second, integrative self-control theory shares the general assumption underlying the COM-B model that different interventions can specifically target different components, including the link between education/persuasion and motivational aspects (such as goal commitment), the link between training and capacity, and the link between environmental restructuring and opportunity (enactment constraints), even though, again, the link between intervention strategies and target component is somewhat more fine-grained in integrative self-control theory due to its particular focus. Third, a major difference between integrative self-control theory and Michie and colleagues’ approach is that the latter outlines links between interventions and various public policy categories such as regulation, fiscal policies, legislation, environmental/social planning, or service provision, thus connecting the psychological determinants to the fields of implementation science and public policy making.

11.5.2 Emerging New Challenges Integrative self-control theory may be a useful conceptual background in thinking about largely unexplored and challenging issues for future research on self-control interventions. First, the interplay of external policy measures with internal components of self-control, such as shortterm desire strength, long-term goal commitment, conflict awareness, and self-control motivation, is poorly understood. How exactly do various public policy measures such as educational campaigns, incentives, and behavioral nudges that impose constraints on behavioral choices affect internal key components of self-control? Second, to what extent can people “extract a lesson” from behavioral nudges regarding selfcontrol that benefits their future use of selfcontrol strategies? More generally, to what extent can short-term behavioral changes instigated by nudges lead to real changes in self-concept that propagate enduring, sustainable behavior?

158

WILHELM HOFMANN, SIMONE DOHLE, AND KATHI DIEL

Third, how may emerging technologies (such as via continuous self-assessment of problematic situations/contexts or geospatial sensing thereof) aid in the development/ improvement of goal commitment, self-control motivation, and selfcontrol capacity? Inspired by the recent “quantified self” movement in the field (Swan, 2013), does the daily collection of healthy/unhealthy behaviors via experience sampling provide further insights into the user’s own behavioral patterns over time – a perspective most people typically may not take? Such basic feedback functionality could be further enhanced with additional components designed to boost motivation (e.g., temporal or social comparison information; rewards) and safeguards against self-licensing, including means to increase awareness of maladaptive patterns of reasoning that may act as self-licenses (De Witt Huberts et al., 2014; see also Chapter 29, this volume). Finally, taking a more long-term perspective, when behavioral outcomes of multiple contentrelated self-control episodes are arranged in temporal order, systematic regulatory patterns may emerge. On the macro level, these patterns may be classified along a spectrum ranging from one extreme of chronic self-control failure, to some ratio of “balancing” or “moderation” (e.g., a dieter resisting three out of four temptations on average), to the other extreme of chronic selfcontrol. Such a long-term perspective touches on what may be some of the most exciting, and challenging, questions at the frontier of selfcontrol research – for example, what sorts of emotional and cognitive “lessons” do people extract from their past self-control successes and failures and how do these influence future behavior (Baumeister et al., 2007; Dohle & Hofmann, 2019; Hofmann & Fisher, 2012)? According to which principles do people mentally account for their “vices” and “virtues” over time and within and across domains? Finally, is there a golden mean of balancing desire and self-control goal enactment in a given domain (e.g., eating,

drinking, saving, etc.) that maximizes people’s long-term well-being? Integrative self-control theory may aid understanding of such long-term dynamics and macrolevel effects in that it may stimulate thinking about how the key internal and external parameters of an individual self-control episode influence each other over time: First, the components of the model are likely to have lagged effects on each other. Second, people may utilize proactive self-control strategies to make subsequent selfcontrol easier (Fujita, 2011; Hofmann & Kotabe, 2012; Trope & Fishbach, 2000). Specifically, as they extract lessons from their self-control failures and successes, they may avoid tempting places, install reminders of conflict, pre-commit in public, arrange for more rest, and so on, thus selectively shaping specific components of integrative self-control theory in advance. Third, people may hold specific beliefs about how to best balance temptation and self-control goal enactment in their daily lives. Such beliefs may even account for strong fluctuations in self-control motivation from one episode to the next. As understanding of the temporal dynamics of the specific psychological components of selfcontrol increases, so will general understanding of how people navigate their desire-laden worlds over time.

11.6 Conclusions and Outlook Integrative self-control theory provides a theoretical basis for behavioral change interventions of self-control, derived from a seven-component analysis of self-control processes. It attempts to explain the behavioral outcomes (i.e., self-control goal enactment; desire enactment) of a given selfcontrol episode by providing a psychological analysis at the level of an individual episode. As such, the integrative self-control theory may provide a useful framework that is tailored enough to the specific phenomenon of self-control, defined as an intrapsychological conflict among competing

Changing Behavior Using Integrative Self-Control Theory

Pre-Behavior

Control Capacity

159

Behavior

EXERTION CLUSTER

Control Effort

Prevailing Force is Control Effort

Self-Control Success

Control Motivation

Activation

Desire vs. Control Effort

Enactment Constraints

Desire-Goal Conflict Prevailing Force is Desire

Self-Control Failure

Desire

Self-Control Goal

ACTIVATION CLUSTER

Figure 11.1 A diagram of integrative self-control theory (after Kotabe & Hofmann, 2015). Adapted with permission from Kotabe, H. P., & Hofmann, W. (2015). On integrating the components of self-control. Perspectives on Psychological Science, 10, 618–638. https://doi.org/10.1177/1745691615593382

short-term and long-term motivations, and broad enough to be integrative across the somewhat fragmented landscape of basic self-control research and its applied interventions.

References Ajzen, I., & Albarracín, D. (2007). Predicting and changing behavior: A reasoned action approach. In I. Ajzen, D. Albarracín, & R. Hornik (Eds.), Prediction and Change of Health Behavior: Applying the Reasoned Action Approach (pp. 3–21). Mahwah, NJ: Lawrence Erlbaum Associates. Alberts, H. J. E. M., Mulkens, S., Smeets, M., & Thewissen, R. (2010). Coping with food cravings. Investigating the potential of a mindfulness-based intervention. Appetite, 55, 160–163. https://doi .org/10.1016/j.appet.2010.05.044 Allom, V., Mullan, B., & Hagger, M. (2016). Does inhibitory control training improve health

behavior? A meta-analysis. Health Psychology Review, 10, 168–186. https://doi.org/10.1080/ 17437199.2015.1051078 Attwood, A. S., O’Sullivan, H., Leonards, U., Mackintosh, B., & Munafò, M. R. (2008). Attentional bias training and cue reactivity in cigarette smokers. Addiction, 103, 1875–1882. https://doi.org/10.1111/j.1360-0443.2008 .02335.x Baer, R. A. (2003). Mindfulness training as a clinical intervention: A conceptual and empirical review. Clinical Psychology: Science and Practice, 10, 125–143. https://doi.org/ 10.1093/clipsy.bpg015 Baumeister, R. F., Bratslavsky, E., Muraven, M., & Tice, D. M. (1998). Ego depletion: Is the active self a limited resource? Journal of Personality and Social Psychology, 74, 1252. https://doi.org/ 10.1037/0022-3514.74.5.1252 Baumeister, R. F., Vohs, K. D., DeWall, N., & Zhang, L. (2007). How emotion shapes behavior: Feedback,

160

WILHELM HOFMANN, SIMONE DOHLE, AND KATHI DIEL

anticipation, and reflection, rather than direct causation. Personality and Social Psychology Review, 11, 167–203. https://doi.org/10.1177/ 1088868307301033 Berridge, K. C., Robinson, T. E., & Aldridge, J. W. (2009). Dissecting components of reward: “liking,” “wanting,” and learning. Current Opinion in Pharmacology, 9, 65–73. https://doi .org/10.1016/j.coph.2008.12.014 Bernecker, K., Job, V., & Hofmann, W. (2018). Experience, resistance, and enactment of desires: Differential relationships with trait measures predicting self-control. Journal of Research in Personality, 76, 92–101. https://doi.org/10.1016/j .jrp.2018.07.007 Botvinick, M. M., Braver, T. S., Barch, D. M., Carter, C. S., & Cohen, J. D. (2001). Conflict monitoring and cognitive control. Psychological Review, 108, 624–652. https://doi.org/10.1037/0033295X.108.3.624 Brehm, J. W., & Self, E. A. (1989). The intensity of motivation. Annual Review of Psychology, 40, 109–131. https://doi.org/10.1146/annurev .ps.40.020189.000545 Carver, C. S., & Scheier, M. F. (1982). Control theory: A useful conceptual framework for personality– social, clinical, and health psychology. Psychological Bulletin, 92, 111–135. https://doi .org/10.1037/0033-2909.92.1.111 Delaney, L., & Lades, L. K. (2017). Present bias and everyday self-control failures: A day reconstruction study. Journal of Behavioral Decision Making, 30, 1157–1167. https://doi.org/ 10.1002/bdm.2031 De Witt Huberts, J. C., Evers, C., & De Ridder, D. T. (2014). “Because I am worth it”: A theoretical framework and empirical review of a justificationbased account of self-regulation failure. Personality and Social Psychology Review, 18, 119–138. https://doi.org/10.1177/ 1088868313507533 Dohle, S., & Hofmann, W. (2019). Consistency and balancing in everyday health behavior: An ecological momentary assessment approach. Applied Psychology: Health and Well-Being, 11, 148–169. https://doi.org/10.1111/ aphw.12148

Du, J., van Koningsbruggen, G. M., & Kerkhof, P. (2018). A brief measure of social media self-control failure. Computers in Human Behavior, 84, 68–75. https:// doi.org/10.1016/j.chb.2018.02.002 Duckworth, A. L., Gendler, T. S., & Gross, J. J. (2016). Situational strategies for self-control. Perspectives on Psychological Science, 11, 35–55. https://doi.org/10.1177/1745691615623247 Fishbach, A., & Hofmann, W. (2015). Nudging selfcontrol: A smartphone intervention of temptation anticipation and goal resolution improves everyday goal progress. Motivation Science, 1, 137. http://dx.doi.org/10.1037/ mot0000022 Fishbach, A., & Zhang, Y. (2008). Together or apart: When goals and temptations complement versus compete. Journal of Personality and Social Psychology, 94, 547–559. https://doi.org/10.1037/ 0022-3514.94.4.547 Florsheim, P., Heavin, S., Tiffany, S., Colvin, P., & Hiraoka, R. (2008). An experimental test of a craving management technique for adolescents in substance-abuse treatment. Journal of Youth and Adolescence, 37, 1205–1215. https://doi.org/ 10.1007/s10964-007-9232-0 Forman, E. M., Hoffman, K. L., McGrath, K. B., Herbert, J. D., Brandsma, L. L., & Lowe, M. R. (2007). A comparison of acceptance- and control-based strategies for coping with food cravings: An analog study. Behavior Research and Therapy, 45, 2372– 2386. https://doi.org/10.1016/j.brat.2007.04.004 Fujita, K. (2011). On conceptualizing self-control as more than the effortful inhibition of impulses. Personality and Social Psychology Review, 15, 352–366. https://doi.org/10.1177/ 1088868311411165 Fujita, K., & Han, H. A. (2009). Moving beyond deliberative control of impulses: The effect of construal levels on evaluative associations in selfcontrol conflicts. Psychological Science, 20, 799– 804. https://doi.org/10.1111/j.14679280.2009.02372.x Godin, G., & Kok, G. (1996). The theory of planned behavior: A review of its applications to healthrelated behaviors. American Journal of Health Promotion, 11, 87–98. https://doi.org/10.4278/ 0890-1171-11.2.87

Changing Behavior Using Integrative Self-Control Theory

Hanks, A. S., Just, D. R., Smith, L. E., & Wansink, B. (2012). Healthy convenience: Nudging students toward healthier choices in the lunchroom. Journal of public health, 34, 370–376. https://doi .org/10.1093/pubmed/fds003 Hardcastle, S. J., Fortier, M. S., Blake, N., & Hagger, M. S. (2017). Identifying content-based and relational techniques to change behavior in motivational interviewing. Health Psychology Review, 11, 1–16. https://doi.org/10.1080/ 17437199.2016.1190659 Hofmann, W., Adriaanse, M., Vohs, K. D., & Baumeister, R. F. (2014). Dieting and the selfcontrol of eating in everyday environments: An experience sampling study. British Journal of Health Psychology, 19, 523–539. https://doi.org/ 10.1111/bjhp.12053 Hofmann, W., Baumeister, R. F., Förster, G., & Vohs, K. D. (2012). Everyday temptations: An experience sampling study of desire, conflict, and self-control. Journal of Personality and Social Psychology, 102, 1318–1335. https://doi.org/ 10.1037/a0026545 Hofmann, W., & Fisher, R. R. (2012). How guilt and pride shape subsequent self-control. Social Psychological and Personality Science, 3, 682–690. https://doi.org/ 10.1177/1948550611435136 Hofmann, W., & Kotabe, H. P. (2012). A general model of preventive and interventive self-control. Social and Personality Psychology Compass, 6, 707– 722. https://doi.org/10.1111/j.17519004.2012.00461.x Hofmann, W., & Nordgren, L. F. (Eds.). (2015). The Psychology of Desire. New York: Guilford Press. Hofmann, W., Reinecke, L., Meier, A., & Oliver, M. (2017). Of sweet temptations and bitter aftertaste: Self-control as a moderator of the effects of media use on well-being. In L. Reinecke & M. B. Oliver (Eds.), The Routledge Handbook of Media Use and Well-Being: International Perspectives on Theory and Research on Positive Media Effects (pp. 211–222). New York: Routledge. Hofmann, W., Schmeichel, B. J., & Baddeley, A. D. (2012). Executive functions and self-regulation. Trends in Cognitive Sciences, 16, 174–180. https://doi.org/10.1016/j.tics.2012.01.006

161

Houben, K., Havermans, R. C., & Wiers, R. W. (2010). Learning to dislike alcohol: Conditioning negative implicit attitudes toward alcohol and its effect on drinking behavior. Psychopharmacology, 211, 79–86. https://doi.org/10.1007/s00213-010-1872-1 Houben, K., Nederkoorn, C., Wiers, R. W., & Jansen, A. (2011). Resisting temptation: Decreasing alcohol-related affect and drinking behavior by training response inhibition. Drug and Alcohol Dependence, 116, 132–136. https://doi.org/ 10.1016/j.drugalcdep.2010.12.011 Houben, K., Schoenmakers, T. M., & Wiers, R. W. (2010). I didn’t feel like drinking but I don’t know why: The effects of evaluative conditioning on alcohol-related attitudes, craving and behavior. Addictive Behaviors, 35, 1161–1163. https://doi .org/10.1016/j.addbeh.2010.08.012 Houben, K., Wiers, R. W., & Jansen, A. (2011). Getting a grip on drinking behavior: Training working memory to reduce alcohol abuse. Psychological Science, 22, 968–975. https://doi.org/10.1177/ 0956797611412392 Inzlicht, M., Schmeichel, B. J., & Macrae, C. N. (2014). Why self-control seems (but may not be) limited. Trends in Cognitive Sciences, 18, 127–133. https://doi.org/10.1016/j.tics.2013.12.009 Jaeggi, S. M., Buschkuehl, M., Jonides, J., & Perrig, W. J. (2008). Improving fluid intelligence with training on working memory. Proceedings of the National Academy of Sciences of the United States of America, 105, 6829–6833. https://doi.org/ 10.1073/pnas.0801268105 Janz, N. K., & Becker, M. H. (1984). The health belief model: A decade later. Health Education Quarterly, 11, 1–47. https://doi.org/10.1177/ 109019818401100101 Johnson, E. J., & Goldstein, D. (2003). Do defaults save lives? Science, 302, 1338–1339. https://doi.org/ 10.1126/science.1091721 Johnson, B. T., Wolf, L., Maio, G., & Smith-McLallen, A. (2019). Communication-induced persuasion or resistance: Processes and effects of who says what to whom. In B. T. Johnson & D. Albarracín (Eds.), The Handbook of Attitudes (2nd ed., Vol. 1, pp. 557–601). New York, NY: Psychology Press. Kavanagh, D. J., Andrade, J., & May, J. (2005). Imaginary relish and exquisite torture: The

162

WILHELM HOFMANN, SIMONE DOHLE, AND KATHI DIEL

elaborated intrusion theory of desire. Psychological Review, 112, 446–467. https://doi .org/10.1037/0033-295x.112.2.446 Kotabe, H. P., & Hofmann, W. (2015). On integrating the components of self-control. Perspectives on Psychological Science, 10, 618–638. https://doi .org/10.1177/1745691615593382 Kruglanski, A. W., Belanger, J. J., Chen, X. Y., Kopetz, C., Pierro, A., & Mannetti, L. (2012). The energetics of motivated cognition: A force-field analysis. Psychological Review, 119, 1–20. https:// doi.org/10.1037/A0025488 Kruglanski, A. W., Shah, J. Y., Fishbach, A., Friedman, R., Chun, W. Y., & Sleeth-Keppler, D. (2002). A theory of goal systems. In M. P. Zanna (Ed.), Advances in Experimental Social Psychology (pp. 331–378). San Diego, CA: Academic Press. Krönke, K.-M., Wolff, M., Mohr, H. et al. (2018). Monitor yourself! Deficient error-related brain activity predicts real-life self-control failures. Cognitive, Affective, and Behavioral Neuroscience, 18, 622–637. https://doi.org/ 10.3758/s13415-018-0593-5 Lewin, K. (1951). Field Theory in Social Science: Selected Theoretical Papers. Oxford: Harper. Lian, H., Yam, K. C., Ferris, D. L., & Brown, D. (2017). Self-control at work. Academy of Management Annals, 11, 703–732. https://doi.org/10.5465/ annals.2015.0126 Lopez, R. B., Hofmann, W., Wagner, D. D., Kelley, W. M., & Heatherton, T. F. (2014). Neural predictors of giving in to temptation in daily life. Psychological Science, 25, 1337–1344. https:// doi.org/10.1177/0956797614531492 Mahoney, M. J., & Thoresen, C. E. (1972). Behavioral self-control: Power to the person. Educational Researcher, 1, 5–7. https://doi.org/10.3102/ 0013189X001010005 McFall, R. M. (1970). Effects of self-monitoring on normal smoking behavior. Journal of Consulting and Clinical Psychology, 35, 135–142. https://doi .org/10.1037/h0030087 Metcalfe, J., & Mischel, W. (1999). A hot/cool-system analysis of delay of gratification: Dynamics of willpower. Psychological Review, 106, 3–19. https://doi.org/10.1037/0033-295X.106.1.3

Michie, S., Atkins, L., & West, R. (2014). The Behavior Change Wheel: A Guide to Designing Interventions. London: Silverback Publishing. Michie, S., Van Stralen, M. M., & West, R. (2011). The behavior change wheel: A new method for characterising and designing behavior change interventions. Implementation Science, 6, 42. https://doi.org/10.1186/1748-5908-6-42 Milkman, K. L., Minson, J. A., & Volpp, K. G. (2013). Holding the Hunger Games hostage at the gym: An evaluation of temptation bundling. Management Science, 60, 283–299. https://doi .org/10.1287/mnsc.2013.1784 Miller, W. R., & Rollnick, S. (2012). Motivational Interviewing: Helping People Change (3rd ed.). New York: Guilford Press. Mischel, W., Ebbesen, E. B., & Zeiss, A. M. (1972). Cognitive and attentional mechanisms in delay of gratification. Journal of Personality and Social Psychology, 21, 204–218. https://doi.org/10.1037/ h0032198 Mischel, W., Shoda, Y., & Peake, P. K. (1988). The nature of adolescent competencies predicted by preschool delay of gratification. Journal of Personality and Social Psychology, 54, 687–696. https://doi.org/10.1037/00223514.54.4.687 Moffitt, T. E., Arseneault, L., Belsky, D. et al. (2011). A gradient of childhood self-control predicts health, wealth, and public safety. Proceedings of the National Academy of Sciences, 108, 2693–2698. https://doi.org/10.1073/pnas.1010076108 Moskowitz, G. B., & Grant, H. (Eds.). (2009). The Psychology of Goals. New York: Guilford Press. Nielsen, K. S. (2017). From prediction to process: A self-regulation account of environmental behavior change. Journal of Environmental Psychology, 51, 189–198. https://doi.org/10.1016/j .jenvp.2017.04.002 Olesen, P. J., Westerberg, H., & Klingberg, T. (2003). Increased prefrontal and parietal activity after training of working memory. Nature Neuroscience, 7, 75–79. https://doi.org/10.1038/ nn1165 Ozaki, Y., Goto, T., Kobayashi, M., & Hofmann, W. (2017). Counteractive control over temptations: Promoting resistance through enhanced

Changing Behavior Using Integrative Self-Control Theory

perception of conflict and goal value. Self and Identity, 16, 439–459. https://doi.org/10.1080/ 15298868.2016.1269668 Pichert, D., & Katsikopoulos, K. V. (2008). Green defaults: Information presentation and proenvironmental behaviour. Journal of Environmental Psychology, 28, 63–73. https://doi .org/10.1016/j.jenvp.2007.09.004 Schelling, T. C. (1984). Self-command in practice, in policy, and in a theory of rational choice. The American Economic Review, 74, 1–11. Schroeder, S. A. (2007). We can do better: Improving the health of the American people. New England Journal of Medicine, 357, 1221–1228. https://doi .org/10.1056/NEJMsa073350 Shipstead, Z., Redick, T. S., & Engle, R. W. (2012). Is working memory training effective? Psychological Bulletin, 138, 628–654. https://doi .org/10.1037/a0027473 Sobal, J., & Wansink, B. (2007). Kitchenscapes, tablescapes, platescapes, and foodscapes: Influences of microscale built environments on food intake. Environment and Behavior, 39, 124– 142. https://doi.org/10.1177/0013916506295574 Strecher, V. J., Seijts, G. H., Kok, G. J. et al. (1995). Goal-setting as a strategy for health behaviorchange. Health Education Quarterly, 22, 190– 200. https://doi.org/10.1177/ 109019819502200207 Swan, M. (2013). The quantified self: Fundamental disruption in big data science and biological discovery. Big Data, 1, 85–99. https://doi.org/ 10.1089/big.2012.0002 Tangney, J. P., Baumeister, R. F., & Boone, A. L. (2004). High self-control predicts good adjustment, less pathology, better grades, and interpersonal success. Journal of Personality, 72, 271–324. https://doi.org/10.1111/j.00223506.2004.00263.x Thaler, R. H., & Sunstein, C. R. (2009). Nudge: Improving Decisions about Health, Wealth, and Happiness. New York: Penguin Books. Thaler, R. H., Sunstein, C. R., & Balz, J. P. (2010). Choice architecture. SSRN. http://ssrn.com/ abstract=1583509

163

Thorndike, A. N., Sonnenberg, L., Riis, J., Barraclough, S., & Levy, D. E. (2012). A 2-phase labeling and choice architecture intervention to improve healthy food and beverage choices. Journal of Public Health, 102, 527–533. https:// doi.org/10.2105/AJPH.2011.300391 Thush, C., Wiers, R. W., Moerbeek, M. et al. (2009). Influence of motivational interviewing on explicit and implicit alcohol-related cognition and alcohol use in at-risk adolescents. Psychology of Addictive Behaviors, 23, 146–151. https://doi.org/10.1037/ a0013789 Trope, Y., & Fishbach, A. (2000). Counteractive selfcontrol in overcoming temptation. Journal of Personality and Social Psychology, 79, 493–506. https://doi.org/10.1037/0022-3514.79.4.493 Van Dillen, L., Papies, E. K., & Hofmann, W. (2013). Turning a blind eye to temptation: How task load can facilitate self-regulation. Journal of Personality and Social Psychology, 104, 427–443. https://doi.org/10.1037/a0031262 Van Gucht, D., Baeyens, F., Vansteenwegen, D., Hermans, D., & Beckers, T. (2010). Counterconditioning reduces cue-induced craving and actual cue-elicited consumption. Emotion, 10, 688–695. https://doi.org/10.1037/A0019463 Webb, T. L., Miles, E., & Sheeran, P. (2012). Dealing with feeling: A meta-analysis of the effectiveness of strategies derived from the process model of emotion regulation. Psychological Bulletin, 138, 775–808. https:// doi.org/10.1037/a0027600 Wiers, R. W., Eberl, C., Rinck, M., Becker, E. S., & Lindenmeyer, J. (2011). Retraining automatic action tendencies changes alcoholic patients’ approach bias for alcohol and improves treatment outcome. Psychological Science, 22, 490–497. https://doi.org/10.1177/ 0956797611400615 Yon, B. A., Johnson, R. K., Harvey-Berino, J., Gold, B. C., & Howard, A. B. (2007). Personal digital assistants are comparable to traditional diaries for dietary self-monitoring during a weight loss program. Journal of Behavioral Medicine, 30, 165–175.

12 Changing Behavior Using the ReflectiveImpulsive Model Roland Deutsch and Fritz Strack

Practical Summary According to the reflective-impulsive model (RIM), two mental systems jointly influence behavior in response to the environment. In doing so, the two systems follow different principles and operate under different conditions. The impulsive system is always active, is confined to associations within its memory, and changes its responses depending on the deprivation of basic needs. It can also be broadly tuned to approach or avoidance behavior. The reflective system primarily operates when people are motivated and able to think and decide thoroughly. It is capable of abstract thinking in terms of logic and rationality and is driven by goals. Behavioral interventions based on the RIM primarily focus on changes of associations in the impulsive system such that desired behavior becomes more likely. Although there are several demonstrations of their effectiveness, there is still ongoing debate about when and why exactly these interventions work.

12.1 Introduction The reflective-impulsive model (RIM; Strack & Deutsch, 2004, 2015) attempts to capture the interplay of multiple psychological processes (e.g., perception, memory and learning, reasoning, motivation) in their interactive control of behavior. As such, the RIM is an integrative theory that is based on earlier theorizing and research within and beyond social psychology. It rests on the assumption that the processes underlying behavior can be broadly categorized into two systems, which are assumed to differ in their degree to which they can operate automatically (i.e., independent of intentions, fast, efficient) and the degree to which they depend on propositional mental representations, which flexibly connect pieces of knowledge and have a subjective truth-value. Importantly, the two

systems are the bearers of several different processes that interact at different stages of information processing (for an overview, see Strack & Deutsch, 2015). Thus, the RIM does not belong to the category of dual-process models that merely describe different types of processing under specific conditions (e.g., Ouellette & Wood, 1998; Petty & Cacioppo, 1986; see also Chapters 15 and 41, this volume) but represents a true dual-systems model that focuses on the interplay between the different mechanisms. The present chapter describes the core structures and processes that characterize the RIM as well as the relevant evidence to support them. Furthermore, it reviews how the RIM was applied to inform behavior change. https://doi.org/10.1017/9781108677318.012

Changing Behavior Using the Reflective-Impulsive Model

12.2 Brief Overview of the Theory and Evidence The RIM assumes that behavior is controlled by two interacting systems of processing: an impulsive system and a reflective system. Both systems have access to a final common pathway to

165

behavior, where behavioral schemata compete for execution.

12.2.1 The Impulsive System The impulsive system is an assembly of psychological mechanisms with properties that enable

Sidebar 12.1 Key terms and concepts of the reflective-impulsive model

Behavioral schema. Within the context of the reflective-impulsive model (RIM), the term behavioral schema refers to memory representations that associatively connect those representations that often co-occurred in the course of actions (i.e., situational and inner conditions, the behavior itself, as well as the consequences of that behavior). Associative representations. Within the context of the RIM, this term refers to connections between elements (e.g., apple, green) in memory that represent nothing more than mutual activation (e.g., activating apple activates green). If multiple elements are connected via associations, this group of associated elements can be referred to as associative cluster. Accessibility. In the context of the RIM, accessibility refers to the ease (i.e., speed, probability) with which memory contents are retrieved and therefore influence further processing. It typically depends on the recency and frequency of prior activation in memory. Syllogistic reasoning. Syllogistic reasoning is a mental activity by which conclusions are drawn from premises according to the rules of syllogistic logic. For example, a premise might be that all alcoholic drinks may cause cancer. Another premise might be that wine is an alcoholic drink. The conclusion would be that wine may cause cancer. Propositional representations. Within the context of the RIM, this term refers to mental representations of states of the world (e.g., this is a green apple) that combine more basic representations (e.g., apple, green) by flexibly assigning semantic relations (e.g., is a) and that have a subjective truth-value. Ironic/paradox effects. In the context of self-control, this term refers to situations where attempts to regulate a behavior in a particular way cause the opposite of what was intended. For example, when the intention not to laugh increases laughing. Evaluative conditioning. A procedure by which neutral stimuli are paired with valent (i.e., positive or negative) stimuli, typically in a repeated manner. A typical result is that the valence of the formerly neutral stimuli becomes more similar to the valence of the valent stimuli. For example, a neutral face that is repeatedly paired with a negative picture will become more negative.

166

ROLAND DEUTSCH AND FRITZ STRACK

behavior under a wide range of conditions. More specifically, the impulsive system is assumed to function under distraction, fatigue, and time pressure as well as when such cognitive strains are absent. In other words, it is relatively independent from cognitive resources and therefore operates in an automatic manner, just like driving a car on an empty freeway. As such, it can be seen as a default “response generator” that is always active. In the next sections, the core psychological processes that are part of the impulsive system are outlined.

one element of an associative cluster receives activation, from inner (e.g., goals, emotions) or outer sources (stimuli in a situation), activation spreads to other elements in proportion to their association strength. In the above-mentioned restaurant example, seeing the restaurant again would be expected to result in an increased activation of past eating in the restaurant as well as the satisfaction that went along with eating, thereby rendering the actual execution of the behavior more likely (cf. Custers & Aarts, 2005; Kahnt et al., 2010).

12.2.1.1 Behavioral Schemata

12.2.1.2 Homeostatic Dysregulation

The central element of the impulsive system is an associative memory that binds representations of outer and inner events depending on the frequency of coactivation. The representations include sensory codes (e.g., sights, odors, sounds), evaluative codes (e.g., pleasant/unpleasant), and motor codes (i.e., those representations that stand for concrete muscle movements). This architecture implies the emergence of habit-like structures (cf. Chapters 13 and 41, this volume) in the impulsive system, which link representations of situations, behaviors, and behavioral outcomes that, in past experience, often co-occurred. Within the model, such structures are referred to as behavioral schemata (see Sidebar 12.1). For example, if a person often entered a certain restaurant in a hungry state, ate schnitzel and potato salad there, and then felt satisfied, all sensory representations relating to the feeling of hunger, the restaurant, and the food will be associated with motor codes relating to entering the restaurant, as well as eating. Also, associative representations of satisfaction will become linked with the other elements of this associative cluster (see Sidebar 12.1). For example, thoughts of a good meal may trigger feelings of happiness and vice versa (cf. Kahnt et al., 2010). Representations in the impulsive system follow the principles of accessibility and associative activation (Higgins, 1996; see Sidebar 12.1). If

Behavioral schemata, as described so far, reflect past experience and current activation through the external situation or inner states. The RIM holds a special role for a certain type of inner state: the dysregulation of organismic needs such as hydration or food supply. Deprivation is assumed to result in a strong inner representation of this condition, typically going along with strong subjective experiences (e.g., hunger, thirst). By means of learning, these representations can become part of behavioral schemata, such that, in a deprived state (e.g., while being thirsty), all behavioral schemata of which the representation of deprivation is part (e.g., schemata about behaviors that were executed in a thirsty state) will receive activation. This is predicted to result in a perceptual readiness for matching situational conditions. As a consequence, situations in which past behavior remedied deprivation (e.g., situations in which fluid is available) as well as the concrete objects related to deprivation (e.g., a bottle of water; a drinking fountain) will be more likely recognized (Aarts, Dijksterhuis, & De Vries, 2001; Wispé & Drambarean, 1953). The mental activation of need-matching situations is also predicted to result in need-congruent interpretations of ambiguous stimuli (Dunning & Balcetis, 2013) and to facilitate the processing of stimuli that are associated with need-related behaviors (Radel & Clément-Guillotin, 2012).

Changing Behavior Using the Reflective-Impulsive Model

For example, being thirsty is predicted to render mental representations of drinkable fluids more accessible in memory. When encountering a stimulus that might or might not represent something drinkable (e.g., a bottle), the ambiguous stimulus will be more likely interpreted as representing something drinkable (e.g., a bottle of water). In tandem with situational inputs and associated reward, deprivation may push that schema above the threshold, thereby preparing for overt behavior. Thus, a hungry state would activate the schema of entering the particular restaurant and eating schnitzel and salad there. If this restaurant happens to be in sight, the schema would receive extra activation that other competing schemata are missing and may therefore be selected for execution.

12.2.1.3 Motivational Orientation Behavioral tendencies generated by the impulsive system depend on a third mechanism, which is called motivational orientation. Motivational orientation can be understood as a general tuning of the impulsive system toward either approaching or avoiding stimuli or situations. When in an approach orientation, positive information is more easily processed, approach behaviors are more easily triggered, and positive affect is predominant. As an example, in an approach orientation, which might be induced by being in a good mood, one’s favorite dessert will be more visually salient and behaviors that result in consumption will be more easily triggered. In an avoidance motivation, negative information is more easily processed, avoidance behaviors are more easily triggered, and negative affect is predominant. As an example, in an avoidance orientation, which might be induced by being anxious, threat stimuli such as angry faces will be more visually salient and social withdrawal in response to such faces will be more easily triggered. The RIM further suggests that a congruent motivational orientation will be induced by any input to the impulsive system that has a valence or represents approach versus avoidance.

167

12.2.2 The Reflective System While the impulsive system operates automatically, it is not able to solve a number of cognitive tasks. Specifically, it cannot represent information in a propositional format, which is necessary to engage in syllogistic reasoning, such as drawing an inference (see Sidebar 12.1). For example, if someone holds the premise that all physical exercise increases heart health and that swimming is physical exercise, then this person can conclude that swimming increases heart health. If the person holds the goal to increase heart health, syllogistic reasoning might result in the intention to engage in swimming. Thus, the RIM has a second, propositional processing system, the reflective system, which requires more cognitive resources than the impulsive system, depends on intentions, and operates at a slower speed.

12.2.2.1 Propositional Representations Propositions are mental representations of states of the world that flexibly combine elements by relations and that the holder considers as being true or false (i.e., they have a truth-value; see Sidebar 12.1). This means that their contents are based on people’s beliefs about reality. In contrast, associations, which are the main representational format of the impulsive system, connect elements merely by associative links without further qualifications (for an extensive discussion of the distinction between propositions and associations, see Deutsch, 2017). So, within the restaurant example in Section 12.2.1.1, an association between a certain restaurant and eating will manifest itself such that, when the representation of the restaurant is activated, a representation of eating will simply “pop into one’s mind.” On a propositional level, however, the restaurant and eating will be combined by more abstract qualifications, perhaps resulting in the belief that “one can eat lunch at this restaurant.” Whereas an association does not imply any knowledge about what is the case (it is a mere

168

ROLAND DEUTSCH AND FRITZ STRACK

cognitive occurrence), the proposition includes the claim of correspondence with reality.

12.2.2.2 Negation Generating and transforming such representations is one of the exclusive capacities of the reflective system. In general, propositional representations and further transformations can apply to an unlimited number of things, situations, or inner states. One area, however, that is particularly important from the perspective of the RIM is negation, which reverses a stated claim of reality correspondence. For example, on a purely associative level, as is present in the impulsive system, the activation of the concepts food, restaurant, good, and not will result in an overall positive response as most of the components are positive. Only after these components have been integrated into the proposition “the food in this restaurant is not good” will there be an overall negative response. Indeed, there is ample evidence that people’s attempts not to engage in a particular mental activity (e.g., don’t think of a white bear) may result in the opposite (Wegner, 1994). Similarly, negations can have paradoxical effects (see Sidebar 12.1), especially under conditions of shallow processing (e.g., Deutsch, Gawronski, & Strack, 2006) or when the intention to negate is missing (e.g., Wiswede et al., 2013). These theoretical assumptions are crucial when it comes to attempts to change behavior via negated imperatives like “you shall not smoke” or “just say no.” The success of such attempts depends on whether conditions for reflective processing are present or not. However, recent evidence has expanded the view on propositional processing in general and negation processing in particular (for an overview, see Deutsch, 2017), suggesting that negations may operate faster and more independent from intentions than previously assumed (Deutsch et al., 2009). Further research and theorizing are needed to further elucidate the boundaries and conditions of such apparently automatic processing of propositions.

12.2.2.3 Behavioral Decisions and Intending Another form of propositional process in the reflective system is that of behavioral decisionmaking. It is similar to expectancy-value models of beliefs and self-efficacy (see Chapters 2 and 3, this volume) and results in the formation of a behavioral intention. As a result, this process shares much in common with expectancyvalue models of behavior (for more detailed descriptions, see Deutsch & Strack, 2008, 2010). A final important process residing within the reflective system is intending, which “bridges” behavioral decisions and overt behavior by automatically activating behavioral schemata relevant for goal-attainment. Intending is similar to the concept of implementation intentions (see Chapter 6, this volume). The relative strength of reflective and impulsive processes crucially determines behavioral outcomes (resembling self-regulation; see Chapter 11, this volume).

12.2.3 Critiques Both dual-process and dual-systems models have received some criticism (e.g., Gigerenzer & Regier, 1996; Kruglanski & Gigerenzer, 2011; Melnikoff & Bargh, 2018; Mitchell, De Houwer, & Lovibond, 2009; Moors, 2014). Some criticisms question the validity of some rather specific assumptions of single theories. For example, there has been an extensive discussion about the assumption of many duality theories that memory associations can be formed in the absence of propositional knowledge about these associations (Mitchell et al., 2009). Other criticisms are more general and question whether it is adequate or even possible to divide the realm of psychological processes into two chunks (e.g., Gigerenzer & Regier, 1996; Kruglanski & Gigerenzer, 2011; Melnikoff & Bargh, 2018; Moors, 2014). Instead, single-process models (Kruglanski et al., 2003), or multi-process models without duality assumptions (Sherman, 2006),

Changing Behavior Using the Reflective-Impulsive Model

169

Syllogistic rules

Noetic decision factual & evaluative

Propositional categorization Pointing & reffering

Reflective system (RS)

Reasoning

Reasoning

Noetic awareness

Behavioral decision

Intending

Spreading activation

“Reasoned action”

{Experiential Perception imagination

Motivational orientation

awareness}

Behavioral schemata

Behavior

“Impulsive action”

Spreading activation Associative store Episodic and semantic links

Impulsive system (IS)

RS IS

Figure 12.1 Overview of the structure of the reflective-impulsive model (from Strack & Deutsch, 2004)

Source. Figure taken from Strack, F., & Deutsch, R. (2004). Reflective and impulsive determinants of social behavior. Personality and Social Psychology Review, 8, 220–247. Copyright © 2004 by Lawrence Erlbaum Associates, Inc. http://dx.doi.org/10.1207/s15327957pspr0803_1

have been proposed. These criticisms notwithstanding, duality models in general, and the RIM in particular, have proved to be highly successful in stimulating integrative research on behavior in the realm of social psychology and beyond (for a discussion of criticisms, see Deutsch & Strack, 2006a, 2006b).

12.3 How Has the ReflectiveImpulsive Model Been Used to Change Behavior? 12.3.1 Typical Targets and Techniques of Interventions The RIM has been used as a basis for behavior change in a large number of applied fields, often in conjunction with other theories, such as

theories of self-control (see Chapter 11, this volume). These applications include attitude change (marketing and advertising), temptation and self-regulation (eating and drinking), health behavior (weight gain, food choice, dental flossing), addiction, consumer behavior (impulse buying), behavior management (nudging, habit formation, procrastination, emotion regulation), sports (exercising, physical activities, learning new motor tasks), computer games and gambling, behavior therapy (fear, disgust, depression), trust in relationships, and aesthetics. In reviewing research from some of these areas, the next section will focus on those processes from the RIM that are considered most distinct and that were most often used as a basis for interventions. More specifically, a focus will be on impulsive processes (e.g., approach/avoidance, deprivation,

170

ROLAND DEUTSCH AND FRITZ STRACK

motivational orientations, evaluative associations, schema activation) and how they interact with their reflective counterparts (e.g., schema activation through intending; schema activation through propositional logic). A typical exemplary intervention based on the RIM is approach/avoidance training. Next, a review of the core techniques and empirical indicators of their short- and long-term effectiveness, as well as review research on moderators of these effects is provided.

12.3.2 Evaluative Associations and Behavior Change Within the RIM, evaluative associations play an important role as precursors of behavior. In the impulsive system, they can instigate motivational orientations and thereby direct behavioral responses. In the reflective system, evaluative associations can serve as the basis of evaluative judgments and, in the end, reasoned actions. It is not surprising that multiple studies have investigated the effectiveness of changing evaluative associations. In particular, evaluative conditioning (De Houwer, Thomas, & Baeyens, 2001) was used to change evaluative associations and, correspondingly, overt behavior toward those objects that were evaluatively associated (see Sidebar 12.1). Although evaluative conditioning has been shown to be effective in changing evaluative associations (Hofmann et al., 2010), results regarding its power to change corresponding behavior have been inconclusive. A recent meta-analysis on eating behavior provides no consistent evidence that evaluative conditioning–induced change of evaluative associations translates into overt behavior (Aulbach, Knittle, & Haukkala, 2019). For example, some findings show that evaluative conditioning results in changes in alcohol and soft-drink consumption (Houben, Havermans, & Wiers, 2010; Houben, Schoenmakers, & Wiers, 2010; Shaw et al., 2016;

Tello et al., 2018), but it has been disputed whether or not the effect was driven by changes in evaluative associations. Also, some of the studies suggest that the effect seems to depend on procedural factors of evaluative conditioning. For example, evaluative conditioning proved to be stronger when stimuli were presented above thresholds for conscious perception, when participants are aware of the relation between neutral and valent stimuli, or when the identical (instead of similar) stimuli are used for learning and later testing (Hofmann et al., 2010). Evidence also exists for evaluative conditioning effects on exercise behavior (Antoniewicz & Brand, 2016), smoking behavior (Măgurean, Constantin, & Sava, 2016), and condom use (Ellis et al., 2015). Evaluative conditioning seems particularly attractive as a technique for behavior change because it can be administered in a relatively rigid, mechanistic way: People simply have to observe pairings of stimuli. At the same time, cumulative evidence implies a more active role for perceivers than earlier theorizing had suggested. More specifically, it seems that perceivers need to be aware of what stimuli are paired for this technique to work. One interpretation of such observations is that it is not mere associations that are driving these effects but instead propositional knowledge that perceivers derive from the observation of pairings (e.g., Mitchell, De Houwer, & Lovibond, 2009). Further research should clarify the relative effectiveness and mediators of persuasion-based versus conditioning-based creation or change of knowledge about the valence of stimuli.

12.3.3 Negation and Behavior Change Negations are often used as means to instigate selfregulation – for example, in prohibitions or advice (e.g., “don’t overeat!”). From the perspective of the RIM, such attempts come with the risk of being ineffective or even counterproductive because processing negations activates the affirmative element

Changing Behavior Using the Reflective-Impulsive Model

that is part of the negation in memory (e.g., “eating”). Thus, for negations to be effective, not only must reflection be in place to extract the intended meaning of the negated imperative; moreover, a way to manage the activation of the negated constructs in the impulsive system is needed, which might go along with tempting evaluations (e.g., eating-positive) and motivational orientations (e. g., eating-approach). For example, the instruction not to smoke must be translated into concrete behavioral consequences. In addition, unwanted behavioral consequences of physical or mental exposure to the drug must be controlled and counteracted. This is particularly important as there are several experiments suggesting that negated information may exert paradox effects on behavior or its immediate precursors by triggering responses that oppose the intended outcome (for a review, see Deutsch, 2017). As an example, a study by Wegner, Ansfield, and Pilloff (1998) showed that participants who had been asked not to move a pendulum in a particular direction did just that if they were distracted by a secondary task (see also Wang, Hagger, & Chatzisarantis, 2020). In a more applied setting, a study by Adriaanse and colleagues (2011) suggests that intentions not to eat can actually increase eating behavior. Evidence also suggests that campaigns, advertisements, or signs using negated phrases such as “no smoking” (Earp et al., 2013; Farrelly et al., 2002) or “vaccination is no risk” (Betsch & Sachse, 2013) might cause opposing attitudes. Although these (and other) studies suggest that behavioral interventions containing negations should be handled with care, there is also growing evidence that, depending on specific conditions, negations might be more easily processed than previously thought. This was the case for highly practiced negations, comparable to expressions such as no way (Deutsch et al., 2006). Likewise, negations that have a clear opposite, comparable to not ill (Mayo, Schul, & Burnstein, 2004), may be processed relatively easily. An important task for future research will

171

be to further elucidate the variables moderating the effectiveness of negated imperatives.

12.3.4 Deprivation and Behavior Change The deprivation mechanism as specified in the RIM has important consequences for changing behavior. According to the mechanism, the deprivation of homeostatic needs increases the accessibility of those behavioral schemata that had previously been successful in fulfilling a need. As a consequence, attention to relevant triggers or cues is increased and the probability of executing corresponding behaviors is heightened. Several studies do suggest that food deprivation induces a motivational state in which food is automatically given greater priority in perception and evaluation (Hoefling et al., 2009; Loeber et al., 2013; Lozano, Crites, & Aikman, 1999; Orquin & Kurzban, 2016; Seibt, Häfner, & Deutsch, 2007; Stockburger et al., 2009) and can increase an individual’s vulnerability to problem behaviors (Nisbett & Kanouse, 1969), although the effects are typically moderated by within-person variables such as impulsivity or obesity status (Mela, Aaron, & Gatenby, 1996). As a potential technique to reduce problem behaviors, having people eat before entering a critical situation was shown to reduce the consumption of unhealthy food (Nederkoorn et al., 2009). As with the effects of deprivation on attitudes, such protective effects of preloading often depend on within-person variables such as bulimia nervosa (Hetherington et al., 2000), restrained eating (Polivy, Coleman, & Herman, 2005), or impulsivity (Nederkoorn et al., 2009).

12.3.5 Approach/Avoidance Exercises As described in Section 12.2.1.3, a central mechanism of the RIM is that of motivational orientation, which tunes the impulsive system toward either approaching or avoiding a given

172

ROLAND DEUTSCH AND FRITZ STRACK

target. Most important, spatial orientation and evaluation are causally connected in a bidirectional fashion. That is, a positive valence elicits tendencies toward approach, a negative valence facilitates avoidance. In reverse, approach associates its target with a positive valence, avoidance with a negative evaluation (for a review, see Krieglmeyer, De Houwer, & Deutsch, 2013). A plethora of studies has tested the effectiveness of interventions based on motivational orientation. More specifically, these interventions aim at strengthening and/or weakening motivational orientation by having people practice specific approach and/or avoidance behaviors toward stimuli (for a review, see Kakoschke, Kemps, & Tiggemann, 2017). In numerous studies with both patients and college students, Wiers and colleagues (e.g., Wiers et al., 2010) successfully reduced alcohol consumption by a focus on the motivational orientation. Using a joystick training task, participants had to push away (avoidance) alcohol pictures. More recently, Rinck et al. (2018) combined approach/avoidance training with an intervention to change attention toward alcohol cues in a large sample of alcohol-dependent patients. At a one-year follow-up, patients who had received both types of training were more successful at giving up alcohol than patients who had received a sham or no training. At the same time, a significant effect was only obtained if both types of training had been applied. By themselves, the attention and the avoidance training had only trivial effects. This finding suggests that, while the underlying mechanisms are clearly supported by the results, the effectiveness of the derived interventions needs to be further investigated. Another domain where RIM-related interventions have proven successful is that of overeating and obesity (for a critical review, see Jones et al., 2018). As in the domain of alcohol consumption, overeating is considered to be partly driven by impulsive components that are not necessarily in line with people’s goals. Thus, an intervention

that addresses eaters’ motivational orientations should be effective. Most recently, this was demonstrated by Mehl and colleagues (2018) who applied the joystick procedure to train participants to approach healthy food pictures and avoid unhealthy ones. They observed that only one training session reduced obese individuals’ approach tendencies for unhealthy food, which stayed weakened for the subsequent days. For normal-weight participants, approach tendencies toward healthy food could be strengthened through the training. Van Dessel, Hughes, and De Houwer (2018) observed effects on unhealthy eating behavior if approach/avoidance training was combined with affective consequences. These and other studies (for a recent meta-analysis, see Aulbach et al., 2019) suggest that automatic approach tendencies toward food can be changed through approach/avoidance training, thus offering additional possibilities for obesity treatment and healthy weight management. Another area of application is smoking cessation. Applying the principle of motivational orientation, Wittekind et al. (2015) conducted a study with 257 smokers and reported a significant reduction in cigarette smoking, cigarette dependence, and compulsive drive as a consequence of approach/avoidance training. However, reviews of the literature suggest that effects might be heterogeneous (Mühlig et al., 2016). Similarly, experimental research also suggests that phobias may be treated by approach/avoidance interventions. For example, Jones et al. (2013) found that fearful individuals who had been induced to approach spider pictures reported less anxiety when encountering live spiders than participants in the control condition. Published evidence, however, suggests that the effects of approach/avoidance training might be less stable than this brief review suggests. Mertens, Van Dessel, and De Houwer (2018) discuss relevant failures to observe effects of approach/avoidance training and provide theoretical and empirical arguments for moderating

Changing Behavior Using the Reflective-Impulsive Model

variables, which might help us to understand when approach/avoidance training is effective. One important variable studied by Mertens et al. (2018) is the functional match between stimulus valence and behavior. More specifically, their observations suggest that having people repeatedly approach very negative stimuli might not be effective, as the act of approaching negative stimuli in itself is aversive and might counteract potentially positive effects of avoidance.

12.4 Conclusion The current chapter provides examples on how interventions in many behavioral domains may operate on the basis of impulsive mechanisms as outlined in the RIM. In particular, inducing agents to either approach or avoid a behavioral target was frequently demonstrated to support therapeutic measures in various addictions and maladaptive behaviors. As the authors of a recent meta-analysis in the domain of overeating conclude, “dual-process models integrate deliberative and impulsive mental systems and predict dietary behaviors better than deliberative processes alone” (Aulbach et al., 2019, p. 179). This observation is in line with findings from some of the cited studies that only obtained weak effects (Rinck et al., 2018) or were able to increase the robustness of impulsive interventions by combining them with reflective consequences (Van Dessel et al., 2018). As Aulbach and colleagues’ findings imply, it is not the replacement of reflective processes by impulsive mechanisms but the integration of both that warrants reliable effects. This is exactly the core of the RIM that explains human behavior as the smooth cooperation and interaction of reflective and impulsive mechanisms. In contrast to other models, the RIM does not assume that either only one or only the other system is active. Instead, the systems are intertwined while specific conditions determine the relative contribution of one or the other.

173

As a consequence, the effectiveness of interventions does not depend on addressing selective mechanisms but on combining them in a way that reflects the dynamics of the problematic situation. This recommendation is particularly important if behaviors are controlled by different levels of automaticity (Vallacher & Wegner, 1987; see also Chapter 13, this volume). That is, if both the quality of actors’ experiences and the impulses that prevent them from reaching a particular goal are jointly operating, interventions from both the reflective and the impulsive system are called for. That is, explicit rewards (or punishments) in combination with facilitating (or inhibiting) impulsive procedures may be the silver bullet of behavioral interventions. This is exactly in the spirit of the RIM and its applied potential.

References Aarts, H., Dijksterhuis, A., & De Vries, P. (2001). On the psychology of drinking: Being thirsty and perceptually ready. British Journal of Psychology, 92, 631–642. http://dx.doi.org/10.1348/ 000712601162383 Adriaanse, M. A., van Oosten, J. M. F., de Ridder, D. T. D., de Wit, J. B. F., & Evers, C. (2011). Planning what not to eat: Ironic effects of implementation intentions negating unhealthy habits. Personality and Social Psychology Bulletin, 37, 69–81. http:// dx.doi.org/10.1177/0146167210390523 Antoniewicz, F., & Brand, R. (2016). Learning to like exercising: Evaluative conditioning changes automatic evaluations of exercising and influences subsequent exercising behavior. Journal of Sport and Exercise Psychology, 38, 138–148. http://dx .doi.org/10.1123/jsep.2015-0125 Aulbach, M. B., Knittle, K., & Haukkala, A. (2019). Implicit process interventions in eating behaviour: A meta-analysis examining mediators and moderators. Health Psychology Review, 13, 179– 208. http://dx.doi.org/10.1080/ 17437199.2019.1571933 Betsch, C., & Sachse, K. (2013). Debunking vaccination myths: Strong risk negations can

174

ROLAND DEUTSCH AND FRITZ STRACK

increase perceived vaccination risks. Health Psychology, 32, 146–155. http://dx.doi.org/ 10.1037/a0027387 Custers, R., & Aarts, H. (2005). Positive affect as implicit motivator: On the nonconscious operation of behavioral goals. Journal of Personality and Social Psychology, 89, 129–142. http://dx.doi.org/ 10.1037/0022-3514.89.2.129 De Houwer, J., Thomas, S., & Baeyens, F. (2001). Association learning of likes and dislikes: A review of 25 years of research on human evaluative conditioning. Psychological Bulletin, 127, 853–869. http://dx.doi.org/10.1037/00332909.127.6.853 Deutsch, R. (2017). Associative and propositional processes from the perspective of the reflectiveimpulsive model. In R. Deutsch, B. Gawronski, & W. Hofmann (Eds.), Reflective and Impulsive Determinants of Human Behavior (pp. 51–68). New York: Routledge. Deutsch, R., Gawronski, B., & Strack, F. (2006). At the boundaries of automaticity: Negation as reflective operation. Journal of Personality and Social Psychology, 91, 385–405. http://dx.doi.org/ 10.1037/0022-3514.91.3.385 Deutsch, R., Kordts-Freudinger, R., Gawronski, B., & Strack, F. (2009). Fast and fragile: A new look at the automaticity of negation processing. Experimental Psychology, 56, 434–446. http://dx .doi.org/10.1027/1618-3169.56.6.434 Deutsch, R., & Strack, F. (2006a). Duality models in social psychology: From dual processes to interacting systems. Psychological Inquiry, 17, 166–172. http://dx.doi.org/10.1207/ s15327965pli1703_2 Deutsch, R., & Strack, F. (2006b). Duality models in social psychology: Response to commentaries. Psychological Inquiry, 17, 265–268. http://dx.doi .org/10.1207/s15327965pli1703_6 Deutsch, R., & Strack, F. (2008). Variants of judgment and decision making: The perspective of the reflective-impulsive model. In H. Plessner, C. Betsch, & T. Betsch (Eds.), Intuition in Judgment and Decision Making (pp. 39–53). Mahwah, NJ: Lawrence Erlbaum Associates. Deutsch, R., & Strack, F. (2010). Building blocks of social behavior: Reflective and impulsive

processes. In B. Gawronski & B. K. Payne (Eds.), Handbook of Implicit Social Cognition: Measurement, Theory, and Applications (pp. 62–79). New York: Guilford Press. Dunning, D., & Balcetis, E. (2013). Wishful seeing: How preferences shape visual perception. Current Directions in Psychological Science, 22, 33–37. http://dx.doi.org/10.1177/0963721412463693 Earp, B. D., Dill, B., Harris, J. L., Ackerman, J. M., & Bargh, J. A. (2013). No sign of quitting: Incidental exposure to “no smoking” signs ironically boosts cigarette-approach tendencies in smokers. Journal of Applied Social Psychology, 43, 2158–2162. http://dx.doi.org/10.1111/jasp.12202 Ellis, E. M., Homish, G. G., Parks, K. A., Collins, R. L., & Kiviniemi, M. T. (2015). Increasing condom use by changing people’s feelings about them: An experimental study. Health Psychology, 34, 941– 950. http://dx.doi.org/10.1037/hea0000205 Farrelly, M. C., Healton, C. G., Davis, K. C., Messeri, P., Hersey, J. C., & Haviland, M. L. (2002). Getting to the truth: Evaluating national tobacco countermarketing campaigns. American Journal of Public Health, 92, 901–907. http://dx.doi.org/ 10.2105/AJPH.92.6.901 Gigerenzer, G., & Regier, T. (1996). How do we tell an association from a rule? Comment on Sloman (1996). Psychological Bulletin, 119, 23–26. http:// dx.doi.org/10.1037/0033-2909.119.1.23 Hetherington, M. M., Stoner, S. A., Andersen, A. E., & Rolls, B. J. (2000). Effects of acute food deprivation on eating behavior in eating disorders. International Journal of Eating Disorders, 28, 272–283. http://dx .doi.org/10.1002/1098-108X(200011)28:3% 3C272::AID-EAT4%3E3.0.CO;2-Q Higgins, E. T. (1996). Knowledge activation: Accessibility, applicability, and salience. In E. T. Higgins & A. W. Kruglanski (Eds.), Social Psychology: Handbook of Basic Principles (pp. 133–168). New York: Guilford Press. Hoefling, A., Likowski, K. U., Deutsch, R. et al. (2009). When hunger finds no fault with moldy corn: Food deprivation reduces food-related disgust. Emotion, 9, 50–58. http://dx.doi.org/10.1037/ a0014449 Hofmann, W., De Houwer, J., Perugini, M., Baeyens, F., & Crombez, G. (2010). Evaluative

Changing Behavior Using the Reflective-Impulsive Model

conditioning in humans: A meta-analysis. Psychological Bulletin, 136, 390–421. https://dx .doi.org/10.1037/A0018916 Houben, K., Havermans, R. C., & Wiers, R. W. (2010). Learning to dislike alcohol: Conditioning negative implicit attitudes toward alcohol and its effect on drinking behavior. Psychopharmacology, 211, 79– 86. http://dx.doi.org/10.1007/s00213-010-1872-1 Houben, K., Schoenmakers, T. M., & Wiers, R. W. (2010). I didn’t feel like drinking but I don’t know why: The effects of evaluative conditioning on alcohol-related attitudes, craving and behavior. Addictive Behaviors, 35, 1161–1163. http://dx.doi .org/10.1016/j.addbeh.2010.08.012 Jones, A., Hardman, C. A., Lawrence, N., & Field, M. (2018). Cognitive training as a potential treatment for overweight and obesity: A critical review of the evidence. Appetite, 124, 50–67. http://dx.doi .org/10.1016/j.appet.2017.05.032 Jones, C. R., Vilensky, M. R., Vasey, M. W., & Fazio, R. H. (2013). Approach behavior can mitigate predominately univalent negative attitudes: Evidence regarding insects and spiders. Emotion, 13, 989–996. http://dx.doi.org/10.1037/ a0033164 Kahnt, T., Heinzle, J., Park, S. Q., & Haynes, J. D. (2010). The neural code of reward anticipation in human orbitofrontal cortex. Proceedings of the National Academy of Sciences of the United States of America, 107, 6010–6015. http://dx.doi.org/ 10.1073/pnas.091283810 Kakoschke, N., Kemps, E., & Tiggemann, M. (2017). Approach bias modification training and consumption: A review of the literature. Addictive Behaviors, 64, 21–28. http://dx.doi.org/10.1016/j .addbeh.2016.08.007 Krieglmeyer, R., De Houwer, J., & Deutsch, R. (2013). On the nature of automatically triggered approach–avoidance behavior. Emotion Review, 5, 280–284. http://dx.doi.org/10.1177/ 1754073913477501 Kruglanski, A. W., Chun, W. Y., Erb, H. P., Pierro, A., Mannetti, L., & Spiegel, S. (2003). A parametric unimodel of human judgment: Integrating dualprocess frameworks in social cognition from a single-mode perspective. In J. P. Forgas, K. D. Williams, & W. von Hippel (Eds.), Social

175

Judgments: Implicit and Explicit Processes (pp. 137–161). New York: Cambridge University Press. Kruglanski, A. W., & Gigerenzer, G. (2011). Intuitive and deliberate judgments are based on common principles. Psychological Review, 118, 97–109. http://dx.doi.org/10.1037/a0020762 Loeber, S., Grosshans, M., Herpertz, S., Kiefer, F., & Herpertz, S. C. (2013). Hunger modulates behavioral disinhibition and attention allocation to food-associated cues in normal-weight controls. Appetite, 71, 32–39. http://dx.doi.org/10.1016/j .appet.2013.07.008 Lozano, D. I., Crites, S. L., & Aikman, S. N. (1999). Changes in food attitudes as a function of hunger. Appetite, 32, 207–218. http://dx.doi.org/10.1006/ appe.1998.0205 Măgurean, S., Constantin, T., & Sava, F. A. (2016). The indirect effect of evaluative conditioning on smoking. Journal of Substance Use, 21, 198–203. Mayo, R., Schul, Y., & Burnstein, E. (2004). “I am not guilty” vs. “I am innocent”: Successful negation may depend on the schema used for its encoding. Journal of Experimental Social Psychology, 40, 433–449. Mehl, N., Mueller-Wieland, L., Mathar, D., & Horstmann, A. (2018). Retraining automatic action tendencies in obesity. Physiology and Behavior, 192, 50–58. http://dx.doi.org/10.1016/j .physbeh.2018.03.031 Mela, D. J., Aaron, J. I., & Gatenby, S. J. (1996). Relationships of consumer characteristics and food deprivation to food purchasing behavior. Physiology and Behavior, 60, 1331–1335. http:// dx.doi.org/10.1016/S0031-9384(96)00241–7 Melnikoff, D. E., & Bargh, J. A. (2018). The insidious number two. Trends in Cognitive Sciences, 22, 668–669. http://dx.doi.org/10.1016/ j.tics.2018.05.005 Mertens, G., Van Dessel, P., & De Houwer, J. (2018). The contextual malleability of approachavoidance training effects: Approaching or avoiding fear conditioned stimuli modulates effects of approach-avoidance training. Cognition and Emotion, 32, 341–349. http://dx.doi.org/ 10.1080/02699931.2017.1308315 Mitchell, C. J., De Houwer, J., & Lovibond, P. F. (2009). The propositional nature of human

176

ROLAND DEUTSCH AND FRITZ STRACK

associative learning. Behavioral and Brain Sciences, 32, 183–198. http://dx.doi.org/10.1017/ S0140525X09000855 Moors, A. (2014). Examining the mapping problem in dual-process models. In J. W. Sherman, B. Gawronski, & Y. Trope (Eds.), Dual-Process Theories of the Social Mind (pp. 20–34). New York: Guilford Press. Mühlig, S., Paulick, J., Lindenmeyer, J., Rinck, M., Cina, R., & Wiers, R. W. (2016). Applying the “cognitive bias modification” concept to smoking cessation: A systematic review. Sucht 62, 333–354. http://dx.doi.org/10.1024/09395911/a000454 Nederkoorn, C., Guerrieri, R., Havermans, R. C., Roefs, A., & Jansen, A. (2009). The interactive effect of hunger and impulsivity on food intake and purchase in a virtual supermarket. International Journal of Obesity, 33, 905–912. http://dx.doi.org/10.1038/ijo.2009.98 Nisbett, R. E., & Kanouse, D. E. (1969). Obesity, food deprivation, and supermarket shopping behavior. Journal of Personality and Social Psychology, 12, 289–294. http://dx.doi.org/10.1037/h0027799 Orquin, J. L., & Kurzban, R. (2016). A meta-analysis of blood glucose effects on human decision making. Psychological Bulletin, 142, 546–567. http://dx .doi.org/10.1037/bul0000035 Ouellette, J. A., & Wood, W. (1998). Habit and intention in everyday life: The multiple processes by which past behavior predicts future behavior. Psychological Bulletin, 124, 54–74. http://dx.doi .org/10.1037/0033-2909.124.1.54 Petty, R. E., & Cacioppo, J. T. (1986). The elaboration likelihood model of persuasion. In L. Berkowitz (Ed.), Advances in Experimental Social Psychology, Vol. 19 (pp. 123–205). New York: Academic Press. Polivy, J., Coleman, J., & Herman, C. P. (2005). The effect of deprivation on food cravings and eating behavior in restrained and unrestrained eaters. International Journal of Eating Disorders, 38, 301–309. http://dx.doi.org/ 10.1002/eat.20195 Radel, R., & Clément-Guillotin, C. (2012). Evidence of motivational influences in early visual perception: Hunger modulates conscious access.

Psychological Science, 23, 232–234. http://dx.doi .org/10.1177/0956797611427920 Rinck, M., Wiers, R. W., Becker, E. S., & Lindenmeyer, J. (2018). Relapse prevention in abstinent alcoholics by cognitive bias modification: Clinical effects of combining approach bias modification and attention bias modification. Journal of Consulting and Clinical Psychology, 86, 1005– 1016. http://dx.doi.org/10.1037/ccp0000321 Seibt, B., Häfner, M., & Deutsch, R. (2007). Prepared to eat: How immediate affective and motivational responses to food cues are influenced by food deprivation. European Journal of Social Psychology, 37, 359–379. http://dx.doi.org/ 10.1002/ejsp.365 Shaw, J. A., Forman, E. M., Espel, H. M. et al. (2016). Can evaluative conditioning decrease soft drink consumption? Appetite, 105, 60–70. http://dx.doi .org/10.1016/j.appet.2016.05.016 Sherman, J. W. (2006). On building a better process model: It’s not only how many, but which ones and by which means? Psychological Inquiry, 17, 173–184. http://dx.doi.org/10.1207/s15327965 pli1703_3 Stockburger, J., Schmälzle, R., Flaisch, T., Bublatzky, F., & Schupp, H. T. (2009). The impact of hunger on food cue processing: An event-related brain potential study. NeuroImage, 47, 1819–1829. http://dx.doi.org/10.1016/j .neuroimage.2009.04.071 Strack, F., & Deutsch, R. (2004). Reflective and impulsive determinants of social behavior. Personality and Social Psychology Review, 8, 220–247. http://dx.doi.org/10.1207/ s15327957pspr0803_1 Strack, F., & Deutsch, R. (2015). The duality of everyday life: Dual-process and dual system models in social psychology. In M. Mikulincer, P. R. Shaver, E. Borgida, & J. A. Bargh (Eds.), APA Handbook of Personality and Social Psychology, Vol. 1: Attitudes and Social Cognition (pp. 891–927). Washington, DC: American Psychological Association. http:// dx.doi.org/10.1037/14341-028 Tello, N., Bocage-Barthélémy, Y., Dandaba, M., Jaafari, N., & Chatard, A. (2018). Evaluative conditioning: A brief computer-delivered

Changing Behavior Using the Reflective-Impulsive Model

intervention to reduce college student drinking. Addictive Behaviors, 82, 14–18. http://dx.doi.org/ 10.1016/j.addbeh.2018.02.018 Vallacher, R. R., & Wegner, D. M. (1987). What do people think they’re doing? Action identification and human behavior. Psychological Review, 94, 3–15. http://dx.doi.org/10.1037/0033-295X.94.1.3 Van Dessel, P., Hughes, S., & De Houwer, J. (2018). Consequence-based approach-avoidance training: A new and improved method for changing behavior. Psychological Science, 29, 1899–1910. http://dx.doi.org/10.1177/0956797618796478 Wang, D., Hagger, M. S., & Chatzisarantis, N. L. D. (2020). Ironic effects of thought suppression: A meta-analysis. Perspectives on Psychological Science. https://doi.org/10.1177/1745691619898795 Wegner, D. M. (1994). Ironic processes of mental control. Psychological Review, 101, 34–52. http:// dx.doi.org/10.1037/0033-295X.101.1.34 Wegner, D. M., Ansfield, M., & Pilloff, D. (1998). The putt and the pendulum: Ironic effects of the mental control of action. Psychological Science, 9, 196–199. http:// dx.doi.org/10.1111/1467-9280.00037

177

Wiers, R. W., Rinck, M., Kordts, R., Houben, K., & Strack, F. (2010). Retraining automatic action-tendencies to approach alcohol in hazardous drinkers. Addiction, 105, 279–287. http://dx.doi.org/10.1111/j.1360-0443 .2009.02775.x Wispé, L. G., & Drambarean, N. C. (1953). Physiological need, word frequency, and visual duration thresholds. Journal of Experimental Psychology, 46, 25–31. http://dx.doi.org/10.1037/ h0054454 Wiswede, D., Koranyi, N., Müller, F., Langner, O., & Rothermund, K. (2013). Validating the truth of propositions: Behavioral and ERP indicators of truth evaluation processes. Social Cognitive and Affective Neuroscience, 8, 647–653. http://dx.doi .org/10.1093/scan/nss042 Wittekind, C. E., Feist, A., Schneider, B. C., Moritz, S., & Fritzsche, A. (2015). The approach-avoidance task as an online intervention in cigarette smoking: A pilot study. Journal of Behavior Therapy and Experimental Psychiatry, 46, 115–120.

13 Changing Behavior Using Habit Theory Sheina Orbell and Bas Verplanken

Practical Summary Habits permit fluent action, preserve cognitive resources for other tasks (e.g., thinking about the day’s tasks at work while driving to work), and ensure action is not forgotten. The utility of a habit lies in the mechanism by which it directs behavior. A habit is formed when behavior is repeatedly performed in a consistent context. After many repetitions, mental representations of habitual action are formed that are automatically activated by context cues, prompts, or events in the environment that serve to “trigger” the behavior or action. Forming a habit is useful in any context where consistent regular performance of an action is desirable. Interventions to create new desired habits rely on consistent pairing of a new behavior with a consistent cue. Any recurring feature of a performance context, such as physical location (e.g., in the bathroom), completion of a previous task (e.g., arrival home from work), and previous action in a scripted sequence (e.g., getting dressed), might come to serve as a habit cue. Interventions to disrupt undesired habits rely on removing the cue, taking advantage of natural discontinuities in the cue context, ignoring the cue, effortful inhibition of the habituated response to the cue, or substitution of a new response to the cue.

13.1 Introduction The habit construct is one of the most enduring in the history of psychology, revisited by psychologists even as dominant paradigms and theoretical perspectives on psychological science shifted over time. For example, William James was an advocate of the necessity and efficiency of habit in everyday life: There is no more miserable human being than one in whom nothing is habitual but indecision. Full half the time of such a man [sic] goes to the deciding or regretting of matters which ought to have been so thoroughly ingrained in him as practically not to exist for his consciousness at all. (James, 1887, p. 447)

James was such an admirer of habits that he proposed that young people be advised to

form as many useful habits as possible. His insights that habits become “ingrained” and occur outside of conscious deliberation are cornerstones of modern theorizing about habit. However, the scientific journey took some years and it was not until 2016 that habit first appeared as an entry in the Annual Review of Psychology (Wood & Rünger, 2016). During the early decades of the twentieth century, psychology was dominated by behaviorism and the view that humans learn stimulus-response associations wherein their repeated actions are controlled by external

https://doi.org/10.1017/9781108677318.013

Changing Behavior Using Habit Theory

factors (Hull, 1943; Skinner, 1938), setting the scene for the idea that habits are externally cued. Behaviorism soon gave way to a view that cognitive factors play an important part in human behavior, so that actions are controlled by (explicitly expressed) mental representations of goals and expectations, exemplified by the development of models such as the theory of reasoned action/planned behavior (Fishbein & Ajzen, 1975, 2011; see Chapter 2, this volume). These and related accounts within the social cognitive tradition, with some exceptions, such as Triandis (1977), viewed habit as redundant. Nonetheless, these theories could not readily accommodate observations that behavior repeated often in the past seemed to direct action more reliably than expressed intentions (Eagly & Chaiken, 1993), prompting renewed interest in the habit construct. Time series or ecological momentary assessment studies confirmed that some 43 percent of daily activity was repeated in the same context and often while people were thinking about something other than the action they were performing, implying a lack of conscious intent (Wood, Quinn, & Kashy, 2002). Further scientific developments in cognitive neuroscience drew attention to distinctive patterns of neural activity associated with habit learning versus planning and deliberative thought. While task performance initially involves activation in the associative loop, with repeated practice activation increases in the sensorimotor loop (Tricomi, Balleine, & Doherty, 2009; Yin & Knowlton, 2006). Together, these developments paved the way to current theorizing about habit. Additional impetus for the development of theory of habit arose from more practical considerations. While motives and expectancies may change behavior briefly, the impact of motivation on sustained behavior change is substantially diminished by regression to previous habitual behavior (Ouellette & Wood, 1998; Webb & Sheeran, 2006). The flipside of this coin is that

179

long-term success of behavior change interventions often requires turning the new behavior into a habit: the very features that make habits difficult to change are desirable for new behaviors to acquire (e.g., Gardner & Lally, 2018; Lally et al., 2010). By the twenty-first century, it was becoming clear that fresh approaches were required to address the difficulties of modifying existing habits and to developing new behavioral habits that would be sustained over time. In the following sections, a contemporary definition and theoretical account of habit is provided (Section 13.2). In addition, empirical evidence concerning mechanisms of habit development and operation are outlined. Consideration is given to how habit theory informs behavior change interventions that (1) create habits and (2) undo habits (Section 13.3). The final section provides an overview of current evidence regarding habit theory–based interventions (Section 13.4).

13.2 Overview of Theory and Evidence Contemporary accounts of habit converge on three key elements (Figure 13.1). A history of action repetition in a consistent cue context slowly results in the formation of a cue-response association in memory. Once formed, the cueresponse association in memory is automatically activated by the context cue when it is encountered, so that action becomes cue-contingent (Orbell & Verplanken, 2010; Wood & Rünger, 2016). Thus, as James anticipated, once a habit is acquired, it is “ingrained.” Habits are ways individuals’ neural networks “remember” recurring contexts and optimal responses to them. They contribute to the experience of continuity in daily life because they render every small decision, such as how to shower or dress, make a cup of tea, what to say, or where to go, a nondecision. This ease of activity in turn contributes, if not exactly to save people from the misery of

180

SHEINA ORBELL AND BAS VERPLANKEN

mundane decision-making, at least to a sense of fluency that stands in contrast to the effortful and tiring process of life experienced in a novel context without habits, such as might be experienced after moving house (Verplanken, 2018). A further feature of habit is goal independence. Although goals and motivations may sometimes play a role in the development of habit, the operation of habit is not sensitive to the existence of reward or fluctuating motivation. Empirical evidence for these features of the development and operation of habit is considered in the following sections.

13.2.1 Habits Are Cue-Contingent Habits are acquired via behavioral repetition in specific cue contexts. Over time, an association is forged in memory between cue and action

(Figure 13.1). Any recurring feature of a performance context, such as physical location (e.g., in the bathroom), completion of a previous task (e. g., arrival home from work), and previous action in a scripted sequence (e.g., getting dressed), might come to serve as a habit cue. Evidence for the context-dependent nature of habits comes from observations of habit discontinuities, where a move to a new location, for example, disrupts previous habits (e.g., Verplanken et al., 2008). Conversely, behavioral slips occur when a context cues unintended action (e.g., Orbell & Verplanken, 2010, Study 2). Habit cues are readily detected in the environment (Orbell & Verplanken, 2010, Study 1) and research has shown that context cues are associated with shorter response latencies to habitual actions in a lexical decision task (Neal et al., 2012).

Response latencies to cues and the cue-action pairs

Repeated performance co-occurs with particular context/cues

Mental association between cue and action is formed

Reward/ reinforcement may be present originally

Figure 13.1 The development of habit

When cue is encountered, action follows swiftly with little concious awareness and is insensitive to reward and goal independent

Changing Behavior Using Habit Theory

13.2.2 Habits Are Goal-Independent The relationship of goals to habits has been the subject of debate (Orbell & Verplanken, 2018; Wood & Rünger, 2016). Nonetheless evidence from learning paradigms, studies of behavior change, and cognitive neuroscience all contribute to the conclusion that established habits operate independently of motivational state. People often continue to eat beyond satiety, watch television after interest has waned, or buy products they do not need. Animal learning paradigms derived from behaviorist principles define the acquisition of habit in terms of reward insensitivity. For example, extended training at a task such as lever pushing for a reward results in habitual behavior that persists even after the reward is withdrawn or rendered nasty (Adams, 1982). Relatedly, brain systems activated during the performance of cue-response habits is localized in the sensorimotor loop, concerned with motor control, whereas the control of goal-directed actions is localized in brain regions in the associative loop (Tricomi, Balleine, & Doherty, 2009; Yin & Knowlton, 2006). These neural systems operate in competition, so that, during habit performance, systems associated with deliberation, planning, or executive control are suppressed. In fact, habits tend to prevent flexible and creative responding (Verplanken, Aarts, & van Knippenberg, 1997). One of the few experimental studies that manipulated both cue context and reward value to demonstrate the independence of motives and habit was conducted by Neal et al. (2011). The authors manipulated the motive to eat popcorn (fresh vs. stale) and the popcorneating cue context (cinema vs. meeting room) while participants with weak or strong cinema popcorn–eating habits ostensibly rated film clips. Participants with strong popcorn-eating habits ate equal quantities of stale or fresh popcorn but only in the cinema context, demonstrating that habit was cued by context, regardless of the reward value of the popcorn. In an alternative

181

context, consumption was moderated by taste, even among those with strong habits. Relatedly, insensitivity to changes in monetary incentives led people to continue to make the same habitual responses in a game (Gillan et al., 2015). There are, however, several ways in which habits and goals may be confabulated (Orbell & Verplanken, 2018; Wood & Rünger, 2016). Many habits have their origins in goal-directed action, and habits may continue to serve those goals, even without conscious awareness. Habits are often acquired as a consequence of intended behavior in the past that has been repeatedly performed in stable cue contexts. For example, the desire for something sweet may lead an individual initially to eat a biscuit with a cup of tea but, after many repetitions, reaching for a biscuit will acquire the characteristics of habit and operate irrespective of hunger. The fluency and ease of action afforded by habits are experienced as pleasant and this may lead people to infer that they intend to act (“I always do it so I must like it”). Habits may also be revealed by the invariance of choices or behaviors across contexts. For instance, Verplanken et al. (1994) presented participants with a series of hypothetical travel destinations and asked them to mention as quickly as possible which mode of travel they would use. The assumption was that the time pressure would elicit schematic, rather than intention-based, responses, which would thus represent the person’s general travel mode habit. The true relationship of habit to declarative intention may also be obscured if people make goal inferences for their habits. Adriaanse et al. (2018) showed that people confabulate, that is, make up reasons for their unexplained behavior, without the intent to deceive and without knowing that the claim is ill-grounded, when induced to behave, without conscious awareness, in ways that are inconsistent with current goals or values. Inference may also stretch to instances where an undesired habit, such as eating chocolate biscuits when watching television, is cued and runs off

182

SHEINA ORBELL AND BAS VERPLANKEN

smoothly, in contradiction to an intention to diet (e.g., Verplanken & Faes, 1999). In these instances, a lack of awareness of the cue-contingency might lead an individual to incorrectly infer some other cause, such as stress at work, and consequently fail to gain traction on their actual food habit cues.

13.2.3 Habits Share Features of Automaticity The cue-response associations formed in memory that are characteristic of habit are generally considered to be activated automatically on detection of the cue context and, in many cases, action itself follows automatically from cue detection. For example, people often report: “I started doing it [the habitual behavior] before I realized I was doing it” (Orbell & Verplanken, 2010; Verplanken & Orbell, 2003) or make behavioral

slips when action flows even in contradiction to intention (see Sidebar 13.1). It is important to distinguish automatic habits from other forms of automatic behavior. Habit automaticity refers to an instance of direct cueing, wherein, on perception of the cue stimulus, an individual directly retrieves the associated action response from memory in a single-step memoryretrieval process (Logan, 1988), without mediation by reflective processes (Neal et al., 2012). Other forms of automatic responding, such as automatic goal pursuit, concept priming, implicit attitudes, schemas, attributions, or first impressions share the properties of lack of awareness and reflection but differ in that they assume spreading activation of knowledge structures in memory (Aarts, 2007; Bargh, 2006). Consequently, activation of a goal may activate a wide range of potential goal-related behaviors, or activation of an implicit attitude will activate a

Sidebar 13.1 Habits cause “behavioral slips”

Strong behavioral habits are provoked by context cues without an individual intending or realizing that action has been initiated. These unwitting behaviors can be referred to as “action slips.” Orbell and Verplanken (2010) illustrated this phenomenon in a field study. In 2007, England introduced a law that banned smoking in public places, including inside pubs (public houses licensed to sell alcohol). Orbell and Verplanken (2010) assessed the drinking and smoking habits of people in pubs three months before and three months after the introduction of the legislation. Habit strength to smoke when drinking alcohol in a pub was assessed three months before the ban using the self-report habit index (SRHI; Verplanken & Orbell, 2003; see also Sidebar 13.4). Following the ban, smokers were asked if they had ever made a “behavioral slip,” for example by taking out a cigarette, putting it in their mouth, or lighting or nearly lighting it. Forty percent of smokers indicated that this was true – for example, reporting that “it had happened when I was preoccupied talking to someone.” Other examples, such as “I rolled one ready to go outside but then accidently lit it,” illustrate the power of scripted habit sequences. The person reporting this experience believed that they could interrupt the process of rolling and then lighting a cigarette but, in fact, once the scripted sequence was begun, action was automatically cued to the end of the sequence and the cigarette was lit indoors. Habit strength measured before the ban predicted the likelihood of making such a behavioral slip, even after controlling for typical alcohol consumption.

Changing Behavior Using Habit Theory

range of affective attributes associated with the attitude object. In contrast, habits involve a direct cue-behavior association in which context cues a single, specific well-learned behavioral response with a history of repetition (Verplanken & Orbell, 2003; Wood & Rünger, 2016; see Figure 13.1).

13.3 How Has Habit Theory Been Used to Change Behavior? 13.3.1 Theoretical Bases of Habit Creation Habits, by their nature, have the capacity to ensure that new desired behaviors will be enacted consistently and not forgotten (Sidebar 13.2). However, new habits develop relatively slowly (e.g., the range of time it took to

183

characterize a frequently performed behavior as “habit” in one study was 18–254 days; Lally et al., 2010) and require a specific set of circumstances be met (e.g., Gardner & Lally, 2018; Lally & Gardner, 2013) that are conducive to habit formation. First, a specific behavior needs to be identified and learned, so that it can be performed easily, without friction and in an enabling environment. If the behavior requires materials (dental floss to undertake flossing, fruit in the house to promote fruit eating, a bicycle with pumped up tires to facilitate cycling to work) these need to be available. Second, a critical cue for action needs to be identified. Third, a plan or commitment to perform the critical action when the critical cue is encountered must be formed and enacted in a consistent manner over a period of time.

Sidebar 13.2 The measurement of habit strength

Measuring the strength of habits is inherently difficult: Since most habits occur more or less automatically, in response to cues that are automatically detected, people may not have much insight in these processes. The currently available instruments can roughly be divided into two categories. The first are measures that are based on habit-related observations. The most direct way is to observe behavior and identify features of habit (e.g., George et al., 2017). Other observation methods capitalize on consequences of habituation, such as indicated by response latencies (e.g., Neal et al., 2012), attentional bias (e.g., Orbell & Verplanken, 2010), action slips and reward devaluation (e.g., de Wit et al., 2012), response frequency under time pressure (e.g., Verplanken et al., 1994), or neuroimaging (e.g., Lehéricy et al., 2005). In many research contexts, not in the least for practical reasons, habit strength is assessed by means of self-report instruments. This has often been done by measures of past behavior. A variant that is more focused on habit is the frequency-in-context measure (e.g., Ji & Wood, 2007), which combines self-reported frequency of past behavior with indications of the stability of the context in which habits are triggered. The most prevalent measure of habit strength to date is the SRHI (Verplanken & Orbell, 2003). The SRHI is a twelve-item generic instrument that breaks down the habit construct in facets, such as the experience of repetition, lack of awareness, lack of control, and efficiency. The instrument thus capitalizes on reporting on the experience of a habit, such as the absence of thinking and deliberation. It has been demonstrated to have excellent psychometric properties and has been applied in a wide variety of behavioral contexts.

184

SHEINA ORBELL AND BAS VERPLANKEN

Habit development is facilitated by identification of cues that are salient and perceptible environmental features rather than those that require conscious attention or interpretation (Verplanken, 2006). For example, event-based cues such as “as soon as I arrive home” or “after my evening meal” are more perceptible than time-based cues (“at midday”) that require conscious monitoring (McDaniel & Einstein, 2000). Many tasks performed in the course of a daily routine may be considered behavioral scripts or well-learned sequences of actions that can be represented and chunked into units. For example, making a cup of coffee may comprise a sequence of actions, including filling the kettle, switching it on, reaching for a cup, putting coffee in the cup, and so on. These behavioral scripts also have task boundaries, and a theoretical question for habit formation concerns where best to place a new habit. Placement in the middle of a script requires conscious attention to the steps in a likely automated habit sequence and increases the likelihood that it will be missed (e.g., I will take a pill while making my cup of coffee), whereas inserting a new action at the end of a script may be a readily perceived small task boundary (“when I drink the coffee, I will take a pill”) (Gardner & Lally, 2018). The cues for action must also occur at the same level of frequency as the desired habit.

13.3.2 Theoretical Bases of Undoing Habits Habits “exist” as mental representations of cueaction pairings in memory. Neuroscientific evidence that different neural networks control habits and goal-directed behavior poses a fundamental difficulty for undoing habits (Graybiel & Smith, 2014) because it seems that the “imprinted” cue-response associations and scripts associated with established habits cannot be deleted. It is no easier to avoid a habituated action coming to mind than it is to avoid reading a word (in one’s own language) when one sees it.

So, if one manages to establish a new habit, at least for a period of time, the old habit still exists (e.g., Walker, Thomas, & Verplanken, 2015) and may reactivate, leading to relapses. Habits are a powerful source of behavioral change resistance. Consequently, theory around changing habit focuses on either preventing activation of the stored cue-response association in memory by limiting exposure to the cue or preventing enactment of the habitual response by ignoring the cue, willful response inhibition, or response substitution (see Figure 13.2). A theoretical prerequisite for both of these sets of process is the identification of critical cues. People are often unaware of the cues with which their habits are associated in memory. One way to address this is via preliminary daily monitoring of habit occurrences and contexts (Verhoeven et al., 2014).

13.3.2.1 Preventing Activation in Memory The execution of a habit is contingent on cues in the behavioral context that trigger the habitual response. Removal of the self from these cues, or the cues from the environment, will, theoretically, prevent occurrence of the habit. However, this is rarely straightforward to achieve. One approach to habit change is to identify moments and situations when such contexts are naturally disrupted or in the process of changing. This might imply a suspension of the habitual response. Such habit discontinuities may take place in various ways (Verplanken, Roy, & Whitmarsh, 2018). One class of events are those where the individual is confronted with an environment that changes. This may, for instance, concern infrastructure (e.g., motorway closures, the relocation of a company, a public transport strike) but may also happen as the result of legislation or large-scale episodes such as an economic crisis. Under those circumstances, the individual may no longer be able to engage in a particular habitual response and has to find alternative courses of action. Another class of events is when the individual changes context. Such changes may

Changing Behavior Using Habit Theory

Limit exposure to cues

Repeated performance co-occurs with particular context/cues

Ignore cue or substitute new action in response to cue

Retrain mental associations

Mental association between cue and action is formed

Habit discontinuity hypothesis: Take advantage of new contexts to learn new habit

185

When cue is encountered, action follows swiftly with little conscious awareness and is insensitive to reward and goal independent

Use monitoring and effortful inhibition to limit responding

Figure 13.2 Strategies to inhibit habit

come with major life course changes, such as school-to-work transitions, residential relocation, starting a family, a sudden illness, or retirement. Again, an individual’s habits are disrupted and one has to negotiate alternative courses of action. Some changes play out over longer time periods and individuals may have an opportunity to prepare for it. Other events may happen more suddenly, in which case the disruption and the need for alternatives are immediate. The disruption of old habits implies that the automatic links between cues and habitual responses are suspended, a process that the social psychologist Kurt Lewin described more than seventy years ago as “unfreezing” (Lewin, 1947). This may have a number of consequences. One is that, as an individual has to find alternatives for an

old habit, one has to adopt a more deliberative mindset (e.g., Gollwitzer, Heckhausen, & Steller, 1990). A habitual mindset is characterized by a degree of “tunnel vision,” that is, a lack of attention to and interest for new information (e.g., Verplanken, Aarts, & van Knippenberg, 1997). In a deliberate mindset, individuals look out for useful information, weigh alternatives against each other, and, in general, engage in a more deliberate decision process. Thus, from a perspective of a change agent, providing that information as part of an intervention may be more effective under deliberate mindset than habitual mindset conditions. A second consequence is that, as habit discontinuities enhance deeper information processing, decision makers are more likely to rely on important goals or values when considering alternative courses of

186

SHEINA ORBELL AND BAS VERPLANKEN

action. For instance, Verplanken et al. (2008) found that environmentally concerned commuters who had moved house in the last year were more likely to commute sustainably compared to environmentally concerned commuters who had not moved house. Finally, a habit discontinuity episode is time-limited. Once the discontinuity has occurred, new habits may establish or old habits may resurface. In other words, the opportunity to influence behavior by capitalizing on habit discontinuities has a limited shelf life. For instance, Verplanken and Roy (2016) found that such a “window of opportunity” only lasted approximately three months (see Sidebar 13.3).

13.3.2.2 Preventing Enactment of the Habit Response Encountering a cue context will automatically activate the associated habit response in memory. However, an opportunity exists to disrupt the execution of habit via conscious control. Constant and demanding goal self-regulation, that is itself ego depleting, may be required to combat strong habits. One approach that might support goals that are counter-habitual is the formation of an implementation intention (Gollwitzer, 1993, 1999; Chapter 6, this volume). Implementation intentions are consciously implemented, goal-directed, self-regulatory strategies

Sidebar 13.3 A field experiment testing the habit discontinuity hypothesis

Verplanken and Roy (2016) tested the habit discontinuity hypothesis in a field experiment, which was conducted in the city of Peterborough (UK): 400 hundred households who had moved house in the previous six months were matched with 400 control households who had not moved but were comparable on a number of key characteristics. In each group, half of the households were given an intervention aimed at promoting sustainable behaviors with respect to water and energy use, transportation, and waste, while the other half served as a control group. Self-reports of twenty-five behaviors were collected at baseline and eight weeks later. In addition, a host of determinants of behavior were assessed at baseline, such as habit strength, intentions, involvement, and values. The intervention consisted of a suite of initiatives, most importantly a face-to-face conversation about which, if any, behaviors a person was willing and able to change and providing bespoke information about those behaviors. Results showed that the intervention was effective in increasing the level of sustainable behaviors, after controlling for the behaviors at baseline and the set of behavioral determinants. Importantly, though, this was only found among those who had relocated. A closer inspection of these results revealed that the intervention effects could only be detected among those who had moved house in the previous three months. This suggests that, in the present circumstances, the “window of opportunity” provided by the habit discontinuity lasted three months after the event that caused the discontinuity. The study was unable to determine when the “window of opportunity” opened. In the case of life course changes such as moving house, starting a family, or retirement, people start preparing for the upcoming change. This suggests that “windows of opportunity” may actually open some time before the discontinuity actually takes place.

Changing Behavior Using Habit Theory

that supplement goal intentions (“I intend to do ‘Z’”). They take the form “If I encounter context/ cue ‘X’ then I will perform behavior ‘Y’” in the service of the goal-directed behavior. These ifthen plans link context opportunities to action so that opportunity to act is not missed. In the context of overcoming habits, so long as an individual can correctly identify the existing cue for their unwanted behavior, it may be possible to form an implementation intention to replace an unwanted habit (habit substitution; eating crisps when watching television) with a plan to eat fruit when watching television, for example. Studies that have explored the utility of these strategies for overcoming habits show that a plan to respond to a cue in a different manner than prescribed by habit creates an opportunity for conscious control of action but does not make the novel response more accessible in memory than the old one (Adriaanse et al., 2011). Plans to respond to cues by attempting to negate a habit cue–response (“If I encounter stimulus ‘X’, I will tell myself not to do ‘Y’”) either reactivate prior associations between cue and response “Y” or maintain perceptual readiness to perceive a habit cue and are consequently ineffective. Plans to ignore cues (“If I encounter cue ‘X’, I will ignore it”) may be more effective in breaking habits but little research has been conducted in samples where evidence has been provided of prior strong habits or with adequate follow-up to substantiate effects (Adriaanse & Verhoeven, 2018; see Figure 13.2). A novel approach to tackling undesired habits might be afforded by training new habits that rely less on the mobilization of conscious goal regulation but employ habit architecture to acquire new habits. Just as research is beginning to suggest ways in which environmental primes might be employed to create choice architecture that “nudges” people to enact their goals (e.g., Marteau, Hollands, & Kelly, 2015; see Chapter 14, this volume), so “habit architecture” may be employed to nudge people to enact new habits. A new habit can simply override an old one. For

187

example, developing a new habit to go to the gym after work can effectively inhibit the old habit of going to the pub after work. Or a new habit of walking to work can effectively inhibit an old habit of driving to work (Lin, Wood, & Monterosso, 2015).

13.4 Evidence Base for the Use of Habit Theory in Changing Behavior Modern habit research is in its relative infancy and has, to date, focused primarily on establishing the nature and operation of habit and its boundary conditions. Habit theory–based behavior change interventions are relatively few (Gardner & Lally, 2018; see Chapter 41, this volume). Habits take time to establish, requiring intervention follow-ups that are sufficient to capture the asymptotic growth of automatic association in memory and to demonstrate the true capacity of interventions to establish enduring behavior change. Orbell and Verplanken (2010, Study 3) provided the first evidence that an implementation intention plan to install a dental flossing habit in a particular cue context increased not only behavioral frequency but also habit strength over a four-week period. Subsequent studies (Beeken et al., 2017; Kaushal & Rhodes, 2015; Judah, Gardner, & Aunger, 2013) have reported similar effects. Gardner and Rebar (2019) conducted a narrative review of fifteen studies in which critical components of habit theory, particularly cue-context repetition and action planning, were included in interventions to modify diet, physical activity, and dental and food hygiene behaviors. The authors concluded that, although habit interventions were acceptable to participants and showed promising effects, the superiority of habit-based interventions versus active, motivationally based controls is not yet established and the power of habit-based interventions to establish long-term sustainable behavior change remains untested.

188

SHEINA ORBELL AND BAS VERPLANKEN

Sidebar 13.4 Using the self-report habit index to measure cue-contingent habit strength

The self-report habit index (SRHI) can and has been used extensively without reference to a cue. However, its use might be advanced by the inclusion of a cue context, if a researcher wishes to track the development of a habit in a habit formation study or establish the strength of a habit prior to an intervention to diminish cue-contingent responding, for example. In that case, the first step is to identify the relevant cue context. The cue context might be a location, a visual cue, or a preceding or concurrent action (see Section 13.3). Habit strength may be assessed with respect to a particular action (“X”) and a particular cue context (“Y”): The cue context is referred to here by the term (Situation “Y”). The SRHI (Verplanken & Orbell, 2003; Sidebar 13.2) may be adapted to serve this purpose as follows: 1. Behavior “X” is something I do frequently (in Situation “Y”). 2. Behavior “X” is something I do automatically (in Situation “Y”). 3. Behavior “X” is something I do (in Situation “Y”) without having to consciously remember. 4. If I do not do Behavior “X” when (in Situation “Y”) this makes me feel weird. 5. Behavior “X” is something I do (in Situation “Y”) without thinking. 6. It would require effort not to do Behavior “X” (in Situation “Y”). 7. Doing Behavior “X” (in Situation “Y”) belongs to my (daily, weekly, monthly) routine. 8. Behavior “X” is something I start doing (in Situation “Y”) before I realize I’m doing it. 9. I would find it hard not to do Behavior “X” (in Situation “Y”). 10. Doing Behavior X (in “Situation Y”) is something I have no need to think about doing. 11. Doing Behavior X (in Situation “Y”) is something that’s typically “me.” 12. Doing behavior X (in Situation “Y”) is something I have been doing for a long time. The scale is assessed on a Likert scale: strongly agree–strongly disagree (Verplanken & Orbell, 2003).

Studies that address the undoing of habits are even scarcer in the published literature. A number of interventions that purport to show behavior change via habit change techniques fail to establish prior strong habits, and some studies that have assessed prior habit have produced null results (Webb, Sheeran, & Luszczynska, 2009), suggesting that strong habits are resilient to

planning intervention. Nonetheless, Verplanken et al. (2016) provide an important critical demonstration of the power of habit discontinuity theory in changing environmental behavior habit in a general population sample. In the clinical literature, habit reversal theory (Azrin & Nunn, 1973), which was developed to treat habits such as hair pulling, nail biting, and skin picking, suggests a

Changing Behavior Using Habit Theory

number of strategies that, together, have clinical effectiveness. Strategies include the development of cue and response awareness via monitoring and the description and detection of early signs that a response is occurring, training an incompatible response, and enhancing motivation for behavior change. A recent meta-analysis suggests that habit reversal therapy is highly effective in reducing these behaviors in these samples, although, it should be noted that, to date, these studies do not establish or measure habit strength according to theoretical principles (Bate et al., 2011; see Sidebars 13.2 and 13.4). Recently, habit reversal theory has been extended to the domain of eating disorders. Steinglass et al. (2018) employed the SRHI (Verplanken & Orbell, 2003) to assess change in behaviors that sustain anorexia and provided the first evidence that a habit-based intervention compared to usual treatment control intervention offers an important new treatment approach to behavioral disorders.

13.5 Conclusion Habit is a powerful determinant of behavior in daily life. Habits are specific cue-action associations in memory that are directly activated when cues are detected. They operate without reference to goals or rewards. Habit interventions are therefore a powerful tool for the development of sustained behavior change. To date, behavior change research has tended to be restricted to small samples and short-term follow-ups, but evidence suggests that interventions that promote repeated action in stable cue contexts will promote habit strength and facilitate enduring behavior change.

References Aarts, H. (2007). Health and goal-directed behaviour: The nonconscious regulation and motivation of goals and their pursuit. Health Psychology Review, 1, 53–82. https://doi.org/10.1080/ 17437190701485852

189

Adams, C. D. (1982). Variations in the sensitivity of instrumental responding to reinforce devaluation. The Quarterly Journal of Experimental Psychology Section B, 34, 77–98. https://doi.org/ 10.1080/14640748208400878 Adriaanse, M., Gollwitzer, P. M., de Ridder, T. D., de Wit, J. B. F., & Kroese, F. M. (2011). Breaking habits with implementation intentions: A test of underlying processes. Personality and Social Psychology Bulletin, 37, 502–513. https://doi.org/ 10.1177/0146167211399102 Adriaanse, M., Kroese, F.M., Weijers, J., Gollwitzer, P. M., & Oettingen, G. (2018). Explaining unexplainable food choices. European Journal of Social Psychology, 48, 15–24. https://doi.org/ 10.1002/ejsp.2273 Adriaanse, M. A., & Verhoeven, A. (2018). Breaking habits using implementation intentions. In B. Verplanken (Ed.), The Psychology of Habit: Theory, Mechanisms, Change, and Contexts (pp. 169–188). Cham: Springer. https://doi.org/ 10.1007/978-3-319-97529-0_10 Azrin, N. H., & Nunn, R. G. (1973). Habit-reversal: A method of eliminating nervous habits and tics. Behaviour Research and Therapy, 11, 619–628. https://doi.org/10.1016/0005-7967(73)90119-8 Bargh, J. A. (2006). What have we been priming all these years? On the development, mechanisms and ecology of nonconscious social behavior. European Journal of Social Psychology, 36, 147– 168. https://doi.org/10.1002/ejsp.336 Bate, K. S., Malouff, J. M., Thorteinsson, E. T., & Bhullar, N. (2011). The efficacy of habit reversal therapy for tics, habit disorders and stuttering: A meta-analytic review. Clinical Psychology Review, 31, 865–871. https://doi.org/10.1016/j .cpr.2011.03.013 Beeken, R. J., Leurent, B., Vickerstaff, V. et al. (2017). A brief intervention for weight control based on habit formation theory delivered through primary care: Results from a raondomised controlled trial. International Journal of Obesity, 41, 246–254. https://doi.org/10.1038/ijo.2016.206 de Wit, S., Watson, P., Harsay, H. A., Cohen, M. X., van de Vijver, I., & Ridderinkhof, K. R. (2012). Corticostriatal connectivity underlies individual differences in the balance between habitual and

190

SHEINA ORBELL AND BAS VERPLANKEN

goal-directed action control. The Journal of Neuroscience, 32, 12066–12075. https://doi.org/ 10.1523/JNEUROSCI.1088-12.2012 Eagly, A. H., & Chaiken, S. (1993). The Psychology of Attitudes. Orlando, FL: Harcourt Brace Jovanovich College. Fishbein, M., & Ajzen, I. (1975). Belief, Attitude, Intention and Behavior: An Introduction to Theory and Research. Reading, MA: Addison-Wesley. Fishbein, M., & Ajzen, I. (2011). Predicting and Changing Behavior: The Reasoned Action Approach. New York: Psychology Press. Gardner, B., & Lally, P. (2018). Modelling habit formation and its determinants. In B. Verplanken (Ed.), The Psychology of Habit: Theory, Mechanisms, Change, and Contexts (pp. 207– 228). Cham: Springer. https://doi.org/10.1007/ 978-3-319-97529-0_12 Gardner, B., & Rebar, A. L. (2019). Habit formation and behavior change. Oxford Bibliographies. https:// doi.org/10.1093/OBO/9780199828340-0232 George, C. M., Biswas, S., Jung, D. et al. (2017). Psychosocial factors mediating the effect of the CHoBI7 intervention on handwashing with soap: A randomized controlled trial. Health Education and Behavior, 44, 613–625. https://doi.org/ 10.1177/1090198116683141 Gillan, C. M., Otto, A. R., Phelps, E. A., & Daw, N. D. (2015). Model-based learning protects against forming habits. Cognitive, Affective and Behavioral Neuroscience, 15, 523–536. DOI.: 10.3758/s13415-015-0347-6 Gollwitzer, P. M. (1999). Implementation intentions: Strong effects of simple plans. American Psychologist, 54, 493–503. https://doi.org/ 10.1037/0003-066X.54.7.493 Gollwitzer, P. M. (1993). Goal achievement: The role of intentions. European Review of Social Psychology, 4, 141–185. Gollwitzer, P. M., Heckhausen, H., & Steller, B. (1990). Deliberative and implemental mind-sets: Cognitive tuning toward congruous thoughts and information. Journal of Personality and Social Psychology, 59, 1119–1127. https://doi.org/ 10.1037/0022-3514.59.6.1119 Graybiel, A. M., & Smith, K. S. (2014). Good habits, bad habits. Scientific American, 310, 38–43.

Hull, C. L. (1943). Principles of Behaviour: An Introduction to Behaviour Theory. Oxford: Appleton-Century. James, W. (1887). The laws of habit. The Popular Science Monthly, 31, 433–451. Ji, M. F., & Wood, W. (2007). Purchase and consumption habits: Not necessarily what you intend. Journal of Consumer Psychology, 17, 261–276. https://doi .org/10.1016/S1057-7408(07)70037-2 Judah, G., Gardner, B., & Aunger, R. (2013). Forming a flossing habit: An exploratory study of the psychological determinants of habit formation. British Journal of Health Psychology, 18, 338– 353. https://doi.org/10.1111/j.20448287.2012.02086.x Kaushal, N., & Rhodes, R. E. (2015). Exercise habit formation in new gym members: A longitudinal study. Journal of Behavioral Medicine, 38, 652– 663. https://doi.org/10.1007/s10865-015-9640-7 Lally, P., & Gardner, B. (2013). Promoting habit formation. Health Psychology Review, 7(sup1), S137-S158. https://doi.org/10.1080/ 17437199.2011.603640 Lally, P., van Jaarsveld, C. H. M., Potts, H. W. W., & Wardle, J. (2010). How are habits formed: Modelling habit formation in the real world. European Journal of Social Psychology, 40, 998– 1009. https://doi.org/10.1002/ejsp.674 Lehéricy, S., Benali, H., Van de Moortele, P. F. et al. (2005). Distinct basal ganglia territories are engaged in early and advanced motor sequence learning. PNAS, 102, 12566–71. https://doi.org/ 10.1073/pnas.0502762102 Lewin K. (1947). Frontiers in group dynamics: Concept, method and reality in social science; social equilibria and social change. Human Relations, 1, 5–41. https://doi.org/10.1177/ 001872674700100103 Lin, P.-Y., Wood, W., & Monterosso, J. (2016). Healthy eating habits protect against temptations. Appetite, 103, 432–440. https://doi.org/10.1016/j .appet.2015.11.011 Logan, G. D. (1988). Toward an instance theory of automatization. Psychological Review, 95, 492– 527. https://doi.org/10.1037/0033-295X.95.4.492 Marteau, T. M., Hollands, G. J., & Kelly, M. P. (2015). Changing population behaviour and reducing

Changing Behavior Using Habit Theory

health disparities: Exploring the potential of “Choice Architecture” interventions. In R. M. Kaplan, M. L. Spittel, & D. H. David (Eds.), Population Health: Behavioral and Social Science Insights (pp. 105–126). AHRQ Publication No. 15–002. Rockville, MD: Agency for healthcare Research Quality and Office of Behavioral and Social Sciences Research, National Institutes of Health. McDaniel, M. A., & Einstein, G. O. (2000). Strategic and automatic processes in prospective memory retrieval: A multiprocess framework. Applied Cognitive Psychology, 14, S127–S144. https://doi .org/10.1002/acp.775 Neal, D. T., Wood, W., Labrecque, J. S., & Lally, P. (2012). How do habits guide behavior? Perceived and actual triggers of habits in daily life. Journal of Experimental Social Psychology, 48, 492–498. https://doi.org/10.1016/j.jesp.2011.10.011 Neal, D. T., Wood, W., Wu, M., & Kurlander, D. (2011). The pull of the past: When do habits persist despite conflict with motives? Personality and Social Psychology Bulletin, 37(11), 1428– 1437. https://doi.org/10.1177/ 0146167211419863 Orbell, S., & Verplanken, B. (2010). The automatic component of habit in health behavior: Habit as cue-contingent automaticity. Health Psychology, 29, 374–383. https://doi.org/10.1037/a0019596 Orbell, S., & Verplanken, B. (2018). Progress and prospects in habit research. In B. Verplanken (Ed.), The Psychology of Habit: Theory, Mechanisms, Change, and Contexts (pp. 397– 410). Cham: Springer. https://doi.org/10.1007/ 978-3-319-97529-0_22 Ouellette, J. A., & Wood, W. (1998). Habit and intention in everyday life: The multiple processes by which past behavior predicts future behavior. Psychological Bulletin, 124, 54–74. https://doi .org/10.1037%2F0033-2909.124.1.54 Skinner, B. F. (1938). The Behavior of Organisms. New York: Appleton-Century-Crofts. Steinglass, J. E., Glasofer, D. R., Walsh, E. et al. (2018). Targeting habits in anorexia nervosa: A proof of concept randomized controlled trial. Psychological Medicine, 48, 2584–2591. https:// doi.org/10.1017/s003329171800020X

191

Triandis, H. C. (1977). Interpersonal Behavior. Pacific Grove, CA: Brooks-Cole. Tricomi, E., Balleine, B. W., & O’Doherty, J. P. (2009). A specific role for posterior dorsolateral striatum in human habit learning. European Journal of Neuroscience, 29, 2225–2232. https://doi.org/10.1111/j.14609568.2009.06796.x Verhoeven, A. A., Adriaanse, M. A., de Vet, E., Fennis, B. M., & de Ridder, D. T. (2014). Identifying the “if” for “if-then” plans: Combining implementation intentions with cue-monitoring targeting unhealthy snacking behaviour. Psychology and Health, 29, 1476–1492. https:// doi.org/10.1080/08870446.2014.950658 Verplanken, B. (2006). Beyond frequency: Habit as mental construct. British Journal of Health Psychology, 45, 639–656. https://doi.org/10.1348/ 014466605X49122 Verplanken, B. (2018). Introduction. In B. Verplanken (Ed.), The Psychology of Habit: Theory, Mechanisms, Change, and Contexts (pp. 1–10). Cham: Springer. https://doi.org/10.1007/978-3319-97529-0_1 Verplanken, B., Aarts, H., & van Knippenberg, A. (1997). Habit, information acquisition, and the process of making travel mode choices. European Journal of Social Psychology, 27, 539–560. https://doi.org/10.1002/(SICI)1099-0992 (199709/10)27:5 Verplanken, B., Aarts, H., van Knippenberg, A., & van Knippenberg, C. (1994). Attitude versus general habit: Antecedents of travel mode choice. Journal of Applied Social Psychology, 24, 285–300. https://doi.org/10.1111/j.1559-1816.1994. tb00583.x Verplanken, B., & Faes, S. (1999). Good intentions, bad habits, and effects of forming implementation intentions on healthy eating. European Journal of Social Psychology, 29, 591–604. https://doi.org/ 10.1002/(SICI)1099-0992(199908/09)29:5/ 63.0.CO;2-H Verplanken, B., & Orbell, S. (2003). Reflections on past behavior: A self-report index of habit strength. Journal of Applied Social Psychology, 33, 1313– 1330. https://doi.org/10.1111/j.1559-1816.2003 .tb01951.x

192

SHEINA ORBELL AND BAS VERPLANKEN

Verplanken, B., & Roy, D. (2016). Empowering interventions to promote sustainable lifestyles: Testing the habit discontinuity hypothesis in a field experiment. Journal of Environmental Psychology, 45, 127–134. https://doi.org/10.1016/ j.jenvp.2015.11.008 Verplanken, B., Roy, D., & Whitmarsh, L. (2018). Cracks in the wall: Habit discontinuities as vehicles for behavior change. In B. Verplanken (Ed.), The Psychology of Habit: Theory, Mechanisms, Change, and Contexts (pp. 189– 206). Cham: Springer. https://doi.org/10.1007/ 978-3-319-97529-0_11 Verplanken, B., Walker, I., Davis, A., & Jurasek, M. (2008). Context change and travel mode choice: Combining the habit discontinuity and selfactivation hypotheses. Journal of Environmental Psychology, 28, 121–127. https://doi.org/10.1016/ j.envp.2007.10.005 Walker, I., Thomas, G. O., & Verplanken, B. (2015). Old habits die hard: Travel habit formation and decay during an office relocation. Environment and Behavior, 47, 1089–1106. https://doi.org/doi .org/10.1177/0013916514549619

Webb, T. L., & Sheeran, P. (2006). Does changing behavioral intentions engender behavior change? A meta-analysis of the experimental evidence. Psychological Bulletin, 132, 249–268. https://doi.org/0.1037/00332909.132.2.249 Webb, T., Sheeran, P., & Luszczynska, A. (2009). Planning to break unwanted habits: Habit strength moderates implementation intention effects on behavior change. British Journal of Social Psychology, 48, 507–523. https://doi.org/10.1348/ 014466608X370591 Wood, W., Quinn, J. M., & Kashy, D. A. (2002). Habits in everyday life: Thought, emotion and action. Journal of Personality and Social Psychology, 83, 1281–1297. https://doi.org/10.1037/00223514.83.6.1281. Wood, W., & Rünger, D. (2016). Psychology of habit. Annual Review of Psychology 67, 289–314. https://doi.org/10.1146/annurev-psych-122414033417 Yin, H. H., & Knowlton, B. J. (2006). The role of the basal ganglia in habit formation. Nature Reviews Neuroscience, 7, 464–476.

14 Changing Behavior by Changing Environments Theresa M. Marteau, Paul C. Fletcher, Gareth J. Hollands, and Marcus R. Munafò

Practical Summary There is a growing recognition that much behavior is influenced by the contexts in which a behavior occurs. This has shifted attention toward less-theorized interventions that involve changing cues – physical, digital, social, and economic – in environments. This chapter focuses on changing cues in small-scale physical environments – sometimes known as choice architecture or nudge interventions. Despite attracting much interest, little is known about how they work. From examining three interventions – increasing the relative availability of healthier food options, reducing the size of glassware, and putting graphic warning labels on food and alcohol products – research suggests that no single theory explains the effects of these interventions. Advancing the science of changing behavior at scale by changing environments requires robustly designed field studies to estimate effect sizes, complemented by laboratory studies testing mechanisms as a basis for optimizing interventions and developing theoretical understanding.

14.1 Introduction Theories adopted from psychology have been highly influential in guiding interventions to change behavior. Social cognition approaches, with their roots in subjective expected utility models (Von Neumann & Morgenstern, 1947) and focus on beliefs and evaluations, have achieved good prediction of intention to engage in a range of behaviors (Fishbein & Ajzen, 2010) but far lower efficacy in predicting actual behavior, particularly for routine and habitual behaviors (Webb & Sheeran, 2006). The growing recognition that such theories – with their emphasis on conscious, intentional control of behavior – omit nonconscious processes that activate and regulate much behavior has led researchers to explore and adopt alternative approaches that provide more precise

prediction (Hollands, Marteau, & Fletcher, 2016; Marteau, Hollands, & Fletcher, 2012). The shift away from social cognition models has coincided with increased application of behavior change interventions that likely tap into nonconscious processes by, for example, altering cues in environments and that have the potential to change behavior at scale. Such interventions are commonly known as choice architecture or nudge interventions (Thaler & Sunstein, 2008). Such behavior change at scale is increasingly recognized as essential to address global priorities such as health and environmental issues. Theoretical approaches to behavior change that focus on nonconscious approaches have been informed by ideas from machine learning and reinforcement learning https://doi.org/10.1017/9781108677318.014

194

` THERESA M. MARTEAU, PAUL C. FLETCHER, GARETH J. HOLLANDS, AND MARCUS R. MUNAFO

theories that are central to current neuroscientific models of learning and decision-making (see Sutton & Barto, 1998). This work has led to an influential distinction between model-free and model-based behavior. This distinction can provide a useful theoretical framework for considering how choice architecture interventions may work. The focus of this chapter is on choice architecture interventions, with the following aims: 1. Describe the evidence for the effectiveness of choice architecture interventions to change behavior. 2. Describe the theoretical basis for choice architecture interventions by considering first their mechanisms of action. 3. Outline the research priorities to better understand the theoretical basis of choice architecture interventions to inform optimally effective behavior change by changing environments.

14.2 Brief Overview of Salient Theory Choice architecture or nudge interventions captured the public imagination as well as the interest of researchers and policy makers with the publication of Thaler and Sunstein’s (2008) book on the subject (see also Chapter 42, this volume). It describes the unobtrusive cues in people’s everyday environments that shape our behavior – often without our awareness – be it the etching of a housefly in a urinal to “improve aim” or painting chevrons on a road to create an illusion of speed to slow drivers. These observations attesting to the sensitivity of humans to cues in their environment seem novel but are actually based on more than a century of psychological theory and observation. Despite the common use of the term nudge theory, nudging may be too broad and general to merit the term “theory.” It is best considered as a general framework for interventions aimed at changing behavior by altering the context in which the

behavior occurs and provides a basis for exploring pertinent mechanisms and their theoretical basis as a means of advancing the science of behavior change. The choice architecture approach can be understood as a form of situationism – a perspective in which behavior is regarded as principally determined by external, environmental factors rather than by personality traits or motives – and recognition of the automatic or nonconscious bases of most behaviors (Bowers, 1973). While situationism provides a broad preexisting framework within which to locate choice architecture interventions, it is not a theory in the formal sense. Choice architecture would also fit within the more precisely specified dual-process models of behavior. Such models conceptualize behavior as regulated by two sets of interacting processes: essentially, nonconscious and conscious (Oreskes & Conway, 2010; see also Chapter 12, this volume). While these models vary in their details, they share core features linking literatures on associative learning (de Wit & Dickinson, 2009), habit (Wood & Neal, 2016), and the neuroscience of reward (O’Doherty, Cockburn, & Pauli, 2017). The concept of dual processes continues to be developed to provide a more nuanced and falsifiable account of behavior operating through both nonconscious and conscious pathways (Finkel, 2014). Such theoretical dichotomies – situation vs. motivation, conscious vs. nonconscious, automatic vs. reflective – are likely oversimplifications (Melnikoff & Bargh, 2018). For example, individuals’ motivational states may interact with the situations in which they find themselves. For instance, the absence of food at a meeting will lead to the strong prediction that no food will be consumed during the meeting but, if food is present, then how much is consumed would likely vary depending on whether the meeting was held immediately before lunch, when hunger levels are likely to be higher, or after lunch, when they are likely to be lower. Regarding dual-process

Changing Behavior by Changing Environments

dichotomies, any behavior can be described at a number of levels involving a range of processes, conscious and nonconscious (Hollands et al., 2016). For example, individuals may be aware of eating a bag of crisps and aware of having seen an advertisement for such crisps but unaware of the causal link between the two. In summary, choice architecture approaches highlight both the role of context as a powerful predictor of behavior and that conscious processing is not necessary for changing behavior, important starting points for promising interventions for changing behavior at scale. There may be an absence of an explicit explanatory framework or theory but there are numerous implicit ties to existing theories of human behavior, as explored in Appendix 14.1.

14.3 Choice Architecture and Behavior Change Choice architecture was developed as a general framework to set out underlying principles that can be applied to real-world problems rather than to comprehensively theorize and delineate the ways in which its principles could be applied to specific situations. Inevitably, this means that subsequent use of the term has lacked precision when applied to the development, evaluation, and synthesis of interventions to change behavior. The consequence of having no agreed definition and consistently applied terminology is particularly problematic for evidence synthesis. Sole prioritization evidence of choice architecture will result in incomplete, at best, and incoherent, at worst, attempts to synthesize existing evidence (Szaszi et al., 2018). While there have been attempts to map the core characteristics of choice architecture interventions, these have tended to focus on broad theoretical principles rather than describing ways in which environments can be altered (e.g., Johnson et al., 2012). In order to highlight specific interventions and their underlying mechanisms, an operational definition is

195

required, applicable to a given intervention context. The typology of interventions in proximal physical micro-environments (TIPPME; Hollands, Bignardi et al., 2017) provides one such definition, proceeding to map out the corresponding conceptual terrain concerning interventions to change behavior in a systematic manner. This was developed specifically to apply to the selection, purchase, and consumption of food, alcohol, and tobacco, although it could be applied to other behaviors. A simplified version of TIPPME is presented in Appendix 14.2. The typology outlines six intervention types or ways to alter either the properties or the placement of objects or stimuli within proximal, sensorily perceptible, physical micro-environments. Placement can be manipulated in terms of whether a given object is present (availability) and where it is located within an environment (position). In turn, the properties of objects present within a given environment can be manipulated in respect to their functionality, presentation, size, and the information available about them. This typology overlaps substantially with common understanding and general principles of choice architecture interventions and is used as the basis for describing interventions in this chapter. Choice architecture remains a useful shorthand term to capture this conceptual basis. For most of the six intervention types, there is a substantial body of relevant research examining their effectiveness in changing behavior, particularly in the health domain (Hollands et al., 2013; Szaszi et al., 2018). A brief summary of evidence for three exemplar choice architecture interventions is provided in the next section to illustrate how they have been, and can be, used. The examples concern interventions to reduce food, alcohol, and tobacco consumption as that is where there is a substantive evidence base. These behaviors remain the predominant focus of these types of interventions (Szaszi et al., 2018). In other domains that concern consumption behaviors, which also involve engagement with specific

196

` THERESA M. MARTEAU, PAUL C. FLETCHER, GARETH J. HOLLANDS, AND MARCUS R. MUNAFO

objects or products and where the evidence has been systematically reviewed, there remains a relative absence of evidence, such as proenvironmental behaviors, including meat reduction (Bianchi et al., 2018) and medication adherence (Hollands, McDermott et al., 2015).

14.3.1 Availability Availability is defined in TIPPME as adding or removing some or all products or objects to increase, decrease, or alter their range, variety, or number. Interventions aimed at increasing the relative availability of healthier food options provide good examples of availability placement to change behavior (Pechey et al., 2019). There is currently a small evidence base for these interventions, suggestive of large potential effects but with considerable uncertainty. Only six studies met the inclusion criteria for a recent Cochrane Review assessing the impact on selection or consumption of altering the availability of food, alcohol, and tobacco products (Hollands et al., 2019). Meta-analysis found that providing fewer options of a targeted food or food category resulted in a moderate reduction in consumption of those foods but with a substantial degree of uncertainty and a large reduction in their selection. This effect was also seen in the first field study to date attempting to isolate the impact of increasing the proportion of lower energy options available across a broad range of food categories in worksite cafeterias (Pechy et al., 2019). This saw a reduction of 7 percent in energy purchased per day from targeted food categories across six cafeterias, with no significant impact on revenue.

14.3.2 Size Size is defined in TIPPME as altering the size or shape of products or objects. An example of an intervention altering this aspect of the typology is changing the size of wine glasses. A small literature has emerged examining the effect of

manipulating wine glass size. In a series of studies using a multiple treatment reversal design, wine glass size was changed in five bars and restaurants in Cambridge, England between 2015 and 2018, with glass capacities ranging from 250 ml to 510 ml. When combining data from bars and restaurants in meta-analyses, there was no evidence of an effect on wine sales of increasing wine glass size, as was the case for only bars. However, for restaurants only, sales increased with increasing glass size (Pilling et al., 2019).

14.3.3 Information According to TIPPME, information is defined as adding, removing, or changing words, symbols, numbers, or pictures that convey information about the product or object or its use. A prominent example of manipulating information in choice architecture interventions is through adding graphic health warnings to food, tobacco, and alcohol products (see also Chapter 34, this volume). Warning labels are currently mandated for use on tobacco packaging in many countries. There is substantial evidence demonstrating their impact on cessation-related behaviors (Hammond, 2011), with evidence indicating labels that produce negative emotions are most effective (Cho et al., 2018). This is demonstrated by a greater effect of graphic pictorial warnings than text-only warnings, with absolute effect sizes varying across studies (Brewer et al., 2016; Hammond, 2011), including in socially and materially deprived groups in whom smoking rates in higher-income countries are generally higher (Thrasher et al., 2012). Uncertainty remains around the types of images that exert the greatest effect, with pictures depicting the most severe adverse consequences of smoking most likely to be believed and judged most effective (Maynard et al., 2018). By contrast, there is an almost complete absence of evidence of the effect of warning

Changing Behavior by Changing Environments

labels beyond tobacco (Clarke et al., 2019). A small number of studies focused on sugar-sweetened beverages suggest that warning labels are promising for changing related behaviors beyond smoking, including selection of a sugary drinks with text-only warning labels (Bollard et al., 2016), and intentions to purchase with pictorial labels (Bollard et al., 2016; Mantzari et al., 2018).

14.4 Mechanisms of Effect In many cases, current mechanistic explanations of choice architecture intervention effects are limited and incomplete. For example, neurobiological explanations of a “plate size” effect on food consumption expressed in terms of interactions between basal ganglia and ventromedial prefrontal cortex, though interesting, would be difficult to take into account when designing such an intervention. Conversely, different competing mechanistic models may be proposed to explain the effect. For example, the reduced selfserving as a result of a smaller plate size could be explained in terms of altered affordance (Osiurak, Rossetti, & Badets, 2017) or the effect might be attributable to the altered reward value of a given portion size (Schultz, 2006), potentially operating outside of awareness. Alternatively, a mixture of both processes may operate simultaneously or the predominant process may be determined by contexts and intrapersonal factors. Importantly, knowing the likely mechanism at this level of description would have different implications for how the intervention was taken forward, under what circumstances it would be likely to work, and how it might be modified by other factors.

14.4.1 The Model-Based/Model-Free Framework The picture is complex and demands a systematic consideration of possible factors contributing to choice architecture interventions. An influential

197

framework that has its roots in the neuroscience of reinforcement learning may provide an appropriate explanatory system for the mechanisms underpinning these types of interventions. It is based on a theoretical consideration of behaviors emerging from a complex bidirectional interaction between an agent and its environment, with the agent receiving information about the current state of the environment and acting accordingly to maximize reward. Crucially, the framework offers a simplified classification of behaviors into model-free and model-based (Daw et al., 2011) and provides a framework indicating specific points at which the agent-environment interaction may be altered (for details, see Appendix 14.3, supplemental materials). A schematic of the model is presented in Figure 14.1. The model comprises an agent – an individual acting in a given environment – with a repertoire of possible actions (A1, A2, A3, . . . An). The overall goal of the agent is to adjust its statedependent actions to maximize its rewards. A given action produces one or more of a set of possible outcomes (O1, O2, O3, . . . On) that are associated, probabilistically, with a reward value and, potentially, a change in the environmental state. Following a given state-action-outcome cycle, the agent may update its probability for actions in a model-free way. Simply put, a high reward following a given action in a given state will strengthen the probability of repeating that action in that state. The agent may also shape its actions over the longer term by using a growing experience of state transitions – model-based learning – enabling it to pursue a longer-term planned strategy that allows for low reward actions to be chosen if they ultimately move the agent into a state that can maximize reward in the longer term. Note that, while this distinction may map loosely on to a conscious/nonconscious distinction, with model-based learning more likely to involve conscious processing, it is by no means necessary that model-free processing is nonconscious and model-based is conscious.

198

` THERESA M. MARTEAU, PAUL C. FLETCHER, GARETH J. HOLLANDS, AND MARCUS R. MUNAFO

Model-Free Learning

*4 Action

*2

*3

Outcome

A1 A2 A3

O1 O2 O3

Reward Value

*5

Agent

*6 *1 An

On

Environmental State

Model-Based Learning

Figure 14.1 Schematic illustration of a reinforcement learning model of behavior indicating possible

points at which interventions may exert their effects Note. 1 = agent’s model; 2 = probabilities of actions; 3 = action-outcome pairing; 4 = range of outcomes; 5 = value of outcome; 6 = environmental state.

The model of agent-environment interaction suggests a framework for considering points at which interactions may exert their effects, although any given intervention may act at multiple points within the model. Each component of the model is outlined in the next sections. *1 Agent’s Model. The agent’s model broadly refers to the degree to which the agent comprehends the regularities of its world in terms of how the states of that world may change as a consequence of its actions. Strategies for enhancing model-based learning and behavior may allow an agent to develop behaviors based on planning around longer-term goals as, for example, may be the case with precommitting to avoiding alcohol consumption in the evening by not purchasing it during the day. Model-based behavior tends to be

flexible, adaptive, and more able to encompass multiple temporal and spatial features of an environment. *2 Probabilities of Actions. Interventions aimed at shifting an agent’s pattern of behaviors by, for example, making certain actions more probable (e.g., glass shape affecting mouth shape or “embouchure” and, in turn, volume consumed of a beverage that the agent intends to drink). Other interventions may more indirectly act at this point by, for example, increasing reward value (*5) thereby enhancing the vigor with which certain actions may be performed. *3 Action-Outcome Pairing. An environmental change that influences the relationship between an action and outcome. For example, changing convenience (e.g., the effort required to acquire a snack by, for example,

Changing Behavior by Changing Environments

increasing its proximity to the agent) can impact on the degree of effort with which an action may must be performed in order to acquire the outcome and thereby increase the chances that it is performed. *4 Range of Outcomes. By controlling the range of possible outcomes, certain actions will be attenuated, either through a change in statedependent responding or, related to *3, by rendering certain actions fruitless. Simply put, if an unhealthy option is just not available, actions that are directed toward obtaining it will diminish. *5 Value of Outcome. Altering the reward value of outcomes is likely to lead to an update in reinforcement learning and an ensuing change in behavior. *6 Environmental State. As well as producing more or less rewarding outcomes, actions also change the environment. Many actions are concatenated in ways that produce a series of state transitions before any reward is forthcoming. Interventions may therefore be aimed at influencing how actions change environmental states. An example of this would be a plate size manipulation in which fewer spoonfuls of a food were required to cover a relative amount of the plate.

14.4.2 Applications of the ModelBased/Model-Free Framework The following sections outline how the framework might be applied in thinking about possible mechanisms by which choice architecture interventions may exert their effects. It is important to recognize that this framework, although it bestows a number of advantages (see Appendix 14.3), does not offer clear-cut definitions of where interventions may be acting. Any given intervention is likely to be operating at several points, and even the seemingly straightforward question of whether a behavior is purely model-

199

free or model-based is unlikely to be answered definitively, given that any particular behavior is likely to show elements of both. However, the value of this framework lies not in its (limited) capacity to parse behaviors into a rigid structure but rather in how it encourages explicit consideration of dynamic adaptive agents whose behaviors are governed by both explicit and implicit attempts to optimize outcomes within complex, changing environments. Moreover, it does so in a way that unites behavioral observations with a growing literature on neurobiology and computational neuroscience that is being developed within the same framework. In doing so, it is expected that it is possible to generate hypotheses about mechanisms and thereby to guide the sorts of complementary laboratory studies that could ultimately provide credible underpinning mechanisms by which choice architecture interventions act.

14.4.2.1 Increasing the Relative Availability of Healthier Food Options There are several points at which an increase in the availability of a healthy food option may exert an effect on selection and consumption (see Figure 14.1). First, at the most basic level, an intervention may act in part by simply making an action more probable (*2 in Figure 14.1). This may relate to the idea of affordance, whereby the design, number, or arrangement of objects may prompt or cue particular actions. In this example, the more product options that are available within a given range, the more likely it is that a person will encounter an option they are willing to select or consume. As well as making an action more likely to be selected than competing alternatives, increased availability enhances the chance that an action will obtain the manipulated outcome (*3 in Figure 14.1). Put simply, if a product is placed more obviously and conveniently within reach, as well as prompting the action, it will enhance the

200

` THERESA M. MARTEAU, PAUL C. FLETCHER, GARETH J. HOLLANDS, AND MARCUS R. MUNAFO

probability that the action will yield the desired outcome (e.g., picking up an apple rather than a chocolate bar). Increasing or manipulating the possible range of potential outcomes can prime or prompt actions (*4 in Figure 14.1). For example, given that stimuli may have an “incentive salience” (Berridge & Robinson, 2016), an increased predominance of outcomes in the perceptual field will increase their salience and thus the likelihood that an outcome-prompted action will occur. This may be related to the “mere exposure” effect, whereby appeal of an object is increased by being repeatedly exposed to it (Zajonc, 2001). Enhanced outcome availability (*5 in Figure 14.1) may increase the reward value for at least two reasons: First, increased availability will reduce the cost, in terms of effort, of obtaining the item and hence increase the value to that item. More indirectly, the increased availability may imply an updated social norm. Since it is known to be inherently rewarding to act in line with or to signal or achieve a social identification, then this might make the product more valuable and hence desirable. Any or all of the above mechanisms are credible and not necessarily mutually exclusive. They are directly testable and may suggest promising ways to augment interventions. For example, if one component of the availability effect lies in a reduced acquisition effort, this could easily be tested by comparing items that were equally available but demanded different levels of effort. If this proves to be a significant factor, then enhancing the availability effect on choice by manipulating the acquisition effort could be investigated.

14.4.2.2 Changing the Size of Wine Glasses Altering, for example, the shape of a wine glass may produce subtle but important effects on the nature of the action (*2 in Figure 14.1) and an ensuing difference in the relationship between the action and the outcome (*3 in Figure 14.1). Taken together, these effects relate to affordance (see

Section 14.4.2.1). As an example, an inwardsloping glass will have a smaller rim and therefore is associated with a different embouchure that might exert an effect on drinking behaviors, including sip size, speed of consumption, and volume consumed. Reducing the glass size, which would simply not allow for larger volumes of wine to be poured, could also have a direct effect on subsequent drinking behavior. This again can be considered as an effect on the set of actions (*2 in Figure 14.1) available to the agent: A smaller volume simply precludes the possibility of drinking more. The same volume of wine poured into larger glasses is perceived as smaller (Pechey et al., 2015), resulting in more wine being poured to achieve a similar serving size to that achieved in a smaller glass (*5 in Figure 14.1). This leads to a similar valuation being achieved by different volumes of wine (Schultz, 2006), with the agent unaware of any difference in volumes of wine. Glasses to optimize portion size effects might be designed to afford less capacity or to activate greater reward value when holding smaller serving sizes or both.

14.4.2.3 Adding Graphic Health Warnings to Food and Alcohol Products A simple intervention would entail adding to a standard product a health warning label presenting an image, commonly graphically aversive, of the harm associated with overconsumption of that product accompanied by explanatory text information. Several mechanisms might be active in producing a reduction in the selection and consumption of that product. An attenuation of an existing state or stimulusdependent action probability (*2 in Figure 14.1). Simply put, if the agent experiences a different, more aversive, outcome, then the associated action may become less likely simply by virtue of its association with the outcome. Note that this effect on the action could persist irrespective of whether the outcome continued to be aversive or

Changing Behavior by Changing Environments

not consciously represented. This would arise from a model-free or stimulus-response updating process and mediated by a reduction in the reward value of the outcome in question. The addition of the graphic image and health information changes the nature of the outcome (*4 in Figure 14.1). Related to this, the reward value (*5 in Figure 14.1) of that outcome may reduce making the outcome less desirable and the action ultimately less likely. Taken together, these effects may be seen as simple conditioning produced by the information presented in the warning label. It may also have a more profound high-level effect by altering the salience of the health-harming properties of the item. Through provision of information relating to these consequences, there may be a change in the agent’s perception of the longer-term behavioral consequences, that is, the model-based updating (*1 in Figure 14.1). Note that, while both of the above mechanistic effects are possible and may act in a complementary way to change behavior, it is possible to dissociate them experimentally and, potentially, to use this information to optimize ensuing interventions. For example, if health warning labels exert their effect through a purely associative (or “model-free”) mechanism, based on associating a valued stimulus with an aversive stimulus regardless of whether the warning concerns health, then the outcome need not be causally related to the behavior but simply an aversive image that will shift the reward value of the outcome and diminish the likelihood of the related actions. Conversely, if the intervention works by devaluing the outcome, then the warning labels should emphasize the relationship between the action and the outcome, depicting consequences that are causally related to engaging in the behavior.

14.5 Research Priorities The choice architecture literature, despite drawing on a long theoretical tradition, is relatively

201

young in terms of systematic assessment. It is easy to overestimate what is currently known by extrapolating from a few preliminary studies (e.g., Marchiori, Adriaanse, & De Ridder, 2017) or by assuming that a promising intervention will generalize from the laboratory to the real world (e.g., Ben-Shahar, 2017) or from one real-world setting to another (e.g., Pechey et al., 2017). In order to avoid these errors, building a robust evidence base of behavior change interventions should involve, although not necessarily be limited to, (1) a summary of existing studies, preferably synthesized using systematic methods; (2) laboratory studies to understand mechanism; and (3) replicated field studies conducted in realworld environments to estimate effect sizes. Ensuring that the results of these studies are robust is also critical, and applying open science principles – for example, preregistration of study protocols and data sharing – will help to improve the quality of studies and robustness of the evidence they generate (Munafò et al., 2017).

14.5.1 Evidence Synthesis The vast volume of published research necessitates appropriate and precise syntheses to summarize what is currently known of behavioral interventions. Systematic review methods remain the gold standard for the comprehensive synthesis of evidence. However, given the scale and complexity of the scientific literature, increasing attention is being focused on both technological and conceptual developments that can enable more efficient cumulative synthesis (e.g., O’Connor et al., 2019). The latter includes the development of classification systems to encode and curate research knowledge on interventions and the relationships that determine their effects (e.g., Norris et al., 2019). Hollands, Shemilt et al.’s (2015) review on the impact of changing portion, package, or tableware size on the selection and consumption of food, alcohol, and tobacco illustrates the value of

202

` THERESA M. MARTEAU, PAUL C. FLETCHER, GARETH J. HOLLANDS, AND MARCUS R. MUNAFO

synthesizing existing evidence. Estimates based on the meta-analysis of the food studies suggested that “downsizing” portions, packages, and tableware could reduce energy intake in UK adults by between 12 percent to 16 percent (279 kcal) per day (Marteau et al., 2015). The review also revealed important gaps in the evidence. For example, most studies were laboratory-based, highlighting the need for more field studies to determine whether the results of laboratory studies generalize to the real world and to estimate effect sizes. Such field trials are beginning to be conducted (e.g., Hollands et al., 2018). The three tobacco studies from Hollands, Shemilt et al.’s (2015) analysis manipulated only cigarette length, highlighting a need for studies on the impact of other characteristics – for example, cigarette pack size – on smoking (Blackwell et al., 2019). Finally, no studies concerning alcohol met the inclusion criteria, an important evidence gap that is now being filled (e.g., Pechey et al., 2016).

14.5.2 Laboratory Studies Real-world behaviors can be modeled in laboratory settings to examine the mechanisms underlying the impact of an intervention on behavior and to conduct proof-of-concept experiments of interventions. For example, proxy outcomes in laboratory studies of behavior change can include urges to drink or smoke rather than actual drinking or smoking behavior. Alternatively, actual behavior can be measured, albeit in artificial conditions, for example by allowing ad lib drinking while engaged in a neutral background task, such as watching television. While only specific aspects of real-world behavior can be modeled in the laboratory, these aspects can be more precisely controlled and measured in greater depth (Maynard et al., 2017; Stothart et al., 2016). For example, the influence of glass shape on drinking behavior can be considered within the model-based/model-free framework and perhaps,

thereby, may inform laboratory experiments aimed at understanding mechanisms. If the effect is driven by the influence of the shape of the glass on the action, leading to smaller sip size, this will be apparent in assessments of drinking microbehavior, likely remaining outside of the awareness of the actors. Alternatively, the effect may lie in altered perceived volume. If perceived volume is used as a cue – likely acting nonconsciously – to guide drinking rate, then the introduction of curved or sloped glasses may distort this effect, resulting in a higher rate of consumption than expected because the cue no longer reliably predicts the outcome. These two different putative mechanisms may then inspire differing predictions about shifts in the effect of glass shape with repeated exposure (e.g., the use of straight glasses in contexts where curved glasses are the norm). If the effect acts at a basic level by shaping the action, then one would predict that this effect will persist since such a basic model-free response would not be updated with experience, although compensatory behaviors such as increased sipping frequency may emerge. Alternatively, if the effect arises from distorted perception, then there may be a learning effect. From the perspective of the model-based modelfree framework, the agent updates the model of the world based on a new relationship between sipping and volume change. While this latter effect could be considered a “model-based” change, it should be acknowledged that the model in this case is a fairly low-level one compared to how the term might generally be used. Understanding which explanation best accounts for observed data will have important downstream implications such as whether the effects of public health interventions based on mandating certain glass shapes are likely to persist or diminish over time. The laboratory can therefore serve as an efficient approach for generating initial proof-ofconcept of putative interventions to select promising ones to progress to real-world tests.

Changing Behavior by Changing Environments

Effect sizes estimated from laboratory studies are unlikely to translate directly to real-world settings, where multiple influences on behavior operate simultaneously in a dynamic way (Mitchell, 2012). Moreover, participants in laboratory studies are typically not representative of the general population, which may further limit the generalizability of findings. However, effect sizes in the laboratory may be a useful metric to select candidate interventions for real-world trials. It can also allow interventions that have been found to be effective to be interrogated in controlled conditions in order to identify the “active ingredient” of the intervention – in other words, the mechanism that underlies the effect.

14.5.3 Field Studies Field studies are those that are conducted in realworld settings, such as shops, schools, restaurants, and bars, providing ecologically valid contexts for studying behavior and the impact of interventions designed to change it. The distinction between real-world and laboratory settings is not, however, always clear-cut. For example, some researcher-designed restaurants are used to study interventions but are not open to the general public (e.g., the laboratory-restaurant at Wageningen University). There are also examples of naturalistic laboratories (e.g., so-called bar labs). Ultimately, however, promising behavior change interventions tested in laboratory settings need to be tested in real-world settings, across a range of contexts – for example, a range of bars that serve customers with different demographic profiles – in order to demonstrate their effectiveness. Such studies place constraints on the outcome measures that can be collected, which will most often be aggregate measures of behavior such as product sales (Pechey et al., 2016; Vasiljevic et al., 2017), which serve as proxies for individual behavior change. Field studies generally use a wider range of designs than those found in

203

laboratory studies in which the individual is generally the unit of randomization. These include cluster, randomized, stepped wedge (e.g., Vasilijevic et al., 2017), and multiple treatment reversal designs (e.g., Pechey et al., 2016; see also Chapters 21 and 22, this volume). As the evidence base develops, it is important to consider how it is most effectively implemented to achieve behavior change at scale. While some interventions might be implemented through voluntary action, most will likely require regulation or legislation. This will require addressing the drivers of “policy inertia,” namely the failure of policy makers worldwide to implement effective policies to change behavior at scale to tackle such global challenges as obesity and climate change (Swinburn et al., 2019). This will include addressing the main drivers of policy inertia, which include industry opposition to policies that threaten their markets (Oreskes & Conway, 2010) and low public demand for change (Diepeveen et al., 2013).

14.6 Summary and Conclusion This chapter provides the first systematic attempt to map choice architecture interventions onto existing theories and theoretical frameworks of human behavior by exploring the mechanisms by which some of the interventions have their effects. While much psychological and behavioral science can be located in one or more theories, most research does not take place within well-articulated, theoretical frameworks – broad bodies of connected theories (Muthukrishna & Henrich, 2019). These include reinforcement learning (Dayan & Niv, 2008), dual-inheritance theory (Henrich, 2016), and Bayesian models of cognition (Griffiths, Kemp, & Tenenbaum, 2008). Working within such theoretical frameworks – with formally defined assumptions and predictions – is central to developing a science of human behavior and including its change. No one theory explains all interventions that can be

204

` THERESA M. MARTEAU, PAUL C. FLETCHER, GARETH J. HOLLANDS, AND MARCUS R. MUNAFO

considered examples of choice architecture. Beyond single theories, learning theory – and in particular model-based and model-free reinforcement learning – could provide a useful theoretical framework for explaining, predicting, and developing choice architecture interventions.

References Ben-Shahar, O. (2017). More failed nudges: Evidence of ineffective “behaviorally informed” disclosures. Jotwell. https://contracts.jotwell.com/ more-failed-nudges-evidence-of-ineffective-beha viorally-informed-disclosures/ Berridge, K. C., & Robinson, T. E. (2016). Liking, wanting, and the incentive-sensitization theory of addiction. American Psychologist, 71, 670–679. https://doi.org/10.1037/amp0000059 Bianchi, F., Garnett, E., Dorsel, C., Aveyard, P., & Jebb, S. A. (2018). Restructuring physical microenvironments to reduce the demand for meat: A systematic review and qualitative comparative analysis. Lancet Planet Health, 2, e384–e397 https://doi.org/10.1016/S2542-5196(18)30188-8 Blackwell, A. K. M., Lee, I., Scollo, M., Wakefield, M., Munafò, M. R., & Marteau, T. M. (2019). Should cigarette pack sizes be capped? Addiction. https:// doi.org/10.1111/add.14770 Bollard, T., Maubach, N., Walker, N., & Ni Mhurchu, C. (2016). Effects of plain packaging, warning labels, and taxes on young people’s predicted sugar-sweetened beverage preferences: An experimental study. International Journal of Behavioral Nutrition and Physical Activity, 13, 95. https://doi.org/10.1186/s12966-016-0421-7 Bowers, K. S. (1973). Situationism in psychology: An analysis and a critique. Psychological Review, 80, 307–336. https://doi.org/10.1037/h0035592 Brewer, N. T., Hall, M. G., Noar, S. M. et al. (2016). Effect of pictorial cigarette pack warnings on changes in smoking behavior: A randomized clinical trial. JAMA Internal Medicine, 176, 905– 912. https://doi.org/10.1001/ jamainternmed.2016.2621 Cho, Y. J., Thrasher, J. F., Yong, H. H. et al. (2018). Path analysis of warning label effects on negative

emotions and quit attempts: A longitudinal study of smokers in Australia, Canada, Mexico, and the US. Social Science and Medicine, 197, 226–234. https://doi.org/10.1016/j.socscimed.2017.10.003 Clarke, N., Pechey, E., Kosite, D. et al. (2019). Impact on selection and consumption of health warning labels on food, alcoholic and non-alcoholic drinks: Protocol for a systematic review. PsyArXiv. https://doi.org/10.31234/osf.io/jt52m Crockett, R. A., King, S. E., Marteau, T. M. et al. (2018). Nutritional labelling for healthier food or non-alcoholic drink purchasing and consumption. Cochrane Database of Systematic Reviews, 2, CD009315. https://doi.org/10.1002/14651858 .CD009315.pub2 Daw, N. D., Gershman, S. J., Seymour, B., Dayan, P., & Dolan, R. J. (2011). Model-based influences on humans’ choices and striatal prediction errors. Neuron, 69, 1204–1215. https://doi.org/10.1016/j .neuron.2011.02.027 Dayan, P., & Niv, Y. (2008). Reinforcement learning: The good, the bad and the ugly. Current Opinion in Neurobiology, 18, 185–196. https://doi.org/ 10.1016/j.conb.2008.08.003 de Wit, S., & Dickinson, A. (2009). Associative theories of goal-directed behaviour: A case for animal-human translational models. Psychology Research, 73, 463–476. https://doi.org/10.1007/ s00426-009-0230-6 Diepeveen, S., Ling, T., Suhrcke, M., Roland, M., & Marteau, T. M. (2013). Public acceptability of government intervention to change health-related behaviours: A systematic review and narrative synthesis. BMC Public Health, 13, 756 https://doi .org/10.1186/1471-2458-13-756 Finkel, E. J. (2014). The I3 Model: Metatheory, theory, and evidence. In J. M. Olson & M. P. Zanna (Eds.), Advances in Experimental Social Psychology, Vol. 49 (pp. 1–104). New York: Academic Press. Fishbein, M., & Ajzen, I. (2010). Predicting and Changing Behavior: The Reasoned Action Approach. New York: Psychology Press. Griffiths, T. L., Kemp, C., & Tenenbaum, J. B. (2008). Bayesian models of cognition. In R. Sun (Ed.), The Cambridge Handbook of Computational Psychology (pp. 59–100). Cambridge: Cambridge University Press.

Changing Behavior by Changing Environments

Hammond, D. (2011). Health warning messages on tobacco products: A review. Tobacco Control, 20, 327–337. https://doi.org/10.1136/tc.2010.037630 Henrich, J. P. (2016). The Secret of Our Success: How Culture Is Driving Human Evolution, Domesticating Our Species, and Making Us Smarter. Princeton: Princeton University Press. Hollands, G. J., Bignardi, G., Johnston, M. et al. (2017). The TIPPME intervention typology for changing environments to change behaviour. Nature Human Behavior, 1, 0140. https://doi.org/10.1038/ s41562-017-0140 Hollands, G. J., Cartwright, E., Pilling, M. et al. (2018). Impact of reducing portion sizes in worksite cafeterias: A stepped wedge randomised controlled pilot trial. International Journal of Behavioral Nutrition and Physical Activity, 15, 78. https://doi.org/10.1186/s12966018-0705-1 Hollands, G. J., Marteau, T. M., & Fletcher, P. C. (2016). Non-conscious processes in changing health-related behaviour: A conceptual analysis and framework. Health Psychology Review, 10, 381–394. https://doi.org/10.1080/17437199 .2015.1138093 Hollands, G. J., McDermott, M. S., Lindson-Hawley, N. et al. (2015). Interventions to increase adherence to medications for tobacco dependence. Cochrane Database of Systematic Reviews, CD009164 https://doi.org/10.1002/14651858 .CD009164.pub2 Hollands, G. J., Shemilt, I., Marteau, T. M. et al. (2013). Altering choice architecture to change population health behaviour: A large-scale conceptual and empirical scoping review of interventions within micro-environments. www.repository.cam.ac.uk/ handle/1810/245108 Hollands, G. J., Shemilt, I., Marteau, T. M. et al. (2015). Portion, package or tableware size for changing selection and consumption of food, alcohol and tobacco. Cochrane Database of Systematic Reviews, 9, CD011045 https://doi.org/10.1002/ 14651858.CD011045.pub2 Johnson, E. J., Shu, S. B., Dellaert, B. G. C. et al. (2012). Beyond nudges: Tools of a choice architecture. Marketing Letters, 23, 487–504. https://doi.org/10.1007/s11002-012-9186-1

205

Mantzari, E., Vasiljevic, M., Turney, I., Pilling, M., & Marteau, T. (2018). Impact of warning labels on sugar-sweetened beverages on parental selection: An online experimental study. Preventive Medicine Reports, 12, 259–267. https://doi.org/ 10.1016/j.pmedr.2018.10.016 Marchiori, D. R., Adriaanse, M. A., & De Ridder, D. T. D. (2017). Unresolved questions in nudging research: Putting the psychology back in nudging. Social and Personality Psychology Compass, 11, e12297. https://doi.org/10.1111/spc3.12297 Marteau, T. M., Hollands, G. J., & Fletcher, P. C. (2012). Changing human behavior to prevent disease: The importance of targeting automatic processes. Science, 337, 1492–1495. https://doi .org/10.1126/science.1226918 Marteau, T. M., Hollands, G. J., Shemilt, I., & Jebb, S. A. (2015). Downsizing: Policy options to reduce portion sizes to help tackle obesity. BMJ, 351, h5863. https://doi.org/10.1136/bmj.h5863 Maynard, O. M., Brooks, J. C., Munafò, M. R., & Leonards, U. (2017). Neural mechanisms underlying visual attention to health warnings on branded and plain cigarette packs. Addiction, 112, 662–672. https://doi.org/10.1111/add.13699 Maynard, O. M., Gove, H., Skinner, A. L., & Munafò, M. R. (2018). Severity and susceptibility: Measuring the perceived effectiveness and believability of tobacco health warnings. BMC Public Health, 18, 468. https://doi.org/10.1186/ s12889-018-5385-x Melnikoff, D. E., & Bargh, J. A. (2018). The mythical number two. Trends in Cognitive Science, 22, 280–293, https://doi.org/10.1016/j.tics .2018.02.001 Mitchell, G. (2012). Revisiting truth or triviality: The external validity of research in the psychological laboratory. Perspectives on Psychological Science, 7, 109–117. https://doi.org/10.1177/ 1745691611432343 Munafò, M. R., Nosek, B. A., Bishop, D. V. M. et al. (2017). A manifesto for reproducible science. Nature Human Behavior, 1, 0021. https://doi.org/ 10.1038/s41562-016-0021 Muthukrishna, M., & Henrich, J. (2019). A problem in theory. Nature Human Behaviour, 3, 221–229. https://doi.org/10.1038/s41562-018-0522-1

206

` THERESA M. MARTEAU, PAUL C. FLETCHER, GARETH J. HOLLANDS, AND MARCUS R. MUNAFO

Norris, E., Finnerty, A. N., Hastings, J., Stokes, G., & Michie, S. (2019). A scoping review of ontologies related to human behaviour change. Nature Human Behavior, 3, 164–172. https://doi.org/ 10.1038/s41562-018-0511-4 O’Connor, A. M., Tsafnat, G., Gilbert, S. B. et al. (2019). Still moving toward automation of the systematic review process: A summary of discussions at the third meeting of the International Collaboration for Automation of Systematic Reviews (ICASR). Systematic Reviews, 8, 57. https://doi.org/10.1186/s13643019-0975-y O’Doherty, J. P., Cockburn, J., & Pauli, W. M. (2017). Learning, reward, and decision making. Annual Reviews of Psychology, 68, 73–100. https://doi .org/10.1146/annurev-psych-010416-044216 Oreskes, N., & Conway, E. M. (2010). Merchants of Doubt: How a Handful of Scientists Obscured the Truth on Issues from Tobacco Smoke to Global Warming. New York: Bloomsbury. Osiurak, F., Rossetti, Y., & Badets, A. (2017). What is an affordance? 40 years later. Neuroscience and Biobehavioral Reviews, 77, 403–417. https://doi .org/10.1016/j.neubiorev.2017.04.014 Pechey, R., Attwood, A. S., Couturier, D. L. et al. (2015). Does glass size and shape influence judgements of the volume of wine? PLoS ONE, 10, e0144536. https://doi.org/10.1371/journal .pone.0144536 Pechey, R., Cartwright, E., Pilling, M. et al. (2019). Impact of increasing the proportion of healthier foods available on energy purchased in worksite cafeterias: A stepped wedge randomized controlled pilot trial. Appetite, 133, 286–296. https://doi.org/10.1016/j.appet.2018.11.013 Pechey, R., Couturier, D. L., Hollands, G. J., Mantzari, E., Munafò, M. R., & Marteau, T. M. (2016). Does wine glass size influence sales for on-site consumption? A multiple treatment reversal design. BMC Public Health, 16, 390. https://doi .org/10.1186/s12889-016-3068-z Pechey, R., Couturier, D. L., Hollands, G. J., Mantzari, E., Zupan, Z., & Marteau, T. M. (2017). Wine glass size and wine sales: A replication study in two bars. BMC Research Notes, 10, 287. https:// doi.org/10.1186/s13104-017-2610-0

Pilling, M., Clarke, N., Pechey, R., Hollands, G., & Marteau, T. (2019). Wine glass size and sales: A meta-analysis of studies in bars and restaurants. Unpublished manuscript, King’s College London. Schultz, W. (2006). Behavioral theories and the neurophysiology of reward. Annual Reviews of Psychology, 57, 87–115. https://doi.org/10.1146/ annurev.psych.56.091103.070229 Stothart, G., Maynard, O., Lavis, R., & Munafò, M. (2016). Neural correlates of cigarette health warning avoidance among smokers. Drug and Alcohol Dependence, 161, 155–162. https://doi .org/10.1016/j.drugalcdep.2016.01.025 Sutton, R. S., & Barto, A. G. (1998). Reinforcement Learning: An Introduction. Cambridge, MA: MIT Press. Swinburn, B. A., Kraak, V. I., Allender, S. et al. (2019). The global syndemic of obesity, undernutrition, and climate change: Lancet commission report. Lancet, 393, 791–846. https://doi.org/10.1016/ S0140-6736(18)32822-8 Szaszi, B., Palinkas, A., Palfi, B., Szollosi, A., & Aczel, B. (2018). A systematic scoping review of the choice architecture movement: Toward understanding when and why nudges work. Journal of Behavioral Decision Making, 31, 355– 366. https://doi.org/10.1002/bdm.2035 Thaler, R., & Sunstein, C. (2008). Nudge: Improving Decisions About Health, Wealth, and Happiness. New Haven, CT: Yale University Press. Thrasher, J. F., Carpenter, M. J., Andrews, J. O. et al. (2012). Cigarette warning label policy alternatives and smoking-related health disparities. American Journal of Preventive Medicine, 43, 590–600. https://doi.org/10.1016/j.amepre.2012.08.025 Vasiljevic, M., Cartwright, E., Pechey, R. et al. (2017). Physical micro-environment interventions for healthier eating in the workplace: protocol for a stepped wedge randomised controlled pilot trial. Pilot Feasibility Studies, 3, 27. https://doi.org/ 10.1186/s40814-017-0141-z Von Neumann, J., & Morgenstern, O. (1947). Theory of Games and Economic Behavior. Princeton: Princeton University Press. Webb, T. L., & Sheeran, P. (2006). Does changing behavioral intentions engender behavior change?

Changing Behavior by Changing Environments

A meta-analysis of the experimental evidence. Psychological Bulletin, 132, 249–268. https://doi .org/10.1037/0033-2909.132.2.249 Wood, W., & Neal, D. T. (2016). Healthy through habit: Interventions for initiating and maintaining health

207

behavior change. Behavioral Science and Policy, 2, 71–83. https://doi.org/10.1353/bsp.2016.0008 Zajonc, R. B. (2001). Mere exposure: A gateway to the subliminal. Current Directions in Psychological Science, 10, 224–228.

15 Changing Behavior Using Integrated Theories Martin S. Hagger and Kyra Hamilton

Practical Summary A vast number of theories have been applied to predict behavior. The theories can be useful in identifying the potentially changeable factors or constructs that can be reliably related to behavior through various processes and can be targeted in behavior change interventions. Making sense of the numbers of theories and their constituent constructs presents considerable challenges for those interested in changing behavior. Theory integration may assist in narrowing the number of constructs and processes to a manageable number that are optimally effective in determining behavior. In addition, despite claims of broad applicability, research has shown that existing theories are not optimally effective in predicting behavior across behaviors, contexts, and populations, with gaps or boundary conditions that limit their application. Theory integration may assist by providing additional constructs and processes to address these limitations. A number of integrated theories have been developed and have shown promise in identifying determinants of behavior. They have also been used as the basis for behavior change interventions. However, research applying integrated theory-based interventions is sparse and more research is needed, particularly of interventions that target change in separate constructs from integrated theories.

15.1 Introduction A myriad of social psychological theories has been applied to predict behavior as part of the endeavor to identify viable targets for behavior change interventions. Many of the theories adopt a social cognition approach, which assumes that motivation toward, and actual enactment of, behavior is a function of individuals making reasoned decisions based on their processing of available socially determined information. While such theories have been Martin S. Hagger’s contribution was supported by a Finnish Distinguished Professor (FiDiPro) award from Business Finland (1801/31/2015). https://doi.org/10.1017/9781108677318.015

effective in explaining variance in behaviors (e.g., Armitage & Conner, 2001; Hagger, Koch et al., 2017; Milne, Sheeran, & Orbell, 2000), the vast number of theories, and the factors or constructs of which they comprise, presents problems in synthesizing this research and hinders efforts at identifying core sets of constructs that likely account for variance in target behaviors in relevant contexts and populations. A further problem is that the sets of constructs and associated processes identified in social cognition theories often fail to account for a substantive proportion of the variance in behavior. Solutions to these problems may lie in synthesizing the content of multiple theories to develop new integrated theories that provide more effective

Changing Behavior Using Integrated Theories

209

Sidebar 15.1 What is an integrated theory?

Definitions of a theory in the psychological and behavioral science literature abound. For example, Kerlinger and Lee (1999, p. 11) define a theory as “a set of interrelated (concepts), definitions, and propositions that present a systematic view of phenomena by specifying relations among variables, with the purpose of explaining and predicting phenomena” (see also Davis et al., 2015; Head & Noar, 2014). An integrated theory draws constructs and specified relations from more than one existing theory to arrive at a new theory. Prentice-Dunn and Rogers (1986) coined the term hybrid theory and give the example of the integration the health belief model and social cognitive theory to produce protection motivation theory. There are numerous reasons or goals for the integration of theories (for details, see Section 15.2); often the focus is on addressing gaps or boundary conditions in theories to arrive at comprehensive but optimally parsimonious explanations. Theory integration is therefore predicated on the basis that no theory can be considered definitive and thus should be viewed as modifiable on the basis of evidence and new information coming to light that challenge its propositions or lead to its partial falsification. Many theorists consider theories as “living” entities that summarize current knowledge and are therefore open to modification as knowledge from observation accumulates that either supports predictions or challenges them and signals the need for emendations that contribute to the knowledge summary. Theory integration is therefore part of the process in which theories develop and evolve to provide more efficacious explanations of outcomes such as behavior and processes involved.

behavioral prediction with optimal parsimony. The aim of the present chapter is to chart research on the development and application of the integration of social psychological theories to predict and, ultimately, inform efforts to change behavior. The chapter will outline some of the key approaches to theory integration (for a definition of integrated theories, see Sidebar 15.1), describe illustrative integrative theories and review their application to behavior change, outline the utility of basing behavior change interventions on integrated theories, and provide recommendations for future research on integrated theories applied to behavior change.

15.2 Approaches to Integration There are numerous advantages of theory integration. Theory integration can assist in addressing

the gaps or boundary conditions of theories that place limitations on their capacity to explain behavior (Trafimow, 2012). Integration can also assist in managing and reducing redundancy across theories by identifying sets of core constructs that are optimally effective in predicting behavior and with high parsimony (Hagger, 2009). In addition, integrated theories can assist in identifying constructs to target in behavior change interventions and that are implicated in mechanisms of change (see Chapter 20, this volume). The integration of constructs and mechanisms from multiple theories to arrive at a modified theory is, therefore, one possible solution to addressing the shortcomings of an individual theory. The modified theory can then be subject to new tests to assess its adequacy in accounting for variance in behavior and

210

MARTIN S. HAGGER AND KYRA HAMILTON

explaining the processes involved (Hagger, Gucciardi, & Chatzisarantis, 2017). There are numerous approaches to theory integration. Four prominent approaches are summarized next: (1) additional constructs; (2) the core constructs approach; (3) consensus-based approaches; and (4) utility-based approaches. More detailed treatment of each approach can be found in Appendix 15.1 (supplemental materials). A basic approach to theory integration has been the addition of constructs from another theory to augment explanation and address shortcomings in the prediction of existing theory. There are many examples of how social cognition theories have been augmented to include additional constructs. For example, the health belief model (Rosenstock, 1974; see Chapter 4, this volume) and the theory of reasoned action (Fishbein & Ajzen, 1975) were augmented to include self-efficacy from social cognitive theory (Bandura, 1986; see Chapter 3, this volume) to produce integrated theories labeled protection motivation theory (Rogers, 1975; see Chapter 4, this volume) and the theory of planned behavior (Ajzen, 1991; Chapter 2, this volume), respectively. These integrated theories have become well-used theories in their own right and are not often acknowledged as integrated theories. As with all theories, an important consideration when adding constructs is the importance of clarity of specification. The ad hoc addition of constructs without clear specification of how they relate to other constructs constitutes a “weak” theory, consistent with definitions of theory (see Sidebar 15.1) and with theory quality criteria (Davis et al., 2015). The core constructs approach is a further approach to theory integration. Rather than augmenting an existing theory by adding constructs, the core constructs approach aims to collapse the myriad constructs from theories with similar content but different labels into core categories, based on a review of their content and definitions (Block, 1995; Hagger, 2014). The core constructs

are then incorporated into a theory or model that specifies relations among them derived from their constituent theories. Numerous authors have applied the core constructs approach to theory integration (e.g., Abraham & Sheeran, 2000; McMillan & Conner, 2007; Noar & Zimmerman, 2005; Protogerou, Johnson, & Hagger, 2018). For example, McMillan and Conner’s (2007) content analysis of theories applied to behavior change identified four overarching categories of construct: attitudes (with affective and evaluative subcategories); self-representations; norms (with injunctive and descriptive subcategories); control perceptions; and dispositions to act. Dispositions to act, which reflect motivation and intention measures, are depicted at the apex of the representation, consistent with theories such as protection motivation theory and the theories of reasoned action and planned behavior. While the core constructs approach has utility in summarizing constructs across multiple theories, and relations among them, no study has explicitly proposed sets of predictions among the constructs and tested them empirically. The approach should therefore be viewed as informative of integration rather than yielding a specific integrated theory. Another approach to identifying commonalities and redundancy across theories is to use expert consensus. Such approaches have value because they provide a relatively rapid means of tapping into the in-depth and broad knowledge of theory content and mechanisms held by experts. This can be particularly useful in summarizing and synthesizing the content of different theories applied to behavior change. For example, a group of leading theorists pooled their knowledge of theories applied to behavior change to develop an integrated model to predict HIV prevention behaviors (Fishbein et al., 2001) (see the “major theorists’ model” described in Section 15.3 and Appendix 15.2, supplemental materials). Similarly, Michie and colleagues (e.g., Carey et al., 2019; Connell et al., 2018) used expert consensus to develop a system to classify and describe the content of

Changing Behavior Using Integrated Theories

theories applied to behavior change. For example, Davis et al. (2015) identified theories applied to behavior change through a systematic review of current literature. These were subsequently assessed for quality and their frequency of use and content. Subsequent research applied expert consensus to specify links between constructs from identified theories and specific behavior change techniques, that is, the methods used to affect change in theoretical constructs and that comprise the “active ingredients” of behavior change interventions (Michie et al., 2011). The research has yielded a searchable database of links between constructs and techniques (Connell et al., 2018). The database not only is useful to inform further theory integration but also can help inform interventions to change behavior based on integrated theories by identifying potential means to change target constructs from the theories. A final approach to integration is a utility-based approach. Two important utility-based reasons for theory integration are the reduction of redundancy and increasing complementarity (Hagger, 2009). Reducing redundancy addresses the imperative of reducing constructs from different theories into a core set of constructs with clear labels, definitions, operationalizations, and means to measure them, as in the core constructs approach. Such efforts address so-called jangle fallacies in theories due to multiple constructs with similar content but different labels. Identifying how constructs and processes from different theories provide complementary explanations of behavior is a further goal of theory integration. Such a process likely begins with the recognition of limitations or gaps in the scope of the explanations of behavior offered by a particular theory or boundary conditions that delimit its predictions. The latter may imply the recognition of certain conditions or contextual variables that may determine the extent to which particular theoretical constructs or processes are relevant to predicting and explaining the particular behavior and those that may be redundant. For example, risk perceptions feature prominently in many social

211

cognition theories applied to predict behaviors, particularly behaviors that present a risk to health (Rogers, 1975; Rosenstock, 1974). However, research has suggested that such perceptions are seldom relevant in determining the behavior in populations that do not perceive themselves as vulnerable or are not directly at risk (e.g., perceived risk of cardiovascular disease as a determinant of physical activity participation in younger populations; Zhang et al., 2019). In which case, risk perceptions may not be relevant in a theory applied to that particular population. Developers of integrated theories must, therefore, be aware of the evidence available for particular theories, or the constructs and mechanisms from particular theories, in the context, behavior, and population of interest and formulate their hypotheses accordingly. The specificity of the theory, and its subsequent generalizability, should be titrated according to the breadth of behavioral phenomena to which it is expected to apply. With respect to the previous example on the role of risk perception as a behavioral determinant, if the theorist expects theory predictions to vary across populations, they may specify auxiliary assumptions to account for such variations (e.g., they may specify the moderation of effects of risk perceptions on behavior by age or perceived vulnerability) or, if not, they may opt to omit risk perceptions from the theory altogether. In summary, the utilitarian approach involves reducing redundancies in constructs across theories by (1) distilling and collapsing constructs of existing theories and (2) optimizing comprehensiveness by examining complementarity in constructs and mechanisms across theories appropriate to the behavior(s), population(s), and context(s) of interest.

15.3 Integrated Theories and Their Contribution to Behavior Change Many integrated theories have been developed and tested based on the approaches outlined in

212

MARTIN S. HAGGER AND KYRA HAMILTON

the previous sections (e.g., Bagozzi & Warshaw, 1990; Borland, 2017; de Vries et al., 2005; Gerrard et al., 2005; Hagger & Chatzisarantis, 2014; Hall & Fong, 2007; Hamilton, Cox, & White, 2012; Maher & Conroy, 2016; Montaño & Kasprzyk, 2015; Perugini & Conner, 2000). The theories have generally been responses to gaps in knowledge or limitations or boundary conditions of the theory identified in applications or empirical tests of the theory in particular behaviors, contexts, and populations (for a more detailed summary of preeminent integrated theories, see Appendix 15.2, supplemental materials). For example, inclusion of perceived behavioral control in the theory of reasoned action to produce the theory of planned behavior, effectively integrating the construct of self-

efficacy into the theory from social cognitive theory, was to promote the explanatory capacity of the theory for behaviors that are not under complete volitional control of the actor. Progression in the development of integrated theories, therefore, also reflects the current state of knowledge of the determinants of behavior, and the associated processes, and, more broadly, reflects the evolution in behavioral theory. The “major theorists’ model” (Figure 15.1) is a prime example of theory integration to address gaps in extant knowledge of theories adopting the social cognition approach (Conner & Norman, 2015). The authors of five major social cognition theories developed in the 1970s and 1980s aggregated constructs from their theories to form a core set of constructs and associated predictions to

SelfDiscrepancy Skills Advantages/ Disadvantages

Social Pressure

Intention

Behavior

Self-Efficacy Environmental Constraints Emotional Reactions

Figure 15.1 Schematic representation of the major theorists’ model (after Conner & Norman, 2015) Source. Adapted with permission from Conner, M. T., & Norman, P. (2015). Predicting and changing health behaviour: A social cognition approach. In M. T. Conner & P. Norman (Eds.), Predicting and Changing Health Behaviour: Research and Practice with Social Cognition Models (3rd ed., pp. 1–29). Maidenhead: Open University Press.

Changing Behavior Using Integrated Theories

provide more comprehensive predictions of behavior. Intentions were identified as a proximal predictor of behavior, with perceived advantages or disadvantages for performing the behavior, social norms, self-efficacy, emotional reactions, and self-discrepancy as predictors of intentions. In a substantive progression of extant theory, the model specified environmental constraints and skills as additional direct predictors of behavior, reflecting the potential of these constructs as important conditions that determine behavior independent of intentions. Integrated theories have been further developed by including constructs that reflect different processes. Most social cognitive theories assume that individuals’ behavior is a function of intentional or reasoned decision-making based on beliefs and the processing of available information about future performance of the behavior. This is reflected by the mediation of social cognition constructs on behavior by intention, as predicted by theories like the theories of reasoned action and planned behavior, protection motivation theory, and the major theorists’ model. However, theorists recognized that not all behavior is a function of reasoned, intentional decision-making and that individuals often acted on the basis of relatively minimal information, with little dependence on intentional, reasoned processing. This led to the integration of constructs and processes that reflect acting on the basis of nonconscious processing. The primary source for these modifications has been dual-process models, which assume that individuals’ behavior is a function of dual processes that operate in parallel: a reasoned process, consistent with the intentionmediated route, which requires deliberation and is effortful and relatively slow, and an automatic, nonconscious route that requires little deliberation, is rapid and less effortful, and is impulsive and less discerning (see Chapter 12, this volume). Examples of integrated theories that were among the first to specify dual pathways to behavior include the theory of interpersonal behavior

213

(Triandis, 1977) and the integrated behavioral model (Figure 15.2; Kasprzyk, Montaño, & Fishbein, 1998; Montaño & Kasprzyk, 2015). These theories incorporated habits or previous experience into existing social cognitive approaches as direct predictors of behavior unmediated by intentions and represent nonconscious effects on behavior. Habits reflect enacting behavior due to acquired associations between performance of a given behavior in the presence of consistent, stable contexts or cues, which can result in nonconscious activation of the behavior when in the same context or on presentation of the cues (Hagger, 2019; Chapter 13, this volume). Research adopting these theories, as well as tests of other social cognition theories including direct effects of past behavior, has demonstrated pervasive effects on behavior, illustrating the importance of nonconscious processes in determining behavior (e.g., Hagger, Trost et al., 2017; Hamilton et al., 2017; Maher & Conroy, 2016; Presseau et al., 2014). Another example of a theory that adopts a dualprocess approach is the prototype willingness model (Figure 15.3; Gerrard et al., 2005; Gerrard et al., 2006). The model proposes a “reasoned” route to behavior, represented by effects of social cognition constructs from the theory of reasoned action, including attitudes and subjective norms mediated by intentions. Unique to the model is a “social reactive” pathway, a less deliberative route to behavior represented by effects of perceived prototypes of the behavior – representations or images of the typical person who engages in the target behavior mediated by behavioral willingness. This pathway is derived from the elaboration likelihood model (Petty & Cacioppo, 1986), a dual-process model specifying that actions are a function of deliberative and heuristic processes. Willingness is defined as the desire to perform the behavior given the social context and past experience with the behavior. A person holding a prototype of the target behavior as favorable (e.g., viewing the social image of a

214

MARTIN S. HAGGER AND KYRA HAMILTON

Attitudes Affective Beliefs

Affective Attitudes

Knowledge and Skills

Behavioral Beliefs

Instrumental Attitudes

Salience of the Behavior

Norms Normative Beliefs – Others’ Expectations

Injunctive Norm Intention

Normative Beliefs – Others’ Behavior

Behavior

Descriptive Norm

Agency Control Beliefs

Perceived Behavioral Control

Environmental Constraints

Efficacy Beliefs

Self-Efficacy

Habits

Figure 15.2 Schematic representation of the integrated behavioral model (Montaño & Kasprzyk, 2015)

Attitudes/ Perceived Vulnerability Behavioral Intention/ Expectation

Past Behavior

Subjective Norms

Behavior

Behavioral Willingness

Prototypes

Figure 15.3 Schematic representation of the prototype willingness model (Gibbons, Houlihan, &

Gerrard, 2009)

Changing Behavior Using Integrated Theories

person of the same age who smokes cigarettes or drinks alcohol favorably; “they are ‘cool’”) and similar (e.g., they are “very much like me”) is more likely to be willing to enact the behavior through the social reactive pathway. The specification of dual pathways in the model provides an example of how both constructs and processes from multiple theories (e.g., theory of reasoned action, elaboration likelihood model) have been incorporated to provide an explanation of specific types of behavior (Gerrard et al., 2005; van Lettow et al., 2016). A parallel line of research has integrated additional processes in social cognition theories to address the relatively modest effect of intentions on behavior. Studies have demonstrated that the intention-behavior relation is imperfect and typically small-to-medium in size (e.g., Rhodes & de Bruijn, 2013), suggesting that, while some individuals tend to follow through with their intentions to act, a substantive proportion do not. This has resulted in the integration of constructs from the model of action phases that distinguish between two phases of action: a motivational phase, which describes how intentions are formed, and a volitional phase, which describes how intentions are enacted. The motivational phase is represented by intentions and its determinants, as described in many social cognition theories. The volitional phase is represented by the integration of planning constructs such as implementation intentions or “if-then” plans, which function to assist intention enactment. The health action process approach is a preeminent example of an integrated model that incorporates constructs that reflect motivational (e.g., behavioral beliefs, self-efficacy) and volitional (e.g., action planning) processes (see Chapter 7, this volume). Such research has inspired a new breed of integrated theory that derives its predictions from social cognition, dual-process (Chapter 12, this volume), and dual-phase (Chapters 6 and 7, this volume) theories. These theories integrate

215

constructs and pathways that represent multiple processes to arrive at more comprehensive explanations of behavior, as well as the mechanisms involved. Prototypical of such approaches is the integrated behavior change model (Hagger & Chatzisarantis, 2014). The model specifies three important behavioral processes: (1) a reasoned route to behavior represented by the effects of social cognition constructs from the theory of planned behavior on behavior mediated by intentions; (2) a nonconscious route to behavior represented by the direct effects of implicit attitudes and motives; and (3) two action phases, a motivational phase in which intention formation is determined by constructs identified in process (1) and a volitional phase in which intentions are implemented via planning (see Figure 15.4). The model is also unique in that it identifies autonomous forms of motivation from self-determination theory as determinants of the social cognition constructs, consistent with integrated theory research demonstrating that individuals’ beliefs are a function of their autonomous or controlled reasons for acting (see Hagger & Chatzisarantis, 2009). Although the theory is relatively new, it has been subjected to some tests in numerous contexts. Such tests have generally supported model predictions, although the strength of effects, and some of the component pathways, vary across behaviors and contexts (Brown et al., 2017; Caudwell et al., 2019; Hagger, Trost et al., 2017; Hamilton et al., 2017, 2020). The accumulation of evidence testing of model hypotheses in multiple contexts and populations, and for different behaviors, will permit a definitive evaluation of the overall efficacy of the integrated model in providing a comprehensive explanation of behavior.

15.4 How Have Integrated Theories Been Used to Change Behavior? The value of integrated models as means to change behavior lies in the opportunity to identify

216

MARTIN S. HAGGER AND KYRA HAMILTON

multiple constructs that serve as targets for intervention and, as a consequence, to use multiple behavior change methods or techniques to change behavior. Interventions based on these theories need to identify techniques and strategies that are effective in changing the target constructs, consistent with the typical process and logic models of interventions proposed elsewhere (see also Chapters 19 and 20, this volume). For example, an intervention based on the integrated behavioral model (Montaño & Kasprzyk, 2015) would need to target change in social cognition constructs like experiential attitudes and injunctive norms as well as habits in order to claim to be based on the integrated theory. Similarly, interventions based on the integrated behavior change model would need to target constructs reflecting the reasoned and nonconscious routes to behavior. However, it is important that researchers pay due diligence by conducting appropriate formative predictive and experimental research demonstrating that the theory is efficacious in accounting for variance in the target behavior and signal the appropriate constructs and processes that need to be tapped in the intervention in order to evoke behavior change. Such formative research may indicate that the integrated theory is not efficacious in predicting and changing behavior or that only certain constructs or processes have efficacy, in which case the theory may regress to one of the constituent theories. For example, application of the integrated behavior change model in the context of pre-drinking, a maladaptive pattern of alcohol consumption, demonstrated that intentions were a relatively weak predictor of pre-drinking behavior, with habits and implicit alcohol identity as the most salient predictors (Caudwell et al., 2019). Such research suggests that the integrated model may not be appropriate in this context and that nonconscious, implicit processes, alone, should be the target of the intervention. Such formative work is therefore critical to the identification of appropriate target constructs and, in turn, the

behavior change methods or techniques that would make for an optimally efficient and efficacious intervention. This is consistent with those that advocate the application of theories, and the constructs and processes within theories, to specific behaviors and the context and population in which it is to be tested (e.g., Head & Noar, 2014). While there has been a proliferation of conceptual work on the development of integrated theories, and research on their efficacy in predicting behavior, studies testing their application to behavior change intervention is, by comparison, less prevalent. Certain theories that could be considered integrations have received considerably more attention in terms of their role in informing interventions than others. For example, models such as the health action process approach and the prototype willingness model have been the subject of numerous experimental and intervention trials in multiple contexts. Taking the health action process approach as an example, interventions based on the model have included techniques such as providing mastery experiences alongside specifying coping and action plans (Payaprom et al., 2011; Zhou et al., 2015). These techniques target constructs of the model (e.g., self-efficacy, action planning, coping planning) that set it apart from other theories, and, therefore, the intervention can be said to be based on the model. Importantly, the interventions have been effective not only in affecting change in the behavior but also in the intermediate constructs implicated in the mechanism of change proposed by the theory (see Chapter 20, this volume). Similarly, interventions based on the prototype willingness model include techniques promoting prototype favorability and similarity, both essential components of the model, and have been shown to be effective in changing behavior as well as model components implicated in the change process, such as behavioral willingness (e.g., Gerrard et al., 2006). However, there have been relatively few tests of interventions based on integrated theories. There

Changing Behavior Using Integrated Theories

217

Action Planning

Autonomous Motivation

Attitude

Implicit Attitudes

Subjective Norm

Intention

Perceived Behavioral Control

Implicit Motivation

Behavior

Figure 15.4 Schematic representation of the integrated behavior change model (Hagger &

Chatzisarantis, 2014)

have, however, been interventions targeting change in constructs representing processes derived from multiple theories, which have implications for integrated models. For example, Chatzisarantis and Hagger (2009) conducted an intervention aimed at promoting leisure-time physical activity among schoolchildren by providing motivational techniques and messages delivered by physical education teachers. The intervention used multiple techniques targeting change in behavioral beliefs and attitudes from the theory of planned behavior and autonomous motivation from self-determination theory. Teachers were trained to deliver messages promoting the advantages and allaying the disadvantages of participating in physical activity and to deliver the messages and lesson content using autonomy-supportive language. The intervention demonstrated change in physical activity behavior and constructs from both theories, with small-to-medium effect sizes. Aside from providing support for an intervention targeting constructs from the theory of planned behavior and self-determination theory (Hagger & Chatzisarantis, 2009), the intervention also

served to feed forward in informing the development of the integrated behavior change model (Hagger & Chatzisarantis, 2014; see Figure 15.4). An important consideration in the development and evaluation of behavioral interventions, and one that is particularly pertinent to those based on integrated theories and processes from multiple theories, is the use of factorial designs to isolate the unique and interactive effects of different techniques adopted (Chapter 20, this volume). Such tests are important to precisely identify which of the techniques are responsible for behavior change and whether the multiple techniques used work alone or synergistically with others to produce the change. This will assist interventionists in developing interventions that are optimally efficient and efficacious in changing behavior and also assist in providing further evidence that feeds forward in the development of integrated theories. As an example, an intervention study adopting a factorial design using techniques targeting constructs from multiple theories was conducted by Prestwich, Lawton, and Conner (2003). The

218

MARTIN S. HAGGER AND KYRA HAMILTON

intervention aimed to promote increased physical activity participation in student participants. Participants were randomly assigned to one of four intervention groups or conditions: a motivational condition in which participants were presented with a “decisional balance sheet” based on social cognitive approaches such as protection motivation theory; a planning condition comprising a planning intervention based on action control theories; a combined motivation and planning intervention; and a “no intervention” control condition. Findings indicated that both the motivational and the planning techniques led to greater frequency and duration of physical activity, relative to controls, and the combination of motivation and planning techniques led to greater physical activity frequency than seen with the planning-only group. The study exemplifies the importance of isolating different techniques tapping into different processes and evaluating their singular and synergistic effects on behavior change. It is also important that interventions using content based on integrated theories provide an evaluation of the proposed mechanisms by which they are purported to affect behavior change. Consistent with approaches advocating the construction of “logic models” of proposed effects of intervention content on behavior change through targeted constructs identified in theories (see also Chapters 19 and 20, this volume), analyses of intervention mechanisms require researchers to demonstrate the extent to which interventions affect change in measures of both the target behavior and the intermediate target theory-based constructs implicated in the proposed mechanism. In practice, such mechanisms are typically tested using regression-based mediation analyses, which permits researchers to demonstrate whether any effect of the intervention on behavior occurs indirectly through the proposed theorybased constructs. In the context of integrated models, mediation analyses should demonstrate effects of intervention manipulations on multiple

theory constructs related to the targeted mechanisms. For example, Chatzisarantis and Hagger (2009) demonstrated that effects of their autonomy support and behavioral beliefs intervention on leisure-time physical activity occurred through changes in autonomy support and changes in participating children’s perceived autonomy support and intentions, constructs from the integrated theory on which the intervention was based. Similarly, Prestwich et al. (2003) demonstrated that the effects of their intervention using both motivational and planning strategies on physical activity frequency were mediated by participants’ ability to effectively recall their plans. While these examples provide important information to researchers and interventionists applying integrated theories to inform behavior change interventions, such analyses are seldom conducted. It is also worth noting that unbiased mediation analyses require studies that affect change in both the mediating construct and the technique or intervention content (Bullock, Green, & Ha, 2010) and, as such, the collection of data to test mediation effects for behavior change interventions can be challenging.

15.5 Future Behavior Change Research Using Integrated Theories Thus far, research has demonstrated that integrated theories have effectively informed the development of behavior change interventions by, for example, identifying target constructs and outlining the mechanisms involved. Nevertheless, while the literature on the application of integrated theories to behavior change is expanding, evidence supporting their application across behaviors, contexts, and populations is relatively sparse. Important priorities for future research in this area include the need for more systematic descriptions of integrated theory constructs and the interrelations among them, as well as the links between the constructs and the

Changing Behavior Using Integrated Theories

techniques purported to change them; the need for systematic experimental and intervention research that tests the effects of the techniques on both behavior change and the targeted integrated theory constructs; tests of integrated theory-informed interventions that consider context and sample characteristics as potential moderators; and the need to consider whether current integrated theories are optimally comprehensive and parsimonious through systematic research comparing their effects. These issues will be outlined in the subsequent sections.

15.5.1 Precision in Integrated Theory Specification Clarity and specificity in the definitions of constructs in integrated theories, and in the specification of the relations among them, are essential if they are to have value in advancing knowledge on the determinants of behavior and also in informing behavior change interventions. One of the challenges faced by researchers developing integrated theories is the lack of consistency and commonly accepted terminology in describing constructs, as well as means to specify constructs (Connell et al., 2018; Davis et al., 2015). Work aimed at developing core constructs of theories is an important step in developing common sets of definitions of constructs. Researchers have also utilized systems derived from data analytic procedures to specify relations among constructs, such as the diagrammatic forms used in path analyses and structural equation modeling (Hayes, 2018). Such approaches provide formal procedures to specify relations between constructs and have the advantage of having direct correspondence to analytic techniques that may test those relations in empirical analysis. Recent approaches have also done similar work, using systems theory terminology to specify relations among existing theories, and may be useful for the specification of integrated theories (West et al., 2019). Such ongoing work to progress

219

precision in the specification of integrated theories will advance knowledge by facilitating clarity in empirical tests and permitting comparisons across theories.

15.5.2 Linking Constructs and Behavior Change Techniques in Integrated Theories Similarly, consensus-based work aiming to systematically link constructs from theories applied to behavior change and the behavior change techniques will also advance the development of interventions based on integrated theories. Recent research has used expert consensus methods to summarize current theoretical and empirical knowledge of links between theory constructs and behavior techniques from behavior change technique taxonomies (Carey et al., 2019; Connell et al., 2018). The resulting databases provide useful information for interventionists interested in developing behavior change interventions based on integrated theory. Such work may be useful if the integrated theory has not been used extensively as a basis for intervention but its component constructs and associated mechanism have. The database of techniques may therefore provide guidance on the kinds of techniques that may be adopted to change the target constructs of integrated theories and inform the development interventions based on them.

15.5.3 Applying Integrated Theories Many established social cognition theories have a long history of application in multiple behaviors, contexts, and populations, which provides researchers and interventionists with a substantive body of research from which to draw on when developing behavioral interventions and sets of predictions for research testing them. By contrast, integrated theories have seldom had such broad application and, thus, researchers have less formative work on which to base their

220

MARTIN S. HAGGER AND KYRA HAMILTON

interventions and make judgments on whether the theory will be appropriate for particular behaviors and in their population and context of interest. For example, different constructs may be more important in certain contexts and in certain populations (e.g., for unmotivated populations it may make sense to target change in motivational constructs prior to targeting change in volitional constructs). Formative research on the behavior, population, and context of interest to identify salient predictors and piloting of intervention materials is therefore critical.

15.5.4 Optimally Comprehensive and Parsimonious Theories While integrated theories have aimed to provide comprehensive descriptions of behavioral determinants and the associated processes by which they act, it is unlikely that they account for all likely determinants. Furthermore, the inclusion of additional constructs and processes comes at the expense of parsimony, which may not necessarily be met with meaningful increases in explanatory power. The onus is therefore on those developing integrated theories, and the research community at large, to contribute to the development and refinement of the theories by providing high-quality empirical tests of theory premises. Such tests are imperative to the process of updating theories, particularly in identifying constructs and related processes that are ineffective in accounting for variance in behavior and behavior change. The tests may also highlight deficiencies in existing theory and signpost gaps in knowledge and boundary conditions. Testing integrated theories, and ongoing conceptual work, is important in the continual development of integrated theories and can indicate how new constructs and processes from other theories can be incorporated that may assist in improving the theory comprehensiveness. For example, existing theories seldom account for affective processes, despite considerable research traditions in these specific fields that

indicate the importance of these constructs (e.g., Bagozzi & Warshaw, 1990; Rhodes & Gray, 2018). Research adopting the reasoned action approach has incorporated constructs relating to anticipated affect and demonstrated that they have both a direct and an indirect impact on behavior (Lawton, Conner, & McEachan, 2009; Perugini & Conner, 2001). However, more elaborated theories of affective processes need to be considered, given the central role affect has been shown to play in determining impulsive and appetitive behaviors, and recent integrated theories have begun this process (e.g., Brand & Ekkekakis, 2018). The addition of affective processes alongside reasoned (e.g., effects of social cognitions like attitudes mediated by intentions), nonconscious (e.g., direct effects of implicit attitudes and habits), and multiple phase (e.g., distinction between intention formation and volitional phases of action) processes paves the way for an elaborated “tetra-process” integrated theory or framework that may advance knowledge on behavioral determinants and processes and is an avenue for future research. A further issue that has not been fully incorporated in many social cognitive theories is the issue of goal salience and competing goals. Although many theories make the assumption that intentions and beliefs are goal-directed (e.g., Perugini & Bagozzi, 2001; Perugini & Conner, 2000), few incorporate constructs related to the active goals readily being pursued by the individual. Recently, Ajzen and Kruglanski (2019) have integrated the theory of planned behavior (Ajzen, 1991) and goal systems theory (Kruglanski et al., 2002), with a view to incorporating the role that active goals have in determining intentional behavior. The resulting theory of reasoned goal pursuit proposes that the intention to perform a specific behavior is dependent not only on relevant beliefs with respect to the future performance of the behavior, but also on the different active goals and the behavioral options available to the individual (Ajzen & Kruglanski, 2019). The theory is

Changing Behavior Using Integrated Theories

expected to broaden the range of predictions of the theory of planned behavior and also to help guide behavioral interventions. However, the theory remains, as yet, untested and future formative tests of the theory of planned behavior that incorporate measures of active goals and available behavioral options should be conducted in order to evaluate the viability of this theory to explain behavior and guide behavior change.

15.6 Summary and Conclusion The vast number of theories applied to predict behavior presents considerable problems to researchers interested in identifying behavioral determinants and potentially modifiable targets for intervention. Considerable redundancy in content has been identified in constructs from these theories, making it challenging to differentiate between constructs. In addition, claims of generalizability in theories are likely overstated given research charting considerable variability in the predictive power of theories across behaviors, contexts, and populations. Theory integration offers an elegant solution to these problems through the collapsing of constructs with similar or identical content across theories and the incorporation of additional constructs and associated processes into theories to arrive at theories that offer improved predictive capability with minimum expense to parsimony. The present chapter has provided examples of procedures used to integrate theories and examples of theory integration. Theory integration has been at the forefront of theory development and very much charts the evolution of behavioral theory. While literature on the conceptual development of integrated theories and tests using correlational and longitudinal designs has proliferated, there have been relatively few experimental and randomized trials testing behavioral interventions based on integrated theories. Further, tests of existing integrated theory-based behavior change interventions have shown promise, but there is need for

221

more intervention research, particularly adopting factorial designs to the isolated and interactive effects of techniques targeting change in integrated theory components. Further development of integrated theories may lie in the integration of constructs representing processes that are seldom incorporated into social cognition theories, such as affective processes.

References Abraham, C., & Sheeran, P. (2000). Understanding and changing health behaviour: From health beliefs to self-regulation. In C. Abraham & P. Norman (Eds.), Understanding and Changing Health Behaviour: From Health Beliefs to SelfRegulation (pp. 3–24). Amsterdam: Harwood Academic Publishers. Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50, 179–211. https://doi.org/10.1016/ 0749-5978(91)90020-T Ajzen, I., & Kruglanski, A. W. (2019). Reasoned action in the service of goal pursuit. Psychological Review, 126, 774–786. https://doi.org/10.1037/rev0000155 Armitage, C. J., & Conner, M. (2001). Efficacy of the theory of planned behaviour: A meta-analytic review. British Journal of Social Psychology, 40, 471–499. https://doi.org/10.1348/014466601164939 Bagozzi, R. P., & Warshaw, P. R. (1990). Trying to consume. Journal of Consumer Research, 17, 127–140. https://doi.org/10.1086/208543 Bandura, A. (1986). Social Foundations of Thought and Action: A Social-Cognitive Theory. Englewood Cliffs, NJ: Prentice-Hall. Block, J. (1995). A contrarian view of the five-factor approach to personality description. Psychological Bulletin, 117, 187–215. https://doi .org/10.1037/0033-2909.117.2.187 Borland, R. (2017). CEOS theory: A comprehensive approach to understanding how to maintain behaviour change. Applied Psychology: Health and Well-Being, 9, 3–35. https://doi.org/10.1111/ aphw.12083 Brand, R., & Ekkekakis, P. (2018). Affective–reflective theory of physical inactivity and exercise. German

222

MARTIN S. HAGGER AND KYRA HAMILTON

Journal of Exercise and Sport Research, 48, 48– 58. https://doi.org/10.1007/s12662-017-0477-9 Brown, D. J., Hagger, M. S., Morrissey, S., & Hamilton, K. (2017). Predicting fruit and vegetable consumption in long-haul heavy goods vehicle drivers: Application of a multi-theory, dual-phase model and the contribution of past behaviour. Appetite, 121(1), 326–336. https:// doi.org/10.1016/j.appet.2017.11.106 Bullock, J. G., Green, D. P., & Ha, S. E. (2010). Yes, but what’s the mechanism? (don’t expect an easy answer). Journal of Personality and Social Psychology, 98, 550–558. https://doi.org/10.1037/ a0018933 Carey, R. N., Connell, L. E., Johnston, M. et al. (2019). Behavior change techniques and their mechanisms of action: A synthesis of links described in published intervention literature. Annals of Behavioral Medicine, 53, 693–707. https://doi.org/10.1093/abm/kay078 Caudwell, K. M., Keech, J. J., Hamilton, K., Mullan, B. A., & Hagger, M. S. (2019). Reducing alcohol consumption during pre-drinking sessions: Testing an integrated behaviour-change model. Psychology and Health, 34, 106–127. https://doi .org/10.1080/08870446.2018.1518527 Chatzisarantis, N. L. D., & Hagger, M. S. (2009). Effects of an intervention based on self-determination theory on self-reported leisure-time physical activity participation. Psychology and Health, 24, 29–48. https://doi.org/10.1080/08870440701809533 Connell, L. E., Carey, R. N., de Bruin, M. et al. (2018). Links between behavior change techniques and mechanisms of action: An expert consensus study. Annals of Behavioral Medicine. https://doi.org/ 10.1093/abm/kay082 Conner, M. T., & Norman, P. (2015). Predicting and changing health behaviour: A social cognition approach. In M. T. Conner & P. Norman (Eds.), Predicting and Changing Health Behaviour: Research and Practice with Social Cognition Models (3rd ed., pp. 1–29). Maidenhead: Open University Press. Davis, R., Campbell, R., Hildon, Z., Hobbs, L., & Michie, S. (2015). Theories of behaviour and behaviour change across the social and behavioural sciences: A scoping review. Health

Psychology Review, 9, 323–344. https://doi.org/ 10.1080/17437199.2014.941722 de Vries, H., Mesters, I., van de Steeg, H., & Honing, C. (2005). The general public’s information needs and perceptions regarding hereditary cancer: An application of the Integrated Change Model. Patient Education and Counseling, 56, 154–165. https://doi.org/10.1016/j.pec.2004.01.002 Fishbein, M., & Ajzen, I. (1975). Belief, Attitude, Intention and Behavior. Reading, MA: Addison-Wesley. Fishbein, M., Hennessy, M., Kamb, M. et al. (2001). Using intervention theory to model factors influencing behavior change: Project RESPECT. Evaluation and the Health Professions, 24, 363–384. Gerrard, M., Gibbons, F. X., Brody, G. H., Murry, V. M., Cleveland, M. J., & Wills, T. A. (2006). A theory-based dual-focus alcohol intervention for preadolescents: The strong African American families program. Psychology of Addictive Behaviors, 20, 185–195. Gerrard, M., Gibbons, F. X., Stock, M. L., Lune, L. S. V., & Cleveland, M. J. (2005). Images of smokers and willingness to smoke among African American pre-adolescents: An application of the prototype/willingness model of adolescent health risk behavior to smoking initiation. Journal of Pediatric Psychology, 30, 305–318. https://doi .org/10.1093/jpepsy/jsi026 Hagger, M. S. (2009). Theoretical integration in health psychology: Unifying ideas and complimentary explanations. British Journal of Health Psychology, 14, 189–194. https://doi.org/10.1348/ 135910708X397034 Hagger, M. S. (2014). Avoiding the “déjà-variable” phenomenon: Social psychology needs more guides to constructs. Frontiers in Psychology, 5, 52. https://doi.org/10.3389/ fpsyg.2014.00052 Hagger, M. S. (2019). Habit and physical activity: Theoretical advances, practical implications, and agenda for future research. Psychology of Sport and Exercise, 42, 118–129. https://doi.org/ 10.1016/j.psychsport.2018.12.007 Hagger, M. S., & Chatzisarantis, N. L. D. (2009). Integrating the theory of planned behaviour and self-determination theory in health behaviour: A meta-analysis. British Journal of Health

Changing Behavior Using Integrated Theories

Psychology, 14, 275–302. https://doi.org/10.1348/ 135910708X373959 Hagger, M. S., & Chatzisarantis, N. L. D. (2014). An integrated behavior-change model for physical activity. Exercise and Sport Sciences Reviews, 42, 62–69. https://doi.org/10.1249/ JES.0000000000000008 Hagger, M. S., Gucciardi, D. F., & Chatzisarantis, N. L. D. (2017). On nomological validity and auxiliary assumptions: The importance of simultaneously testing effects in social cognitive theories applied to health behavior and some guidelines. Frontiers in Psychology, 8, 1933. https://doi.org/10.3389/ fpsyg.2017.01933 Hagger, M. S., Koch, S., Chatzisarantis, N. L. D., & Orbell, S. (2017). The common-sense model of self-regulation: Meta-analysis and test of a process model. Psychological Bulletin, 143, 1117–1154. https://doi.org/10.1037/bul0000118 Hagger, M. S., Trost, N., Keech, J., Chan, D. K. C., & Hamilton, K. (2017). Predicting sugar consumption: Application of an integrated dualprocess, dual-phase model. Appetite, 116, 147– 156. https://doi.org/10.1016/j.appet.2017.04.032 Hall, P. A., & Fong, G. T. (2007). Temporal selfregulation theory: A model for individual health behavior. Health Psychology Review, 1, 6–52. https://doi.org/10.1080/17437190701492437 Hamilton, K., Cox, S., & White, K. M. (2012). Testing a model of physical activity among mothers and fathers of young children: Integrating selfdetermined motivation, planning, and theory of planned behavior. Journal of Sport and Exercise Psychology, 34, 124–145. https://doi.org/10.1123/ jsep.34.1.124 Hamilton, K., Gibbs, I., Keech, J. J., & Hagger, M. S. (2020). Reasoned and implicit processes in heavy episodic drinking: An integrated dual process model. British Journal of Health Psychology, 25, 189–209. https://doi.org/10.1111/BJHP.12401 Hamilton, K., Kirkpatrick, A., Rebar, A., & Hagger, M. S. (2017). Child sun safety: Application of an integrated behavior change model. Health Psychology, 36, 916– 926. https://doi.org/10.1037/hea0000533 Hayes, A. F. (2018). Introduction to Mediation, Moderation, and Conditional Process Analysis: A Regression-Based Approach (2nd ed.). New York: Guilford Press.

223

Head, K. J., & Noar, S. M. (2014). Facilitating progress in health behaviour theory development and modification: The reasoned action approach as a case study. Health Psychology Review, 8, 34–52. https://doi.org/10.1080/17437199.2013.778165 Kasprzyk, D., Montaño, D. E., & Fishbein, M. (1998). Application of an integrated behavioral model to predict condom use: A protective study among high HIV risk groups. Journal of Applied Social Psychology, 28, 1557–1583. https://doi.org/ 10.1111/j.1559-1816.1998.tb01690.x Kerlinger, F. N., & Lee, H. B. (1999). Foundations of Behavioral Research. Belmont, CA: Wadsworth. Kruglanski, A. W., Shah, J. Y., Fishbach, A., Friedman, R., Chun, W. Y., & Sleeth-Keppler, D. (2002). A theory of goal systems. In M. P. Zanna (Ed.), Advances in Experimental Social Psychology, Vol. 34 (pp. 331– 378). San Diego, CA: Psychology Press. Lawton, R., Conner, M. T., & McEachan, R. (2009). Desire or reason: Predicting health behaviors from affective and cognitive attitudes. Health Psychology, 28, 56–65. https://doi.org/10.1037/a0013424 Maher, J. P., & Conroy, D. E. (2016). A dual-process model of older adults’ sedentary behavior. Health Psychology, 35, 262–272. https://doi.org/10.1037/ hea0000300 McMillan, B., & Conner, M. (2007). Health cognition assessment. In A. B. S. Ayers, C. McManus, S. Newman, K. Wallston, J. Weinman, & R. West (Ed.), Cambridge Handbook of Psychology, Health and Medicine (2nd ed., pp. 260–266). Cambridge: Cambridge University Press. Michie, S., Ashford, S., Sniehotta, F. F., Dombrowski, S. U., Bishop, A., & French, D. P. (2011). A refined taxonomy of behaviour change techniques to help people change their physical activity and healthy eating behaviours: The CALO-RE taxonomy. Psychology and Health, 26, 1479–1498. https:// doi.org/10.1080/08870446.2010.540664 Milne, S., Sheeran, P., & Orbell, S. (2000). Prediction and intervention in health-related behavior: A metaanalytic review of protection motivation theory. Journal of Applied Social Psychology, 30, 106–143. https://doi.org/10.1111/j.1559-1816.2000.tb02308.x Montaño, D. E., & Kasprzyk, D. (2015). Theory of reasoned action, theory of planned behavior, and the integrated behavioral model. In K. Glanz, B. K. Rimer, & K. Viswanath (Eds.), Health Behavior and

224

MARTIN S. HAGGER AND KYRA HAMILTON

Health Education: Theory, Research, and Practice (5th ed., pp. 95–124). San Francisco, CA: JosseyBass. Noar, S. M., & Zimmerman, R. S. (2005). Health behavior theory and cumulative knowledge regarding health behaviors: Are we moving in the right direction? Health Education Research, 20, 275–290. https://doi.org/10.1093/her/cyg113 Payaprom, Y., Bennett, P., Alabaster, E., & Tantipong, H. (2011). Using the health action process approach and implementation intentions to increase flu vaccine uptake in high risk Thai individuals: A controlled before-after trial. Health Psychology, 30, 492–500. https://doi.org/10.1037/a0023580 Perugini, M., & Bagozzi, R. P. (2001). The role of desires and anticipated emotions in goal-directed behaviours: Broadening and deepening the theory of planned behavior. British Journal of Social Psychology, 40, 79–98. https://doi.org/10.1348/ 014466601164704 Perugini, M., & Conner, M. (2000). Predicting and understanding behavioral volitions: The interplay between goals and behaviors. European Journal of Social Psychology, 30, 705–731. https://doi.org/ 10.1002/1099-0992%28200009/10%2930% 3A53.0.CO%3B2-%23 Petty, R. E., & Cacioppo, J. T. (1986). Communication and Persuasion: Central and Peripheral Routes to Attitude Change. New York: Springer. Prentice-Dunn, S., & Rogers, R. W. (1986). Protection motivation theory and preventive health: Beyond the health belief model. Health Education Research, 1, 153–161. https://doi.org/10.1093/her/1.3.153 Presseau, J., Johnston, M., Heponiemi, T. et al. (2014). Reflective and automatic processes in health care professional behaviour: A dual process model tested across multiple behaviours. Annals of Behavioral Medicine, 48, 347–358. https://doi .org/10.1007/s12160-014-9609-8 Prestwich, A., Lawton, R., & Conner, M. (2003). The use of implementation intentions and the decision balance sheet in promoting exercise behaviour. Psychology and Health, 18, 707–721. https://doi .org/10.1080/08870440310001594493 Protogerou, C., Johnson, B. T., & Hagger, M. S. (2018). An integrated model of condom use in subSaharan African youth: A meta-analysis. Health

Psychology, 37, 586–602. https://doi.org/10.1037/ hea0000604 Rhodes, R. E., & de Bruijn, G. J. (2013). How big is the physical activity intention-behaviour gap? A meta-analysis using the action control framework. British Journal of Health Psychology, 18, 296– 309. https://doi.org/10.1111/bjhp.12032 Rhodes, R. E., & Gray, S. M. (2018). Affect in the process of action control of health-protective behaviors. In D. M. Williams, R. E. Rhodes, & M. Conner (Eds.), Affective Determinants of Health Behavior (pp. 21–47). New York: Oxford University Press. Rogers, R. W. (1975). A protection motivation theory of fear appeals and attitude change. Journal of Psychology, 91, 93–114. https://doi.org/10.1080/ 00223980.1975.9915803 Rosenstock, I. M. (1974). Historical origins of the health belief model. Health Education Monographs, 2, 328–335. Trafimow, D. (2012). The role of auxiliary assumptions for the validity of manipulations and measures. Theory and Psychology, 22, 486–498. https://doi .org/10.1177/0959354311429996 Triandis, H. C. (1977). Interpersonal Behavior. Monterey, CA: Brookes/Cole. van Lettow, B., de Vries, H., Burdorf, A., & van Empelen, P. (2016). Quantifying the strength of the associations of prototype perceptions with behaviour, behavioural willingness and intentions: A meta-analysis. Health Psychology Review, 10, 25–43. https://doi.org/10.1080/ 17437199.2014.941997 West, R., Godinho, C. A., Bohlen, L. C. et al. (2019). Development of a formal system for representing behaviour-change theories. Nature Human Behaviour, 3, 526–536. https://doi.org/10.1038/ s41562-019-0561-2 Zhang, C. Q., Zhang, R., Schwarzer, R., & Hagger, M. S. (2019). A meta-analysis of the health action process approach. Health Psychology, 38, 623– 637. https://doi.org/10.1037/hea0000728 Zhou, G., Sun, C., Knoll, N., Hamilton, K., & Schwarzer, R. (2015). Self-efficacy, planning and action control in an oral self-care intervention. Health Education Research, 30, 671–681. https:// doi.org/10.1093/her/cyv032

16 Changing Behavior Using Social Identity Processes Katherine J. Reynolds, Nyla R. Branscombe, Emina Subašić, and Loren Willis

Practical Summary There are several explanations of behavior and change strategies that focus solely on the psychology of the person as an isolated individual. Building on these efforts, this chapter focuses on people’s self-definitions as group members – their social identities and how these shape behavior. Once people’s social identities are “active,” they come to see themselves in terms of the values and beliefs that define the group and act in line with relevant group norms. This analysis of social behavior change opens up new pathways where social and psychological changes that serve to redefine who “we” are and what “we” do – our social selves – can bring about behavioral change. Translating these ideas to applied settings and practices, it is possible that programs and interventions, for example, in work, education, and community settings, that alter the way people identify with groups also will alter the ways in which they behave. Interest in these social identity processes is growing but large-scale field experiments are still needed. This chapter highlights key work that has been done to date and avenues for future advancement.

16.1 Introduction There are many urgent problems that face humanity – from social conflict to ill health and energy overconsumption. Given the causal role of human behavior across these issues, it is timely to focus on the question of behavior change. Psychology has a major role to play as the foundational discipline focused on understanding the human mind and behavior. It is assumed that through understanding the causes of human behavior – what drives a person’s decisions, judgments, and preferences – it will be possible to gain insights into how to shape it. What is often overlooked, however, is that the challenges faced in behavior change go well beyond individual-level drivers and motivation.

The social and behavioral sciences frequently focus on change strategies that are tied to the psychology of the individual in isolation such as awareness campaigns, incentivization, and selfinterest (see also Chapters 28 and 30, this volume). Building on these efforts, this chapter presents theory and research related to people’s group membership and social identification. People live, work, and act in a socially structured system, where there are group-based regularities of language, symbols, perception, cognition, and conduct, and this reality has psychological consequences (Turner & Oakes, 1997). There is a need to recognize that the https://doi.org/10.1017/9781108677318.016

226

´ , AND LOREN WILLIS K A T H E R I N E J . R E Y N O L D S , N Y L A R . B R A N S C O M B E , E M I N A S U B A Sˇ I C

human cognitive system has developed in concert with the experiences of group life and therefore the mind has distinctive “social” capabilities that are key to understanding behavior and behavior change (see also Chapter 44, this volume). This chapter offers an analysis of these social processes with a focus on social behavior change, including social identity, in-group norms, and social influence. Key theoretical ideas are outlined as well as recent research that highlights the contribution of social identity processes to understanding not just behavior but also behavior change. The aim is to offer insights into new pathways for behavior change that may be particularly relevant to policy makers and in areas where what “we” do now needs to change as a way to solve the urgent challenges ahead.

16.2 Brief Overview of Core Constructs 16.2.1 The Existence and Operation of Identity Processes Lewin (1951) recognized that it “is usually easier to change individuals formed into a group than to change any one of them separately” (p. 228). More than sixty-five years later social identity theory (SIT; Tajfel & Turner, 1979) and selfcategorization theory (SCT; Turner et al., 1987; referred to as the social identity perspective/ approach) offer many insights into why and how this is the case. Central to the social identity perspective is the notion of psychological groups to which an individual can not only objectively belong (e.g., gender, ethnicity, nationality, work team, neighborhood, club) but also perceives themselves as being a member, because the group is self-relevant and self-defining (Turner et al., 1987; see Sidebar 16.1 for definitions of social identity theoretically related terms). Individuals can view themselves in terms of idiosyncratic personal attributes (e.g., a personal identity) or in terms of social categories (e.g.,

social identities, such as “Australian,” “woman,” “academic”). The concept of personal identity or the personal self (“I”) is used to describe situations where individuals perceive themselves to be distinct and different from others (that are available for comparison) and social identity or the social self (“we”) refers to an individual’s “knowledge that he [or she] belongs to certain groups together with some emotional and value significance to him [or her] of the group membership” (Tajfel, 1972, p. 31). People behave differently depending on whether they define themselves as “I and me” or “we and us.” Therefore, an important mechanism for change is how people define themselves in a given situation; as self-categorization varies from personal to social levels, and from one social identity to another, so too can beliefs, expectations, and behavior. As people’s perceptions of themselves shift, for example from “American” to “woman” or vice versa, a different set of ingroup norms and values becomes self-defining and, in turn, one’s behavioral repertoire shifts to reflect who “we” are now (e.g., Turner et al., 1994). Along these lines, Trafimow et al. (1997) have demonstrated that bilinguals using either their English or Chinese language leads to differences in their self-concepts and choices (see also Sidebar 16.2). The self can and does change and therefore so too can attitudes, emotions, and behavior. When people self-categorize or identify with a particular in-group, the norms, values, and beliefs that define the group are internalized and as such shape the attitudes and behavior of group members (Turner, 1991; see Sidebar 16.3). The stronger a person’s identification with a particular group, the more likely it is that they will behave intrinsically in ways that align with the characteristics that define the group – the in-group’s norms (see Reynolds, Subašić, & Tindall, 2015). An example of the importance of social identification processes for behavior and behavior change is provided by Folk et al. (2015) who examined predictors of

Changing Behavior Using Social Identity Processes

227

Sidebar 16.1 Brief definitions of theoretical terms used in social identity approaches

Deviance. When a group member’s behavior departs from what is normative or expected within the group. Group identification. An individual’s cognitive and emotional attachment to a particular group. In-group. A group that is perceived to be self-defining and self-relevant in a particular context. In-group norms. Knowledge that is shared by members of a particular group that serve to characterize or define the group and guide members’ thoughts, feelings, and actions. Internalization. An individual’s private acceptance of a set of behavioral norms, values, and beliefs. Out-group. A group that is perceived to be non–self-defining in a particular context. Personal identity. Individuals’ knowledge that they are different from other people, together with some emotional and valued significance to them of the sense of individuality. Prototypical. A position within the in-group that best exemplifies its behaviors, values, and beliefs relative to an out-group. Self-categorization. The process of perceiving the self as an interchangeable member of a category that is defined at a particular level of abstraction (e.g., personal, social, or human; Haslam, 2004). Social identity. A person’s knowledge that they belong to certain social groups and that these groups have some cognitive, emotional, and/or valued significance (Tajfel, 1972). Social influence. The process of privately altering and transforming an individual’s or group’s behaviors, values, or beliefs such that they come to be one’s own. Social norms. The shared rules and expectations within society regarding the acceptable behaviors, values, and beliefs of its members. Social referents. Social referents are prominent in social network analyses and are defined as “individuals who have greater knowledge of typical or desirable behavioral patterns within the community” and, related to this, attract greater attention from other community members (Paluck, Shepard, & Aronow, 2016, pp. 566–567). Stereotyping. A cognitive representation of a group, typically in terms of traits and attributes, that is shared by members of that group or by members of another group.

criminal reoffending and community adjustment (e.g., stable housing, training, employment, savings, volunteerism) among former prisoners who

had been released back into the community. Social identification with the criminal community and the wider community in general was assessed.

228

´ , AND LOREN WILLIS K A T H E R I N E J . R E Y N O L D S , N Y L A R . B R A N S C O M B E , E M I N A S U B A Sˇ I C

Sidebar 16.2 Self-definition varies with shifts from personal to social identity

One of the clearest examples of an intervention that shifts people’s self-concept from personal to social identity is provided by Onorato and Turner (2004). Their research examines the effects of independent and dependent schematics; classifications that, like personality constructs, are considered “relatively unresponsive to changes in one’s social circumstances” (Markus & Wurf, 1987, p. 306). In the study, participants initially were classified as independent schematics (i.e., they possessed core selfschema for being independent) and dependent schematics (i.e., they possessed core self-schema for being dependent). Then, as part of a supposedly unrelated study, participants were placed in an intergroup context in same-sex discussion groups to discuss the degree to which stereotypical traits were more typical of one gender group compared to the other. It was found that dependent male schematics, when social identity was salient, now behaved as independent schematics and independent female schematics behaved as dependent schematics (see also Guimond et al., 2007; Ryan, David, & Reynolds, 2004). Irrespective of participants’ personal self-schemata, both males and females tended to rate themselves in terms of their own gender stereotypes revealing variability in their self-definition. This study shows that it is possible to change people’s self-definition from the personal to the social identity level by making changes to the context such as introducing a comparison group and a particular topic for discussion (gender stereotypes).

The findings indicated that the more the self was overlapping with the criminal community – indicating stronger identification with that group – the higher the reoffending rate one year later. Those who reported more overlap with the wider community had better community adjustment across time. This example demonstrates the importance of social identification in behavior change and how the direction of such change is tied to identity meaning; the qualities that define what it means to be a group member (behaviors, expectations, conduct).

16.2.2 In-Group Norms and Social Influence Social influence processes also are argued to be important in explaining the relationship between social identity and behavior. Individuals belonging to an in-group view other members as being similar to themselves and, furthermore, to hold the correct, appropriate, and right views about

relevant matters (Turner, 1991). This perceived similarity underpins persuasion and influence. There is an expectation that within the in-group there will be a shared view that then motivates people to work through disagreement toward agreement. Resolving differences in perspective (e.g., criticism, new ideas, deviance) emerges from within a higher-order sense of commonality and similarity based on shared social identity. Psychological similarity with others as in-group members and associated social influence provide a mechanism for explaining how who “we” are can shape who “I” am and, centrally, what “I” do. Such insights consider all levels of psychological functioning – individual and group – and can be used to inform attempts to bring about lasting behavior change (Mols et al., 2015; Reynolds et al., 2015). To date, social-identity related research has focused mainly on explaining behavior. In the next section, more recent developments that concern social behavior change are outlined.

Changing Behavior Using Social Identity Processes

16.3 Evidence of Social Behavior Change: Social Identity, InGroup Norms, and Social Influence Processes Much theory and research on social identity and variability in perceptions, attitudes, or behaviors has focused on (1) shifts in the salience of personal and/or social identities and its implications (see Sidebar 16.2); (2) influence attempts from either an in-group or out-group source and/or the strength of identification with a source (in-group, leaders; see Sidebar 16.3); and (3) assessing the relationship between changes in social identity and changes in behavior. Given that this latter research is more recent and relevant to the themes of this handbook, it will be described in more detail.

16.3.1 Group Change to Change Behavior In people’s social lives, there can be changes in the number or type of group memberships, what it means to be a member of the group (identity content), and the strength of psychological connections to groups (group or social identification). These changes can be self-driven. People join new groups often through changes in social roles (parent, wife, manager), by choice or recommendation (e.g., social prescriptions by doctors to improve health and well-being). A longitudinal study by Haslam and colleagues (2016) demonstrates the positive outcomes associated with joining a new group. These researchers have developed a group-based intervention called Groups4Health (G4H) and investigated the impact on the well-being of participants presenting with social isolation and affective disturbance. They assessed identification with a new therapy group and other meaningful groups as well as depression, anxiety, stress, life satisfaction, self-esteem, and social connectedness (loneliness, social functioning) on three occasions

229

(pre-intervention, post-intervention, and sixweek follow-up). Included in these assessments is reference to behavior such as avoiding social activities and days missed at work. In line with predictions, for the majority of participants, there was improvement compared to baseline assessments taken prior to the intervention on all measures except for self-esteem, which remained relatively constant. Furthermore, increased identification with the G4H group and other groups was found to be significantly related to improved well-being across time. This research provides a pathway to address negative mental health and associated behavior through group memberships and improved social connections. It is also possible that external factors trigger changes in group memberships, which can come to have psychological potency. Negative treatment by others in society (e.g., stigma) can be the catalyst for the formation of similarity around a shared grievance (Branscombe, Schmitt, & Harvey, 1999). Political and social change and the associated leadership processes can also redefine intergroup boundaries, providing a context for the emergence of a “new” psychological ingroup (Reicher, Haslam, & Hopkins, 2005; Simon, 2004; Subašić, Reynolds, & Turner, 2008; Vestergren, Drury, & Chiriac, 2018). The key message is that, as group memberships and associated social identity change, so too can behavior.

16.3.2 Evidence for the Relationship Between Social Identity and Behavior Change A central prediction of this analysis of social behavior change is that changes in social identity and strength of social identification across time can lead to attitude and behavior change. Systematic demonstration of these relationships has significant implications for public policy in areas such as education and health. While large-

230

´ , AND LOREN WILLIS K A T H E R I N E J . R E Y N O L D S , N Y L A R . B R A N S C O M B E , E M I N A S U B A Sˇ I C

scale intervention studies are still needed, there are encouraging results from existing work. For example, Reynolds and colleagues have been studying schools, which are an important institution in which to build positive futures for young people. Since 2014, they have surveyed all schools in a local area (eighty plus) on an annual basis, with the ability to link responses from an individual participant across time making it a longitudinal design. The survey is completed by students in grades 5– 12 (from eleven to eighteen years), school staff, and parents and carers (e.g., Lee et al., 2017). More specifically, the aim is to investigate the role of group (school) characteristics, processes, and norms (referred to as the school climate, which includes shared mission, fair processes, quality relationships) and psychological identification with the school (school identification) in explaining and shaping student behavior (e.g., engagement, achievement, aggression) and well-being (depression, anxiety, resilience). Cross-sectional survey results on a pre-2014 sample of four schools have indicated that perceptions of a positive school climate are related to stronger student engagement in learning, academic achievement, and well-being and less student aggression. Academic achievement is measured through the completion of standardized tests, and students are asked about their behavior at school (e.g., “I try to complete my schoolwork on time and to the best of my ability”; “I pushed or shoved other students”). Key findings are that when students experience a positive school climate and they feel the school is a place where they belong, there are fewer instances of aggression and victimization (Turner et al., 2018) and there is higher academic achievement (Reynolds et al., 2017; replicated with the larger post-2014 sample: Maxwell et al., 2017). School climate and school identification explain approximately 9–16 percent of the variance in the behavior of interest. Importantly, in line with predictions, school identification was found to be a significant

mediating factor in explaining the relationship between school climate and these student outcomes (e.g., Bizumic et al., 2009). What this means is that qualities that define the climate of the school (characteristics, processes, and norms) come to have the most impact on behavior when students feel they belong and are connected psychologically to the school as a group. Such findings suggest school social identity processes play a central role in changing student behavior. Early indications from longitudinal analysis of this pre-2014 sample also provide support for the model. Turner et al. (2014) have shown that changes in behavior (e.g., aggression) are related to changes in perceptions of the school climate and increased social identification with the school across time (with this model fitting the data better than the reverse). These same patterns have been observed with respect to student engagement (“I try to complete my schoolwork on time and to the best of my ability”; Reynolds et al., 2019). Using a longitudinal mediation model across three annual waves of data collection, change in perceptions of school climate positively affected the growth in school identification, which, in turn, positively influenced change in students’ engagement. This longitudinal research supports a key prediction of the social identity perspective that changes in social identification across time can lead to attitude and behavior change.

16.3.3 How to Change Social Identity in Schools to Change Behavior Transforming aggregates of individuals into a psychological group is central to social behavior change. It is through this process of becoming part of a social group (e.g., volunteer group, sports or work team, organization, political party) that behavior is directly shaped. The school climate and school identification project has now advanced further to consider how to build psychological group memberships within schools. This serves as a prime example of how strategies

Changing Behavior Using Social Identity Processes

to change behavior using the social identity approach affect motivation and behavior change. There are a number of strategies that can be used to shape the meaning of who “we” are as a group and strength of identification. Many schools have sought to clarify their mission and build consensus across subgroups and individuals: staff, students, and interested parents and community members. These groups have been consulted so as to identify the vision, purpose, and ideal behaviors for staff and students within a particular school (e.g., Haslam, Eggins, & Reynolds, 2003). The extent to which members participate and are involved in planning and decisions affects their identification with the group and their willingness to enact new behavior (e.g., Haslam, 2004; Tyler & Blader, 2000). A whole range of school activities and functions are now shaped by this shared mission (e.g., codifying shared practices, celebration of

231

achievements, championing individuals who exemplify the school’s mission). At one school, to build awareness of its values and norms, rewards for appropriate behavior were introduced. A school certificate was awarded at the end of year 10 based on points students had gained because they acted in line with the school norms over a two-year period (Sugai et al., 2000). Deliberate efforts to recognize students for doing the right thing focus attention on positive behaviors and provide opportunities to reinforce who “we” are as a group. New norms can become accepted and internalized through these efforts. High school settings also have been used to assess norm-based behavior change in areas of bullying and tolerance. Paluck and colleagues (2016) argue that people want to behave in line with what others do but they misperceive the norm, which means they act in normdeviant ways. These researchers developed a

Sidebar 16.3 The importance of social norms and social identification

There is a growing body of research on how norms and social identification relate to changes in behavior. Duffy and Nesdale (2009) found that, while the incidence of bullying was higher for children belonging to groups with a norm supporting bullying in comparison to groups defined by an anti-bullying norm, rates of bullying were greater the more members identified with the group. Another example comes from interventions designed to address heavy or binge drinking at university. Neighbors et al. (2010) measured the perceived drinking norms for four groups of students on their campus, identification with each group, and participants’ own drinking behavior. They found that group identification, as defined by feeling closer to specific groups, moderated the associations between perceived drinking norms in the specific group and one’s own drinking behavior. Heavy drinking was related to the drinking norms within the group but mostly for those who identified more strongly with the group. With respect to assessing actual behavior, Cruwys et al. (2012) have demonstrated that observing the amount of food consumption (low or high) of an in-group or out-group other had an impact on one’s own eating behavior. Where there was a perceived shared group membership that was psychologically salient, individuals conformed to the in-group (but not out-group) norm (see also McGarty et al., 1994). Taken together, these results demonstrate that group norms and one’s psychological connection to the group are important for explaining both behavior and changing behavior.

232

´ , AND LOREN WILLIS K A T H E R I N E J . R E Y N O L D S , N Y L A R . B R A N S C O M B E , E M I N A S U B A Sˇ I C

school-based intervention that revolved around peer and school norms being more clearly communicated and clarified using different mediums (e.g., posters, social media, wearable reminders) and persuasive techniques. A distinctive feature of this research is the use of social network analysis to identify social referents (e.g., role models, agents of influence) who then are enlisted and trained by the researchers as “change agents” and who run activities within the school to promote the school norms (i.e., anti-conflict). It has been demonstrated that such interventions can lead to behavior change, especially for those with stronger network ties to the referents. While there is more work to be done connecting this research to interventions oriented to behavior change, emerging evidence clearly points to the centrality of social identity processes and ingroup norms in shaping behavior.

16.4 Future Directions for Research on Social Identity Processes and Behavior Change As people’s social identities change in a given situation or across time, so too should behavior. It has been proposed in this chapter that attitude and behavior change are a function of (1) changes in psychological group memberships; (2) changes in key group attributes (e.g., norms, goals); and/or (3) the strengthening (or weakening) of social identification with groups. To date, though, there are limited experimental and longitudinal examples of such work, despite its importance for public policy in education and health. Some immediate directions for research are considered in this final section of the chapter. In terms of advancing theory on social identity, there is a need for better integration across existing models. Social norms, for example, feature in many of the dominant models in social psychology, including the theory of planned behavior (e. g., Ajzen, 1991; see also Chapter 2, this volume) and social norm marketing (e.g., Perkins et al.,

2010). There is a growing body of research where group identification (e.g., “how much do you identify with [target group]?”) is measured along with theory of planned behavior constructs (attitudes, subjective norms, and perceived behavioral control). Group identification is expected to moderate the relationship between subjective norms and behavioral intentions and/or behavior. Along these lines, Terry and Hogg (1996) found that the perceived norms of the relevant reference group influenced intentions to engage in regular exercise and use sunscreen but only for individuals who identified strongly with the in-group. Recently, Willis and colleagues have considered another social identity dimension – identity content – which is defined as the level of connection between the behavior of interest and what it means to be a group member (group stereotype). Behaviors that are defining of the group will be prototypical, that is, they denote what it means to be a group member and differentiate the in-group from other out-groups (Hagger et al., 2007; Hogg & Reid, 2006; Reynolds et al., 2015; Willis et al., 2020). Results indicate that, in the context of binge drinking among university students, over and above typical constructs assessed within the theory of planned behavior and group identification, identity content (e.g., to what extent is drinking an important part of being a student at this university) was a significant predictor of binge drinking behavior. There is more work to be done connecting this research to interventions oriented to behavior change. Another direction for future research is to develop better models to explain the relationship between identity continuity and transition and behavior change. A starting point is to systematically examine critical change points in people’s lives where personal and social identities are likely to be disrupted. It is possible to envisage systematic study of the impact of role changes (e. g., promotion, divorce, unemployment, loss of a loved one; Specht, Egloff, & Schmukle, 2011) or self-initiated personal change efforts (therapy,

Changing Behavior Using Social Identity Processes

joining a walking group, volunteering) on behavior change. Group-level psychological factors (identification, in-group norms) need to be assessed at these change points (Reynolds & Branscombe, 2015) in order to map social identity change and behavior change processes. While attention has been directed to health and wellbeing (e.g., Jetten, Haslam, & Haslam, 2012), this analysis should be extended to other domains. There is also much to learn about societal change processes and links to social identity change and behavior change. In these settings, there is likely to be the emergence of new groups and leadership where social connections and networks are reconfigured (e.g., Guimond & de la Sablonniere, 2015). For example, how do national crises (e.g., economic downturns, war) and political movements (e.g., women’s rights, pro-abortion, gay rights, immigration) shape social identities (Canetti & Lindner, 2015; Twenge, 2015) and, in turn, come to impact on behavior change. It is also possible that behavior may be difficult to change the more it is strongly linked to longheld and deep-rooted meaningful social identities or threatens the continuity of the group (Oyserman, 2007). There has been limited consideration of how different identities are reconciled, especially where there are multiple potentially conflicting social identities (Amiot et al., 2015). Disidentification or weakening of certain identifications may be necessary if other identities are to become self-defining and behavior is to change. Importantly, social psychologists also need to educate people about the importance of groups in shaping their behavior. Along these lines, the G4H intervention described in Section 16.3.1 incorporates a number of sessions where participants can become more aware of the role of groups in shaping their lives. Participants learn how to analyze their own group networks and areas that can be further developed and strengthened, identify ways to make the most of existing

233

networks, and manage social groups effectively. Awareness of the group forces that shape behavior can be empowering. Such knowledge about the importance of groups and social identity in shaping attitudes and behavior can also serve a protective function. To the degree that people are only attuned to individual-level drivers of, and motivation behind, behavior, and ignorant of group-based persuasion and influence, they potentially are open to exploitation. As the recent Cambridge Analytica scandal revealed, knowing (or inferring) personality profiles and individuals’ demographic characteristics, social network (friends, interest groups), and patterns of engagement (social media–based likes, reposts) can provide insights into group memberships and their likely psychological and behavioral significance. Such information offers marketing advantages where material designed to change minds and behavior (e.g., voting preferences) can be positioned as if originating from a liked, trusted, in-group source. It is through wider dissemination of the social identity processes outlined in this chapter that it will be possible to increase awareness and ensure active, rather than clandestine, engagement with groups as a means to behavior change.

16.5 Summary and Conclusions In this chapter, the aim has been to make a case for social behavior change through highlighting the importance of social identity, in-group norms, and social influence processes in explaining behavior and shaping it. An explanation of behavior and change strategies has been proposed that moves beyond the psychology of the person as an isolated individual to incorporate group connections. Most importantly, it has been argued that, as social identities and in-group norms change, so too can behavior – that by shifting definitions of who “we” are and what “we” do it is possible to shift what “I” do. These insights may be particularly relevant in addressing

234

´ , AND LOREN WILLIS K A T H E R I N E J . R E Y N O L D S , N Y L A R . B R A N S C O M B E , E M I N A S U B A Sˇ I C

complex social problems where it is necessary to affect large numbers of people effectively and efficiently. Important next steps include new research on social behavior change and more effectively harnessing these social identity insights to address the complex social issues faced by society.

References Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50, 179–211. https://doi.org/10.1016/ 0749-5978(91)90020-t Amiot C. E., de la Sablonniere R., Smith L. G. E., & Smith J. R. (2015). Capturing changes in social identities over time and how they become part of the self-concept. Social and Personality Psychology Compass, 9, 171–187. https://doi.org/ 10.1111/spc3.12169 Bizumic, B., Reynolds, K. J., Turner, J. C., Bromhead, D., & Subašić, E. (2009). The role of the group in individual functioning: School identification and the psychological well-being of staff and students. Applied Psychology: An International Review, 58, 171–192. https://doi.org/10.1111/j.14640597.2008.00387.x Branscombe, N. R., Schmitt, M. T., & Harvey, R. D. (1999). Perceiving pervasive discrimination among African Americans: Implications for group identification and well-being. Journal of Personality and Social Psychology, 77, 135– 149. https://doi.org/10.1037//00223514.77.1.135 Canetti, D., & Lindner, M. (2015). Exposure to political violence and political behavior: Psychological mechanisms of transformation. In K. J. Reynolds & N. R. Branscombe (Eds.), The Psychology of Change: Life Contexts, Experiences, and Identities (pp. 77–89). New York: Psychology Press. Cruwys T., Platow M. J., Angullia S. A. et al. (2012). Modelling of food intake is moderated by salient psychological group membership. Appetite, 58, 754–757. https://doi.org/10.1016/j.appet.2011 .12.002

Duffy, A. L., & Nesdale, D. (2009). Peer groups, social identity, and children’s bullying behavior. Social Development, 18, 121–139. https://doi.org/ 10.1111/j.1467-9507.2008.00484.x Folk, J. B., Mashek, D., Tangney, J., Stuewig, J., & Moore, K. E. (2015). Connectedness to the criminal community and the community at large predicts 1-year post-release outcomes among felony offenders. European Journal of Social Psychology, 46, 341–355. https://doi.org/10.1002/ ejsp.2155 Guimond, S., Branscombe, N. R., Brunot, S. et al. (2007). Culture, gender, and the self: Variations and impact of social comparison processes. Journal of Personality and Social Psychology, 92, 1118–1134. https://doi.org/10.1037/00223514.92.6.1118 Guimond, S., & de la Sablonniere, R. (2015). Psychological metamorphosis: Understanding the effects of institutions, roles, and dramatic social change on individuals. In K. J. Reynolds & N. Branscombe, (Eds.), The Psychology of Change: Life Contexts, Experiences, and Identities (pp. 73– 90). New York: Psychology Press. Hagger, M. S., Anderson, M., Kyriakaki, M., & Darkings, S. (2007). Aspects of identity and their influence on intentional behavior: Comparing effects for three health behaviors. Personality and Individual Differences, 42, 355–367. https://doi .org/10.1016/j.paid.2006.07.017 Hansen, N., & Postmes, T. (2013). Broadening the scope of societal change research: Psychological, cultural, and political impacts of development aid. Journal of Social and Political Psychology, 1, 273–292. https://doi.org/10.5964/jspp.v1i1.15 Haslam, C., Cruwys, T., Milne, M., Kan, C. H., & Haslam, S. A. (2016). Groups 4 Health: Evidence that a social-identity intervention that builds and strengthens social group membership improves mental health. Journal of Affective Disorders, 194, 188–195. https://doi.org/10.1016/j.jad.2016 .01.010 Haslam, S. A. (2004). Psychology in Organizations. London: SAGE. Haslam, S. A., Eggins, R. A., & Reynolds, K. J. (2003). The ASPIRe model: Actualizing social and personal identity resources to enhance

Changing Behavior Using Social Identity Processes

organizational outcomes. Journal of Organizational and Occupational Psychology, 76, 83–113. https://doi.org/10.1348/ 096317903321208907 Hogg, M. A., & Reid, S. A. (2006). Social Identity, selfcategorization, and the communication of group norms. Communication Theory, 16, 7–30. https:// doi.org/10.1111/j.1468-2885.2006.00003.x Jetten, J., Haslam, C., & Haslam, S. A. (Eds.) (2012). The Social Cure: Identity, Health and Well-Being. New York: Psychology Press. https://doi.org/ 10.4324/9781315648569-16 Lee, E., Reynolds, K. J., Subašic, E. et al. (2017). Development of a dual school climate and school identification measure – student (SCASIM-St). Contemporary Educational Psychology, 49, 91– 106. https://doi.org/10.1016/j.cedpsych.2017 .01.003 Lewin, K. (1951). Field Theory in Social Science: Selected Theoretical Papers. New York: Harper & Row. Markus, H., & Wurf, E. (1987). The dynamic selfconcept: A social psychological perspective. Annual Review of Psychology, 38, 299–337. https://doi.org/10.1146/annurev. ps.38.020187.001503 Maxwell, S., Reynolds, K. J., Lee, E., Subašic, E., & Bromhead, D. (2017). The impact of school climate and school identification on academic achievement: Multilevel modeling with student and teacher data. Frontiers in Psychology, 8, 2069. https://doi.org/10.3389/fpsyg.2017.02069 McGarty, C., Haslam, S. A., Hutchinson, K. J., & Turner, J. C. (1994). The effects of salient group memberships on persuasion. Small Group Research, 25, 267–293. https://doi.org/10.1177/ 1046496494252007 Mols, F., Haslam, S. A., Jetten, J., & Steffens, N. K. (2015). Why a nudge is not enough. European Journal of Political Research, 54, 81–98. https:// doi.org/10.1111/1475-6765.12073 Neighbors, C., LaBrie, J. W., Hummer, J. F. et al. (2010). Group identification as a moderator of the relationship between perceived social norms and alcohol consumption. Psychology of Addictive Behaviors, 24, 522–528. https://doi.org/10.1037/ a0019944

235

Onorato, R. S., & Turner, J. C. (2004). Fluidity in the self-concept: The shift from personal to social identity. European Journal of Social Psychology, 34, 257–278. https://doi.org/10.1002/ejsp.195 Oyserman, D. (2007). Social identity and self-regulation. In A. W. Kruglanski & E. T. Higgins (Eds.), Social Psychology: Handbook of Basic Principles (2nd ed., pp. 432–453). New York: Guilford Press. Paluck, E. L., Shepherd, H., & Aronow, P. M. (2016). Changing climates of conflict: A social network experiment in 56 schools. Proceedings of the National Academy of Sciences, 113, 566–571. https://doi.org/10.1073/pnas.1514483113 Perkins, H. W., Linkenbach, J. W., Lewis, M. A., & Neighbors, C. (2010). Effectiveness of social norms media marketing in reducing drinking and driving: A statewide campaign. Addictive Behavior, 35, 866–874. https://doi.org/10.1016/j .addbeh.2010.05.004 Reicher, S., Haslam, S. A., & Hopkins, N. (2005). Social identity and the dynamics of leadership: Leaders and followers as collaborative agents in the transformation of social reality. The Leadership Quarterly, 16, 547–568. https://doi .org/10.1016/j.leaqua.2005.06.007 Reynolds, K. J. & Branscombe, N. R. (2015). Advancing the psychology of change. In K. J. Reynolds & N. Branscombe, (Eds.), The Psychology of Change: Life Contexts, Experiences, and Identities (pp. 265–281). New York: Psychology Press. Reynolds, K. J., Lee, E., Subašic, E., Klik, K., & Bromhead, D. (2019). Social identity processes and person change: A longitudinal investigation of student engagement in learning. Unpublished manuscript, Australian National University. Reynolds, K. J., Lee, E., Turner, I., Bromhead, D & Subašic, E. (2017). How does school climate impact on academic achievement? An examination of social identity processes. School Psychology International, 38, 78–97. https://doi .org/10.1177/0143034316682295 Reynolds, K. J., Subašić, E., & Tindall, K. (2015). The problem of behavior change: From social norms to an ingroup focus. Social and Personality Psychology Compass, 9, 45–56. https://doi.org/ 10.1111/spc3.12155

236

´ , AND LOREN WILLIS K A T H E R I N E J . R E Y N O L D S , N Y L A R . B R A N S C O M B E , E M I N A S U B A Sˇ I C

Ryan, M. K., David, B., & Reynolds, K. J. (2004). Who cares? The effect of gender and context on the self and moral reasoning. Psychology of Women Quarterly, 28, 246–255. https://doi.org/10.1111/ j.1471-6402.2004.00142.x Simon, B. (2004). Identity in the Modern Society: A Social Psychological Perspective. Oxford: Blackwell. Specht, J., Egloff, B., & Schmukle, S. C. (2011). Stability and change of personality across the life course: the impact of age and major life events on mean-level and rank-order stability of the Big Five. Journal of Personality and Social Psychology, 101, 862–882. https://doi.org/10.1037/a0024950 Subašić, E., Reynolds, K. J., & Turner, J. C. (2008). The political solidarity model of social change: Dynamics of self-categorization in intergroup power relations. Personality and Social Psychology Review, 12, 330–352. https://doi.org/ 10.1177/1088868308323223 Sugai, G., Horner, R. H., Dunlap, G. et al. (2000). Applying positive behavior support and functional behavioral assessment in schools. Journal of Positive Behavior Interventions, 2, 131–143. Tajfel, H. (1972). La catégorization sociale [Social categorization]. In S. Moscovici (Ed.), Introduction à la psychologie sociale [Introduction to Social Psychology], Vol. 1 (pp. 272–302). Paris: Larousse. Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup conflict. In W. G. Austin & S. Worchel (Eds.), The Social Psychology of Intergroup Relations (pp. 33–47). Monterey, CA: Brooks/Cole. Terry, D. J., & Hogg, M. A. (1996). Group norms and the attitude-behavior relationship: A role for group identification. Personality and Social Psychology Bulletin, 22, 776–793. https://doi.org/ 10.1177/0146167296228002 Trafimow, D., Silverman, E. S., Fan, R. M.-T., & Fun Law, J. S. (1997). The effects of language and priming on the relative accessibility of the private self and the collective self. Journal of CrossCultural Psychology, 28, 107–123. https://doi.org/ 10.1177/0022022197281007 Turner, I., Reynolds, K. J., Lee, E., Subašić, E., & Bromhead, D. (2018). Understanding aggression

and victimization: Negative binomial modelling with supportive school climate, mental health, and social identity mediation. Translational Issues in Psychological Science, 4, 380–402. https://doi .org/10.1037/tps0000174 Turner, J. C. (1991). Social Influence. Buckingham: Open University Press. Turner, J. C., Hogg, M. A., Oakes, P., Reicher, S. D., & Wetherell, M. (1987). Rediscovering the Social Group: A Self-Categorization Theory. Oxford: Blackwell. Turner, J. C., & Oakes, P. J. (1997). The socially structured mind. In C. McGarty & S. A. Haslam (Eds.), The Message of Social Psychology: Perspectives on Mind in Society (pp. 355–373). Malden, MA: Blackwell. Turner, J. C., Oakes, P. J., Haslam, S. A., & McGarty, C. A. (1994). Self and collective: Cognition and social context. Personality and Social Psychology Bulletin, 20, 454–463. https://doi.org/10.1177/ 0146167294205002 Twenge, J. M. (2015). The age in which we live and its impact on the person. In K. J. Reynolds, & N. R. Branscombe (Eds.), Psychology of Change: Life Contexts, Experiences, and Identities (pp. 44–58). New York: Psychology Press. Tyler, T. R., & Blader, S. L. (2000). Cooperation in Groups: Procedural Justice, Social Identity, and Behavioral Engagement. Philadelphia: Psychology Press. https://doi.org/10.4324/9780203782842 Turner, I., Reynolds, K. J., Lee, E., Subašić, E., & Bromhead, D. (2014). Well-being, school climate and the social identity process: A latent growth model study of bullying perpetration and peer victimization. School Psychology Quarterly, 29, 320–335. https://doi.org/10.1037/spq0000074 Vestergren, S., Drury, J., & Chiriac, E. H. (2018). How collective action produces psychological change and how that change endures over time: A case study of an environmental campaign. British Journal of Social Psychology, 57, 855–877. https://doi.org/10.1111/bjso.12270 Willis, L., Lee, E., Klik, K., & Reynolds, K. J. (2020). The theory of planned behavior and social identity processes: A new look in the context of student binge drinking. Manuscript submitted for publication. Europe’s Journal of Psychology.

17 Changing Behavior Using Ecological Models Jo Salmon, Kylie D. Hesketh, Lauren Arundell, Katherine L. Downing, and Stuart J. H. Biddle

Practical Summary People’s behavior is influenced by factors at multiple levels. For example, people’s engagement in behaviors that are good or bad for their health can be influenced by their own preferences and beliefs, by whether their friends or family also engage in these behaviors, by their physical environment, by workplace policies, and by government mandates or laws. Ecological models have been developed to understand and influence human behavior at the individual, social, and environmental levels. These models are often paired with individual theories that focus on feelings, beliefs, and behaviors. However, few programs have focused on influences at the physical environment, organizational, and policy levels combined. More research on the effectiveness of programs that have focused on behavior change at all levels of the ecological model is needed. Future research should focus on how ecological models can be used to deliver interventions to all relevant groups to improve population reach.

17.1 Introduction Behavior change involves targeting or influencing many aspects of people’s lives. For example, changing people’s voting patterns in a political election not only will involve persuasion through information but could be strongly influenced by individuals’ social and economic circumstances as well as their local environment. Similarly, enabling people to eat healthy foods likely requires much more than persuasion and advertising. Pricing, attractiveness, knowledge, availability, and accessibility are all likely to have an important role. Companies pay a great deal of money to have their products placed in highly accessible locations in supermarkets (Thornton et al., 2012). For example, placing confectionery next to the checkout and creating demands from

children to purchase such products – so-called pester power – can be very frustrating for parents (see also Chapters 14 and 42, this volume). What these examples illustrate is that behavior change is likely to be enhanced through an approach that tackles multiple contexts and levels of influence. Typically, these have been described as “social ecological models,” widely used in the research and practice of behavior change. Ecology refers to how living organisms relate to each other and the environment around them. As the term “ecological” incorporates the social environment, the inclusion of the term “social” as a prefix is redundant. Consistent with Sallis and Owen (2015), the term “ecological models” is used in the current chapter. This chapter provides a brief https://doi.org/10.1017/9781108677318.017

238

JO SALMON ET AL.

outline of ecological models and their history and defines their purpose and important components. Evidence of the use and effectiveness of ecological models for understanding and changing behavior from early childhood to adulthood is summarized. In addition, the chapter presents examples of the use of ecological models in different fields, disciplines, and contexts – although the majority of examples are in the health behavior context. The chapter also identifies theories that ecological models have frequently been paired with. The strengths, limitations, and future research needs with respect to ecological models applied to behavior change are also discussed.

17.2 Brief Overview of Ecological Models The origins of contemporary ecological frameworks that are used for behavior change stem from psychological thinking at least as far back as the 1930s. The German-born psychologist Kurt Lewin is usually credited with the development of “ecological psychology” through his field theory approach published four years after his death (Lewin, 1951). His work fifteen years prior on “typological psychology” provides the foundation for recognizing person-environment interactions as important facets of human behavior (Lewin, 1936). Writing more than eighty years ago, Lewin (1936) referred to his nowfamous equation stating that behavior is a function of the person (P) and environment (E); “the behavior depends on the state of the environment and that of the person: B = f(PE). In this equation P and E are not independent variables” (p. 166). Lewin considered perceptions of the environment to be more important than direct effects. Barker’s (1968) “real-world” studies in ecological psychology at the Midwest Field Station concluded that it is not possible to predict human behavior without knowing the environmental setting or situation that the person is in. This is consistent with Gibsonian ecological psychology

(Gibson, 1979) or ecological dynamics (Davids et al., 2016). Gibson introduced the notion of “affordances,” which refer to the functional and relational properties of an environment. These could be positive or negative, depending on the people and context. For example, upward moving escalators in a train station offer some people the chance to rest but others the opportunity to get some exercise through walking or even choosing the adjacent stairs (see Chapter 14, this volume). Table 17.1 provides examples of ecological models that have been developed in behavior change research since Lewin’s seminal work. While there is no single or definitive ecological model, some models have been used more than others. For example, Bronfenbrenner’s (1977) “bio-ecological” model of micro-, meso-, exo-, and macro-system influences proposes that behavior can be influenced at multiple levels. A good example of the application of this approach to behavior change is that of reduction of smoking rates in high-income countries. Most levels of the model will have been targeted and affected, leading to significant shifts in social and cultural norms (macro-system) with respect to tobacco consumption. Following Bronfenbrenner’s approach, McLeroy et al.’s (1988) social ecological model has been frequently used in health promotion. There are two main concepts with this model: that there are multiple levels of influence on behavior and reciprocal causation. McLeroy and colleagues identified five levels of influence (Table 17.1); however, factors such as culture, social class, and economics are not accounted for. Albert Bandura also proposed the concept of “reciprocal determinism” where all three variables – behavior, person, environment – interact with each other in reciprocal ways. He criticized Lewin’s approach as being too simplistic; “personal and environmental factors do not function as independent determinants, rather they determine each other” (Bandura, 1977, p. 9). He argued that the relative influence of each factor will vary across settings and for different

Changing Behavior Using Ecological Models

Table 17.1 Examples of ecological models that have been developed and used in predicting and changing behavior Ecological Model

Components

Lewin (1936) Field theory

• • • •

Life space Environment Person Behavior

Barker (1968) • Behavior settings – Environmental/ecological physical and social psychology situations • Microsystems Bronfenbrenner (1977) • Mesosystems Ecological systems • Exosystems theory • Macrosystems • Behavior Bandura (1977; 1986) • Person Social learning theory Social cognitive theory • Environment (mainly social) • Reciprocal determinism • Intrapersonal McLeroy et al. (1988) • Interpersonal Social ecological model • Organizational • Community • Public policy • Personal attributes Stokols (1992) • Physical environments Social ecology model • Social environments for health promotion • Environments are multidimensional (i.e., varying levels of complexity and scale) • Human-environment interactions occur at multiple levels and are reciprocal

behaviors. Unlike Bandura’s (1977) social cognitive theory (see Chapter 3, this volume), a challenge for many ecological models is they are “high-level” general frameworks that do not explicitly explain or guide methods or interventions to change behavior. They have been defined as a “metaconcept,” which should guide research

239

and practice rather than generate specific hypotheses to be tested (Richard et al., 2011). Ecological models have been further developed to guide behavioral interventions. For example, Stokols’s (1992) social ecology model for health promotion recognizes that there are reciprocal relationships between individual behavior and the social and physical environments at multiple levels of influence (e.g., child, school, family, culture, whole populations). An example of a disciplinespecific application of an ecological model is seen in the youth physical activity promotion model (YPAP; Welk, 1999). The Y-PAP proposes “enabling” factors, such as an adolescent’s fitness, skills, access to facilities and programs, and environment; “predisposing” factors, which include self-efficacy, perceived competence, enjoyment, beliefs, and attitudes; and “reinforcing” factors, including family, peer, and coach influences. Moderating characteristics such as personal demographics and socioeconomic position are also considered in this model. Moreover, ecological models have been used to guide implementation research. For example, May et al. (2016) argue that a “benefit” of the lack of specificity of ecological models is their increased “elasticity,” which is important for adapting to different contexts. Context is a critical ingredient of successful translation and implementation of health promotion programs (Koorts et al., 2018). In spite of these adaptations, many ecological models fail to identify specific variables and how the broader levels of influence interact across levels to change behavior. To overcome these shortcomings, general ecological models have often been used in conjunction with behavior change theories.

17.3 How Have Ecological Models Been Used to Change Behavior? Ecological models have primarily been used to identify multiple levels of influence by exploring (usually theory-guided) correlates or determinants

240

JO SALMON ET AL.

1. Establish links between behavior and health

2. Measure behavior

3. Identify influences on behavior

4. Evaluate interventions

5. Translate research into practice

Figure 17.1 Behavioral epidemiology framework

of behaviors. The importance of correlates research for informing the development of interventions is highlighted by the behavioral epidemiology framework (Figure 17.1). This framework proposes a systematic sequence of five phases of studies on health-related behaviors, with the aim of leading to evidence-based interventions and translation (Sallis et al., 2000). Phase 3 of this framework, identifying influences on the behavior, suggests that demographic correlates (i.e., individual-level factors such as sex, age, ethnic group, socioeconomic position) can be useful for identifying those most in need of intervention (see Sidebar 17.1). Additionally, this phase suggests that modifiable psychological, social, and environmental factors that may influence behavior should be identified. Within behavior change research, ecological models have commonly been used in the earlier phases of the behavioral epidemiology framework, particularly phase 3, to help frame understanding of behaviors and targets for intervention. Interventions (phase 4 of the framework) typically utilize evidence derived from ecological models about how behavior is developed and influenced to identify targets and settings for behavior change strategies (see Chapters 19 and 21, this volume). Subsequently, interventions that target multiple levels of influence (i.e., multilevel interventions or programs) are developed; however, they are often paired with a behavior change theory to

guide the development of the interventions (see Chapters 18 and 28, this volume). As an example, Figure 17.2 outlines the correlates commonly associated with health behaviors. The behaviors include sedentary behaviors in young children (Hinkley et al., 2010), children (Salmon, Tremblay et al., 2011), and adults (O’Donoghue et al., 2016); physical activity in children (de Vet et al., 2011) and adults (Trost et al., 2002); healthy eating in children (de Vet et al., 2011); adolescent pregnancy (Corcoran, 1999); smoking cessation in adolescents and young adults (Cengelli et al., 2012); and nicotine replacement therapy for adolescent smoking cessation (King et al., 2018). Correlates of health behaviors during key time periods, such as after-school (Arundell et al., 2016) and in special population groups such as children with autism spectrum disorder (Jones et al., 2017), are also identified. The most commonly reported correlates are within the individual domain, with fewer in the social, physical, and policy environments (Bauman et al., 2012). Age, gender, and socioeconomic status (SES) are consistently related to participation in health behaviors (Figure 17.2). The existence of barriers is also related to participation in health behaviors, with some barriers common across behaviors (e.g., cost and access) and some behavior-specific (e.g., addiction to nicotine replacement therapy for smoking cessation). Behavioral history, knowledge, attitudes, and intentions are also behavior-specific and commonly reported. The role of peers and family in influencing behavior is evidenced within the social environment. Support to perform a behavior, coparticipation in the behavior, and behavioral beliefs of peers and family (e.g., importance of participating in the behavior, whether it is normal to participate in the behavior) are often reported as behavioral correlates across a variety of health behaviors. Rules and regulations are also important, particularly parental rules for children’s behaviors such as physical activity and sedentary behaviors. School and workplace policies in support of

Changing Behavior Using Ecological Models

241

Environmental Policies supporting health behavior

Area of residence (rural vs urban)

Access to facilities and services

Environment and facility aesthetics Social

Safety

Family situation/ composition

Family and peer behavioral beliefs

Rules and redulations

Family and peer support

Family and peer co-participation

Gender Age Socio-economic status (SES)

Barriers Behavioral knowledge

Individual

Behavioral history Behavioral attitudes

Behavioral intentions Body Mass Index (BMI)

Figure 17.2 Common correlates associated with health behaviors

health behaviors have also been shown to be related to healthy eating, physical activity, and sedentary behavior (Bauman et al., 2012). Although fewer correlates in the physical and policy environments have been consistently identified, the behavior setting plays an important role in participation in health behaviors. The behavior setting may refer to the workplace, neighborhood, school, or home, where policies, equipment, facilities, or services either encourage (e.g., sporting equipment at school, professional and education regulation and standards) or inhibit (small outdoor yard at home, unhealthy food options

available at work) participation in health behaviors. Although ecological models are typically paired with behavioral theories, the conceptual frameworks and theories underlying research and how these have been paired tend not to be explicitly described. Table 17.2 provides a summary of common pairings that have been reported or implied in published research, with some example references. An example of a study pairing an ecological approach with social cognitive theory (see Chapter 3, this volume) is presented in Sidebar 17.2.

242

JO SALMON ET AL.

Sidebar 17.1 Examples of research on the correlates of behavior using an ecological framework Study. Systematic review to identify factors related to nicotine replacement therapy for adolescent smoking cessation (King et al., 2018). Population. Children under eighteen years of age. Behavior. Smoking cessation. Overview of findings. Fifty-one articles identified factors from studies at each level of the social-ecological model: intrapersonal (n = 20); interpersonal (n = 2); organizational (n = 7); community (n = 11); public policy (n = 14). Study. Systematic review of factors influencing ethical issues in nursing practice to inform the development of a framework to discuss, debate, and create interventions to address ethical issues (Davidson et al., 2018). Population. Nurses. Behavior. Ethical considerations in nursing practice. Overview of findings. Eight dimensions that influence ethical practice are identified as individual factors (patients and families); individual factors (nurses); relationships between health care professionals; relationships between patients and nurses; organizational health care context; professional and education regulation and standards; community; and social, political, and economic.

Table 17.2 Pairings of ecological models and theories Theory Paired With

Target Behavior (Examples)

Organizational theory Social cognitive theory

School-based mental health services among children in poverty (Cappella et al., 2008) Obesity prevention trial in preschool children (Fitzgibbon et al., 2011) Physical activity promotion in African American preschoolers (Annesi et al., 2013) Obesity prevention and school readiness trial (Winter & Sass, 2011) Transform-Us! Physical activity and sedentary behavior program (Salmon, Arundell et al., 2011) Obesity prevention trial in preschool children (Fitzgibbon et al., 2011) Bullying and cyberbullying prevention program (Ortega-Baron et al., 2019)

Behavioral choice theory Self-determination theory Empowerment theory

17.4 Evidence Base for the Use of the Ecological Model in Changing Behavior In acknowledging the varying and interrelated levels of influence on behavior, a 2001 report from the Division of Health Promotion and Disease Prevention within the Institute of Medicine at the US National Academy of Sciences recommended an ecological approach

for behavior change interventions and for the development of recommendations (Committee on Capitalizing on Social Science and Behavioral Research to Improve the Public’s Health, 2001). An ecological approach that considers the individual and their environment is also supported by the World Health Organization (WHO) through their global recommendations for physical activity and health and their guide for population-based approaches to increase physical activity (WHO,

Changing Behavior Using Ecological Models

243

Sidebar 17.2 Example of how the ecological model is paired with social cognitive theory

Wang et al. (2016) described the development of the H2GO! study, a communitybased intervention to reduce sugar-sweetened beverage consumption among youth, guided by the ecological model and the social cognitive theory. As briefly described in Table 17.1, social cognitive theory (see Chapter 3, this volume) proposes that behavioral, cognitive/personal, and environmental factors interact with each other reciprocally to influence behavior (Bandura, 1986). The H2GO! study targeted theoretical constructs from social cognitive theory, including knowledge, attitudes (outcome expectations, self-efficacy), behavioral capabilities and skills (selfmonitoring, problem-solving, self-regulation skills), and behaviors. The strategies targeting these constructs, in the individual, social, and physical environment levels of the ecological model, are shown in Table 17.3. Table 17.3 Strategies and constructs targeted by the H2GO! intervention adapted from Wang et al. (2016) Theoretical Constructs Targeted Intervention Strategies and Activities Individual level Enactive mastery experiences Modeling, self-monitoring Persuasive communication Active learning Reinforcement Didactic instruction Social level Family support Peer modeling Group-based guided practice Physical environmental level Environmental restructuring

Behavioral Knowledge Attitudes Capabilities and Skills Behaviors ✓ ✓ ✓

✓ ✓ ✓ ✓ ✓

✓ ✓ ✓

✓ ✓ ✓

2004). The WHO’s report summarizing the interventions that are most effective at improving physical activity and diet also indicates that multicomponent interventions that consider the local context and environment are most successful (WHO, 2009). To examine how ecological models have been applied to physical activity determinants and intervention research over thirty years, Richard and

✓ ✓ ✓ ✓ ✓

✓ ✓ ✓ ✓

✓ ✓ ✓

✓ ✓ ✓ ✓

colleagues (2011) conducted an analysis of papers published in eight health promotion, health education, and physical activity/exercise/sport journals during three 2-year periods (1988–1990, 1998– 2000, 2007–2009). They concluded that, over time, ecological models are being integrated into such research more frequently, more levels of influence were being examined concurrently, and the level of influence being targeted was

244

JO SALMON ET AL.

increasing to where organizational and political targets were often included (Richard et al., 2011). In a review of 157 intervention papers published over two decades targeting nutrition, physical activity, smoking, sexual behavior, alcohol/ substance use, disease screening, and other behaviors, Golden and Earp (2012) found that these interventions typically focused on factors at the individual and interpersonal level, with few focusing on institutional, community, or policy factors. Fewer than 10 percent of studies used ecological models to inform their intervention design, and only 63 percent of articles focused on just one or two levels of the ecological model. Nutrition interventions were significantly more likely to focus on three or more levels than studies that focused on other health behaviors. Studies that focused on the school setting were also more likely to include multiple levels compared to studies that targeted behavior change in, for example, health care, community, or family settings. The fidelity and success of the interventions were not reported in this review. Cushing and colleagues (2014) conducted a systematic review of health promotion interventions that used ecological models targeting child and adolescent smoking, physical activity, and diet and also performed a meta-analysis to determine the short- and longer-term impact of these studies on these health behaviors. There were ninety-six independent samples included in the review, which found an overall averaged standardized mean difference of g = 0.20. Although this effect is considered “small,” it was also evident, although smaller, at approximately twelve months’ follow-up (g = 0.07). There were strongest effects among dietary interventions that targeted schools and communities (g = 0.71) and among physical activity interventions that targeted individuals and families (g = 0.44). There were insufficient smoking interventions to allow comparisons of ecological model components. Ecological models have been used to identify the characteristics of interventions targeting

individual behaviors (e.g., tobacco control, Richard et al., 2002; breast and cervical cancer screening, Holden et al., 1998; unhealthy weight gain, Kellou et al., 2014), from which domains of the ecological model they belong, and where multiple domains are targeted. For example, Kellou et al. (2014) reviewed the factors within interventions promoting physical activity to prevent unhealthy weight among children (six to twelve years) using an ecological model. Based on fifty-four studies published in the previous five years, 43 percent of interventions targeted individual and/or interpersonal determinants of physical activity only, 48 percent targeted three or four domains including at least one environmental determinant (at the institutional level), and 9 percent were multilevel interventions within the community. The review concluded that intervention programs that target factors at multiple levels of the ecological framework have the most potential for preventing obesity (Kellou et al., 2014). Example intervention strategies to address targets in different levels of the ecological model are shown in Table 17.4 and Sidebar 17.3.

17.5 Summary of the Overall Evidence Research has identified the ecological correlates of behaviors based on ecological models (e.g., King et al., 2018), and the models have been used to inform and provide a structure for intervention strategies and targets. The most commonly reported correlates are within the individual domain (e.g., barriers, knowledge, intentions, attitudes), with fewer correlates identified in the social (e.g., family and peer support and beliefs, family structure, rules and regulations), physical (e.g., access to facilities, area of residence, safety), and policy environments (e.g., supportive health behavior policies in the workplace or school settings). Evidence suggests that targeting all domains within ecological models holds the greatest

Changing Behavior Using Ecological Models

245

Table 17.4 Example intervention targets and strategies across the domains of the ecological model Ecological Model Level

Intervention Targets

Intervention Strategies

Individual

Knowledge Attitudes Intentions Self-efficacy Social support (practical and emotional) Attitudes of social networks Attitudes of institution/school members/leaders Institutional/school culture Institutional/school policies Institutional/school capacity Community capacity Community services Home physical environment Institutional physical environment Community physical environment Attitudes of policy makers Capacity for advocacy Social norms Public policy

Education Skill enhancement Goal-setting Self-monitoring Education/skill enhancement for social network Education/skill enhancement for institution/school members/leaders Changes to institutional/school policies or services Education/skill enhancement for general community

Social

Physical environment

Policy

Changes to home environment Changes to institutional environments Changes to community environments Changes to community services Education/skill enhancement for policy makers Development (or modification) of public policies

Sidebar 17.3 Examples of interventions that have used an ecological framework

The Transform-Us! Intervention Study. Transform-Us! (Carson et al., 2013; Salmon, Arundell et al., 2011; Yildirim et al., 2014). Population. Children aged between nine and eleven years. Behavior(s). Physical activity and sedentary behavior via a thirty-month school- and home-based intervention. Setting. School and home. Theory. Social cognitive theory, behavioral choice theory, and ecological systems theory.

Overview of the Intervention at Each Level of the Framework Individual. Children’s self-efficacy, enjoyment, behavioral self-management and monitoring, and behavioral expectations and expectancies were targeted through classroom teacher–delivered key learning class lessons, standing class lessons, short active breaks, and active homework tasks.

246

JO SALMON ET AL.

Social. Modeling of physical activity and reductions in sedentary behavior by teachers, parents, and siblings was encouraged through teacher professional development sessions and parental newsletters. Parental enforcement of screen time rules was also promoted via newsletters. Environmental. The school environment was targeted via provisions of asphalt line markings, additional sporting and active play equipment, and increases in opportunities for physical activity. Strategies targeting the home environment included increasing the availability of physical activity equipment and reducing opportunities for sedentary behaviors and access to screens (e.g., TV, computer, electronic games). Outcomes. Reductions in children’s sedentary time and improvements in their physical activity at eighteen months and thirty months. Improvements to children’s body mass index (BMI), blood pressure, and some biomarkers were also evident at eighteen months. Transform-Us! is now being upscaled into a statewide program available to all primary (elementary) schools in Victoria, Australia.

The Stand Up Victoria Intervention Study. Stand Up Victoria (Healy et al., 2016). Population. Adults aged eighteen to sixty-five years. Behavior. Sitting. Setting. Workplaces. Theory. Social cognitive theory.

Overview of the Intervention at Each Level of the Framework Individual. Participants’ knowledge, attitudes, and motivation were targeted using staff information sessions, written materials, individual coaching, support phone calls, and selfmonitoring tools. Social (Organizational-Level). Workplaces received management consultation, team champion training, staff information sessions, and management emails to staff to promote changes in the organization culture and policy. Environmental. Participants were provided with sit-to-stand workstations to encourage reduced sitting. Outcomes. Significant reductions in workplace sitting time, prolonged sitting time (≥ 30 minutes), and sitting time during the entire day (work time, non–work time, and non-workdays) at intervention conclusion (three months) and sustained after twelve months.

potential for behavior change (e.g., Kellou et al., 2014; WHO, 2009). Challenges using the ecological model for behavior change include (Richard & Gauvin, 2018) the following: •

Changing the physical environment and policy environment can be difficult and time-

• •

consuming (hence the large focus to date on the intrapersonal level) Multilevel interventions are time- and resourceintensive to develop and deliver Evaluating interventions targeting multiple levels of the ecological model can be challenging due to the complex interrelated nature of the levels

Changing Behavior Using Ecological Models

17.5.1 Recommendations on How Ecological Models Should Be Used to Change Behavior A key challenge remains the partial use of ecological models in interventions. Hawe (2015) cites research arguing that “‘partial paradigm acquisition’ was taking place in prevention science, such as the use of ecological theory in name more than in substance” (p. 309). Behavior change interventions based on ecological models, therefore, need to target all levels of the model. It is also acknowledged that ecological models lack specific components to guide intervention development (Bartholomew et al., 1998). It is recommended that future interventions clearly identify any behavioral theories that have been used in conjunction with ecological models when developing interventions (Bartholomew et al., 1998).

17.5.2 Future Research While the focus of many approaches to behavior change is on individual change, a greater challenge is to consider how ecological models can help change the behavior of entire communities so as to have maximum impact. Bandura (2001) called this “collective agency” where knowledge, self-efficacy, intentions, and skills interact as “group-level property” and are not simply a sum of the individuals within it (see also Chapters 3, 18, and 28, this volume). However, interventions targeting collective agency where individuals within communities have been engaged in social coordination and interdependent efforts are rare and further research is required. Hawe (2015) argues that public health interventions targeting whole-of-community health behaviors (e.g., the Stanford Heart Disease Prevention Project) erroneously applied individual behavior change theories to a community. Therefore, to change the behavior of individuals and of communities, ecological models need to operate within complex systems.

247

It is important that the application of ecological models is undertaken in recognition of the “delicacy of ecological systems” (Wold & Mittelmark, 2018, p. 25). Without such consideration, there is the possibility that a behavior change program could have unintended consequences on a different part of the system. For example, Johnson (2008) argues that an ecological systems approach is needed within complex systems, such as schools. To better understand children’s educational outcomes, she suggests that future research is needed to inform educational policy by “clarifying the multiple layers within the complex educational system using an ecological systems approach and drawing upon the concepts of complexity” (p. 1). Methodological research is also required that enables assessment of the development of community capacity and the development of collaborative relationships between program deliverers and the community (Richard, 2011). Further research is needed to better understand the use of ecological models in real-world contexts and implementation processes (May et al., 2016). Finally, for population reach and maximizing public health impact, it is important to facilitate a better understanding of how ecological models can assist with scaling up interventions and help with understanding context and the many layers of influence on successful outcomes, as well as their likely complexity (Rutter et al., 2017; Rutter et al., 2019).

References Annesi, J. J., Smith, A. E., & Tennant, G. (2013). Cognitive-behavioural physical activity treatment in African-American pre-schoolers: Effects of age, sex, and BMI. Journal of Paediatrics and Child Health, 49, e128–e132. https://doi.org/ 10.1111/jpc.12082 Arundell, L., Fletcher, E., Salmon, J., Veitch, J., & Hinkley, T. (2016). The correlates of after-school sedentary behavior among children aged 5–18 years:

248

JO SALMON ET AL.

A systematic review. BMC Public Health, 16, 1–10. https://doi.org/10.1186/s12889-015-2659-4 Bandura, A. (1977). Social Learning Theory. Englewood Cliffs, NJ: Prentice Hall. Bandura, A. (1986). Social Foundations of Thought and Action: A Social Cognitive Theory. Englewood Cliffs, NJ: Prentice Hall. Bandura, A. (2001). Social cognitive theory: An agentic perspective. Annual Review of Psychology, 52, 1–26. https://doi.org/10.1146/ annurev.psych.52.1.1 Barker, R. G. (1968). Ecological Psychology. Stanford: Stanford University Press. Bartholomew, L. K., Parcel, G. S., & Kok, G. (1998). Intervention mapping: A process for developing theory- and evidence-based health education programs. Health Education and Behavior, 25, 545– 563. https://doi.org/10.1177/109019819802500502 Bauman, A. E., Reis, R. S., Sallis, J. F. et al. (2012). Correlates of physical activity: Why are some people physically active and others not? The Lancet, 380(9838), 258–271. https://doi.org/ 10.1016/S0140-6736(12)60735-1 Bronfenbrenner, U. (1977). Toward an experimental ecology of human development. American Psychologist, 32, 513–531. https://doi.org/ 10.1037/0003-066X.32.7.513 Cappella, E., Frazier, S. L., Atkins, M. S., Schoenwald, S. K., & Glisson, C. (2008). Enhancing schools’ capacity to support children in poverty: An ecological model of school-based mental health services. Administration and Policy in Mental Health and Mental Health Services Research, 35, 395. https://doi.org/10.1007/s10488-008-0182-y Carson, V., Salmon, J., Arundell, L. et al. (2013). Examination of mid-intervention mediating effects on objectively assessed sedentary time among children in the Transform-Us! clusterrandomized controlled trial. International Journal of Behavioral Nutrition and Physical Activity, 10, 62. https://doi.org/10.1186/14795868-10-62 Cengelli, S., Loughlin, J., Lauzon, B., & Cornuz, J. (2012). A systematic review of longitudinal population-based studies on the predictors of smoking cessation in adolescent and young adult smokers. Tobacco Control, 21, 355–362. https:// doi.org/10.1136/tc.2011.044149

Committee on Capitalizing on Social Science and Behavioral Research to Improve the Public’s Health (2001). Promoting health: Intervention strategies from social and behavioral research. American Journal of Health Promotion, 15, 149– 166. https://doi.org/10.4278/0890-1171-15.3.149 Corcoran, J. (1999). Ecological factors associated with adolescent pregnancy: A review of the literature. Adolescence, 34, 603–619. Cushing, C. C., Brannon, E. E., Suorsa, K. I., & Wilson, D. K. (2014). Systematic review and metaanalysis of health promotion interventions for children and adolescents using an ecological framework. Journal of Pediatric Psychology, 39, 949–962. https://doi.org/10.1093/jpepsy/jsu042 Davids, K., Araujo, D., & Brymer, E. (2016). Designing affordances for health-enhancing physical activity and exercise in sedentary individuals. Sports Medicine, 46, 933–938. https://doi.org/10.1007/s40279-016-0511-3 Davidson, P., Rushton, C. H., Kurtz, M. et al. (2018). A social–ecological framework: A model for addressing ethical practice in nursing. Journal of Clinical Nursing, 27, e1233–e1241. https://doi .org/10.1111/jocn.14158 de Vet, E., De Ridder, D. T. D., & de Wit, J. B. F. (2011). Environmental correlates of physical activity and dietary behaviours among young people: A systematic review of reviews. Obesity Reviews, 12, e130–e142. https://doi.org/10.1111/j.1467789X.2010.00784.x Fitzgibbon, M. L., Stolley, M. R., Schiffer, L. A. et al. (2011). Hip-hop to health jr. obesity prevention effectiveness trial: Postintervention results. Obesity, 19, 994–1003. https://doi.org/10.1038/oby.2010.314 Gibson, J. J. (1979). The Ecological Approach to Visual Perception. Hillsdale, NJ: Lawrence Erlbaum Associates. Golden, S. D., & Earp, J. A. (2012). Social ecological approaches to individuals and their contexts: Twenty years of Health Education and Behavior health promotion interventions. Health Education and Behavior, 39, 364–372. https://doi.org/ 10.1177/1090198111418634 Hawe, P. (2015). Lessons from complex interventions to improve health. Annual Review of Public Health, 36, 307–323. https://doi.org/10.1146/ annurev-publhealth-031912-114421

Changing Behavior Using Ecological Models

Healy, G. N., Eakin, E. G., Owen, N. et al. (2016). A cluster randomized controlled trial to reduce office workers’ sitting time: Effect on activity outcomes. Medicine and Science in Sports and Exercise, 48, 1787–1797. https://doi.org/10.1249/ MSS.0000000000000972 Hinkley, T., Salmon, J., Okely, A. D., & Trost, S. G. (2010). Correlates of sedentary behaviours in preschool children: A review. International Journal of Behavioral Nutrition & Physical Activity, 7, 66. https://doi.org/10.1186/14795868-7-66 Holden, D. J., Holliday, J. L., & Moore, K. S. (1998). Health education for a breast and cervical cancer screening program: Using the ecological model to assess local initiatives. Health Education Research, 13, 293–299. https://doi.org/10.1093/ her/13.2.293 Johnson, E. S. (2008). Ecological systems and complexity theory: Toward an alternative model of accountability in education. Complicity: An International Journal of Complexity and Education 5, 1–10. https://doi.org/10.29173/ cmplct8777 Jones, R. A., Downing, K., Rinehart, N. J. et al. (2017). Physical activity, sedentary behavior and their correlates in children with Autism Spectrum Disorder: A systematic review. PLoS ONE, 12, 1– 23. https://doi.org/10.1371/journal.pone.0172482 Kellou, N., Sandalinas, F., Copin, N., & Simon, C. (2014). Prevention of unhealthy weight in children by promoting physical activity using a socioecological approach: What can we learn from intervention studies? Diabetes and Metabolism, 40, 258–271. https://doi.org/10.1016/j.diabet .2014.01.002 King, J. L., Merten, J. W., Wong, T. J., & Pomeranz, J. L. (2018). Applying a social-ecological framework to factors related to nicotine replacement therapy for adolescent smoking cessation. American Journal of Health Promotion, 32, 1291–1303. https://doi.org/10.1177/ 0890117117718422 Koorts, H., Eakin, E., Estabrooks, P., Timperio, A., Salmon, J., & Bauman, A. (2018). Implementation and scale up of population physical activity interventions for clinical and community settings:

249

The PRACTIS guide. International Journal of Behavioral Nutrition and Physical Activity, 15, 51. https://doi.org/10.1186/s12966-018-0678-0 Lewin, K. (1936). Principles of Topological Psychology. London: McGraw Hill. Lewin, K. (1951). Field Theory in Social Science. New York: Harper. May, C. R., Johnson, M., & Finch, T. (2016). Implementation, context and complexity. Implementation Science, 11, 141. https://doi.org/ 10.1186/s13012-016-0506-3 McLeroy, K. R., Bibeau, D., Steckler, A., & Glanz, K. (1988). An ecological perspective on health promotion programs. Health Education Quarterly, 15, 351–377. https://doi.org/10.1177/ 109019818801500401 O’Donoghue, G., Perchoux, C., Mensah, K. et al. (2016). A systematic review of correlates of sedentary behaviour in adults aged 18–65 years: A socio-ecological approach. BMC Public Health, 16, 1–25. https://doi.org/10.1186/s12889-0162841-3 Ortega-Baron, J., Buelga, S., Ayllon, E., MartinezFerrer, B., & Cava, M. J. (2019). Effects of intervention program Prev@cib on traditional bullying and cyberbullying. International Journal of Environmental Research and Public Health, 16, 257. https://doi.org/10.3390/ijerph 16040527 Richard, L., & Gauvin, L. (2018). Building and implementing ecological health promotion interventions. In I. Rootman, S. Dupéré, A. Pederson, & M. O’Neill (Eds.), Health Promotion in Canada: Critical Perspectives on Practice (3rd ed., pp. 67–80). Toronto: Canadian Scholars Press. Richard, L., Gauvin, L., Potvin, L., Denis, J. L., & Kishchuk, N. (2002). Making youth tobacco control programs more ecological: Organizational and professional profiles. American Journal of Health Promotion, 16, 267–279. https://doi.org/ 10.4278/0890-1171-16.5.267 Richard, L., Gauvin, L., & Raine, K. (2011). Ecological models revisited: Their uses and evolution in health promotion over two decades. Annual Review of Public Health, 32, 307–326. https://doi .org/10.1146/annurev-publhealth-031210-101141

250

JO SALMON ET AL.

Rutter, H., Cavill, N., Bauman, A., & Bull, F. (2019). Systems approaches to global and national physical activity plans. Bulletin of the World Health Organization, 97, 162–165. https://doi .org/10.2471/BLT.18.220533 Rutter, H., Savona, N., Glonti, K. et al. (2017). The need for a complex systems model of evidence for public health. The Lancet, 390(10112), 2602– 2604. https://doi.org/10.1016/S0140-6736(17) 31267-9 Sallis, J. F., & Owen, N. (2015). Ecological models of health behavior. In K. Glanz, B. K. Rimer, & K. Viswanath (Eds.), Health Behavior: Theory, Research and Practice (5th ed., pp. 43–64). San Francisco, CA: Jossey-Bass. Sallis, J. F., Owen, N., & Fotheringham, M. J. (2000). Behavioral epidemiology: A systematic framework to classify phases of research on health promotion and disease prevention. Annals of Behavioral Medicine, 22, 294–298. https://doi .org/10.1007/BF02895665 Salmon, J., Arundell, L., Hume, C. et al. (2011). A cluster-randomized controlled trial to reduce sedentary behavior and promote physical activity and health of 8–9 year olds: The Transform-Us! Study. BMC Public Health, 11, 759. https://doi .org/10.1186/1471-2458-11-759 Salmon, J., Tremblay, M. S., Marshall, S. J., & Hume, C. (2011). Health risks, correlates, and interventions to reduce sedentary behavior in young people. American Journal of Preventive Medicine, 41, 197–206. https://doi.org/10.1016/j .amepre.2011.05.001 Stokols, D. (1992). Establishing and maintaining healthy environments: Toward a social ecology of health promotion. American Psychologist, 47, 6–22. https://doi.org/10.1037/0003066X.47.1.6 Thornton, L. E., Cameron, A. J., McNaughton, S. A., Worsley, A., & Crawford, D. A. (2012). The availability of snack food displays that may trigger impulse purchases in Melbourne supermarkets. BMC Public Health, 12, 194. https://doi.org/ 10.1186/1471-2458-12-194

Trost, S. G., Owen, N., Bauman, A. E., Sallis, J. F., & Brown, W. (2002). Correlates of adults’ participation in physical activity: Review and update. Medicine and Science in Sports and Exercise, 34, 1996–2001. https://doi.org/10.1249/ 01.MSS.0000038974.76900.92 Wang, M. L., Lemon, S. C., Clausen, K., Whyte, J., & Rosal, M. C. (2016). Design and methods for a community-based intervention to reduce sugar-sweetened beverage consumption among youth: H2GO! study. BMC Public Health, 16, 1150. https://doi.org/10.1186/s12889-0163803-5 Welk, G. J. (1999). The youth physical activity promotion model: A conceptual bridge between theory and practice. Quest, 51, 5–23. https://doi .org/10.1080/00336297.1999.10484297 WHO (World Health Organization). (2004). A Guide for Population-Based Approaches to Increasing Levels of Physical Activity: Implementation of the WHO Global Strategy on Diet, Physical Activity and Health. Geneva: WHO. WHO (World Health Organization). (2009). Interventions on Diet and Physical Activity: What Works. Summary Report. Geneva: WHO. Winter, S. M., & Sass, D. A. (2011). Healthy and ready to learn: Examining the efficacy of an early approach to obesity prevention and school readiness. Journal of Research in Childhood Education, 25, 304–325. https://doi.org/10.1080/ 02568543.2011.580211 Wold, B., & Mittelmark, M. B. (2018). Healthpromotion research over three decades: The social-ecological model and challenges in implementation of interventions. Scandinavian Journal of Public Health, 46 (Suppl. 20), 20–26. https://doi.org/10.1177/ 1403494817743893 Yildirim, M., Arundell, L., Cerin, E. et al. (2014). What helps children to move more at school recess and lunchtime? Mid-intervention results from Transform-Us! cluster-randomized controlled trial. British Journal of Sports Medicine, 48, 271–277.

18 Changing Behavior Using Theories at the Interpersonal, Organizational, Community, and Societal Levels Robert A. C. Ruiter, Rik Crutzen, Evelyne De Leeuw, and Gerjo Kok

Practical Summary Various factors in an individual’s environment, as well as “social agents” within this environment, can influence an individual’s behavior. Examples include family and peers, professionals, organizations, community members, opinion leaders, and policies. In line with this, planned behavior change programs target not only individuals but also the social agents who exert an influence on these environments. Examples of these kinds of behavior change programs include the formation of patient support groups, the provision of strategies to cope with work stress, the creation of safer bicycle lanes, and advocacy for better child health insurance. Researchers targeting people to behave in a healthier or more energy-efficient way can draw on theories about environmental influences to design behavioral interventions. In this chapter, these theories and their use at the interpersonal, organizational, community, and societal levels are described within the framework of a systematic theory- and evidence-based approach to behavior change.

18.1 Introduction: A ProblemDriven and Socioecological Approach Societal problems, such as the burden of chronic health behavior–related diseases or inefficient resource usage, can be alleviated by changing people’s behaviors. Behavioral scientists and behavior change professionals apply theories that attempt to understand human behavior and how to change it. Addressing behavior change usually requires the coordinated pooling of expertise of individuals from multiple backgrounds and disciplines and at multiple ecological levels. Applied psychology is a primary example of a multidisciplinary field that has informed efforts to solve behavior-driven problems. Although the required expertise within

multidisciplinary planning groups may vary according to the behavioral problem to be addressed, expertise in behavior change is always required. It is worth noting that, within applied psychology, a distinction can be made between two approaches: theory-driven and problem-driven applied psychology (Kok et al., 1996; Ruiter et al., 2013). Theory-driven applied psychology involves testing a theory in an applied setting, for example in schools or organizations, primarily in order to gain insight into the external validity of the theory. Problem-driven applied psychology refers to scientific activities that focus on changing or reducing a practical problem. In problem-driven applied psychology, theories are used but problem-solving is https://doi.org/10.1017/9781108677318.018

252

ROBERT A. C. RUITER, RIK CRUTZEN, EVELYNE DE LEEUW, AND GERJO KOK

the primary focus of this approach, and the criteria for success are formulated in terms of problem reduction. Problem-driven applied psychology thus functions as the ultimate test for the usefulness of psychology, both as a discipline and as a profession. As there are limits to the time and resources available for problem-specific research, it is essential to apply available evidence-based theories related to the problem in the correct manner. Various “social agents” in an individual’s environment, such as family members and peers, professionals in organizations, opinion leaders, and policy makers, can have a pervasive influence when it comes to behavior change. These agents are present in a wide variety of contexts and function at different social ecological levels, including the interindividual, organizational, community, and societal levels. Importantly, the behavior and decisions of these agents are also relevant, and intervention programs can therefore be applied to change the behavior of agents operating at these higher ecological levels in much the same way as they are at the individual level (see Chapter 17, this volume). The socioecological approach to behavior change focuses on the interrelationships between individuals and their interpersonal, organizational, community, and policy environments (Kok, Peters, & Ruiter, 2017; see Chapter 17, this volume). Each of these environmental levels comprise physical, social, and cultural dimensions. Two key assumptions from the socioecological approach can guide the identification of intervention targets for promoting behavior change: (1) behavior influences and is influenced by multilevel environmental factors and (2) individual behavior both shapes and is shaped by the environment (see also Chapter 3, this volume). These assumptions imply that interventions at the various ecological levels should focus on changing the behavior of agents who are in a position to exercise influence over aspects of the environment and, in doing so, bring about behavior change at other environmental levels. For

Figure 18.1 Socioecological approach indicating

different levels of environmental influence on behavior

example, adolescent uptake of smoking might be influenced by peers and parents at the interpersonal environment level and by regulations and access at the community and societal environment levels. The picture that emerges is of a clear – but complex – system of influences on individual behavior change that occur between and within environmental levels. It also presents a rich set of potential contexts in which behavior change interventions can be applied (see Figure 18.1). Smedley and Syme (2000) have shown that intervention packages that work in synergy across environmental levels work better than interventions that operate in isolation at only one level. Programs targeting behavior change – for example, engaging in health-enhancing behaviors, performing ecological behaviors, or saving money – should thus not only target the individual but also consider the environmental factors that influence people’s behavior. In addition to individual-level theories, such as social cognitive models of human behavior (for examples, see Chapters 2, 3, and 4, this volume), environmental-level theories can also inform the design of behavior change

Changing Behavior Using Theories at the Interpersonal, Organizational, Community, and Societal Levels

interventions (see Chapter 17, this volume). However, these higher-level theories have received less attention in the literature on behavior change, particularly the psychological literature. Examples of general environmental-level theories are systems theory (National Cancer Institute, 2007; Chapter 9, this volume), theories of power (Turner, 2005), and empowerment theories (Wallerstein et al., 2015). It is also worth noting that different theories may be useful in terms of informing behavior change at different ecological levels. For example, at the interindividual level, social network (Valente, 2015) and social support theories (Holt-Lunstad & Uchino, 2015), theories of stigma and discrimination (Bos et al., 2013), and diffusion of innovations theory (Brownson et al., 2015) may be useful. At the organizational level, theories of organizational change and organizational development (Cummings & Worley, 2014), and stakeholder frameworks (Kok et al., 2015) may be relevant. At the community level, coalition frameworks (Butterfoss & Kegler, 2012), social capital and community capacity theories (Szreter & Woolcock, 2004), social norms theories (Smelser, 2011), and community organization theories (De Leeuw & Simos, 2017) may come into play. Finally, at the societal level, theories of policy process (Sabatier & Weible, 2014) and multiplex policy network analysis (De Leeuw, Brown, & Gleeson, 2018) may inform behavior change programs. This chapter provides an overview of how to target behavior change at these different ecological levels and how community theories can inform this approach. This is followed by a summary of commonly used theories to promote behavior change at the different environmental levels.

18.2 Theories at the Environmental Level The major environmental-level theories that have been applied to behavior change interventions are summarized in Table 18.1. These theories are

253

organized by level: general environmental, interpersonal, organizational, community, and societal. For each theory, key references are provided, key concepts are presented, an example is given, and the methods for change derived from the theory are listed. For example, one type of theory that can be applied at the organizational level is stakeholder frameworks (Kok et al., 2015; see also Chapter 24, this volume). A stakeholder is any individual, group, or entity that can influence an organization. The more salient a stakeholder is to the organization, and the more central the stakeholder is in the network, the stronger the influence. Salience of a stakeholder, defined as the priority with which decision makers consider the stakeholder’s claim, is determined by the stakeholder’s power, legitimacy, and urgency. The example provided in Sidebar 18.2 illustrates how a multinational firm’s strategic plans were denied by a coalition of stakeholders: a small and seemingly powerless community in Louisiana, supported by media and public sympathy, receiving technical, legal, and other assistance, and applying sophisticated legal and political maneuvering (Berry, 2003). Notice that methods for change should be applied correctly, which means that the parameters for use should be taken into account. Otherwise, the methods might not work and may even be counterproductive, in terms of contributing to behavior change (Bartholomew Eldredge et al., 2016; Kok et al., 2016). In the example provided, a key parameter of use for “increasing stakeholder influence” is that the focal organization, the multinational company, perceives that the external organization or group is one of its stakeholders (in this case, public sympathy for the community’s protest). Appendix 18.1 (supplemental materials) includes a list of environmental-level change methods with definitions, parameters, and examples (derived from Bartholomew Eldredge et al., 2016; see also Chapter 14, this volume).

254

ROBERT A. C. RUITER, RIK CRUTZEN, EVELYNE DE LEEUW, AND GERJO KOK

Table 18.1 Summary of major environmental-level theories including key readings and examples and methods for change Level and Theory

Key Readings/Examples

Method for Change

General environmental level Systems theory (National Cancer Institute, 2007) Theories of power (Turner, 2005)

Empowerment theories (Wallerstein et al., 2015)

Systems change (Best et al., 2012; Anderson et al. (2013); Van Beurden National Cancer Institute, 2007) et al. (2013); Anti-bullying programs (Juvonen & Graham, 2014). Forms of power (Weber, 1947); Types Coercion (Freudenberg & Tsui, 2014; Turner, 2005); Community of power (Wallerstein et al., 2015); development and social action Leadership intervention (Slater & (Minkler & Wallerstein, 2012); Barker, 2019) Agenda setting and creating and enforcing laws and regulations (Clavier & de Leeuw, 2013) Participation (McCullum et al., Empowerment (Aiyer et al., 2014); 2004); Enactive mastery Collective efficacy (Bandura, 1997); experiences (Kelder et al., 2015) Community efficacy example (Commers et al., 2007). Interpersonal level

Social cognitive theory (Kelder et al., 2015; Chapter 3, this volume) Social network and social support theories (Holt-Lunstad & Uchino, 2015; Valente, 2015)

Theories of stigma and discrimination (Bos et al., 2013) Diffusion of innovations theory (Brownson et al., 2015)

Active learning, reinforcement, Social cognitive theory (Bandura, enactive mastery experiences, 1986); Drug intervention (Golden & modeling, facilitation (Bandura, Earp, 2012) 1997; Kelder et al., 2015) Legislation to support social networks Enhancing network linkages; (McLeroy et al., 2001) Developing new network linkages (Holt-Lunstad & Uchino, 2015; Valente, 2015); Use of natural helpers, peers, or lay health workers (Tolli, 2012) Interpersonal contact; Cooperative Interpersonal contact intervention learning (Corrigan & Kosyluk, reduces stigmatization (Corrigan & 2013); Reducing power Kosyluk, 2013) differences (Link & Phelan, 2001) Modeling (Kelder et al., 2015); Diffusion (Wiecha et al., 2004); Sex Technical assistance (Flaspohler education program (Schutte et al., et al., 2008); Mass media role mod2014) eling (Rogers, 2003); Entertainment education (Shen & Han, 2014) Organizational level

Organizational change theories (Cummings & Worley, 2014)

Sense-making (Weick & Quinn, Force field analysis (Lewin, 1947); 1999); Advocacy and lobbying Continuous change model (Brown et (Cummings & Worley, 2014) al., 2014); Sense-making intervention (Yip, Lee, & Tsui, 2015) Continued

Changing Behavior Using Theories at the Interpersonal, Organizational, Community, and Societal Levels

255

Table 18.1 (cont.) Level and Theory

Key Readings/Examples

Method for Change

Organizational level Organizational development theories (Cummings & Worley, 2014)

Stakeholder frameworks (Kok et al., 2015)

Sense-making (Weick & Quinn, Planned change framework (Schein, 1999); Participatory problem2010); Stage theory of solving, organizational diagnosis organizational change (Butterfoss et and feedback, team building and al., 2008); Evidence-based system human relations training, for innovation support (EBSIS) structural redesign (Cummings & intervention (Wandersman et al., Worley, 2014) 2012) Stakeholder influence intervention case Increasing stakeholder influence, study (Berry, 2003) social action (Kok et al., 2015) Community level

Coalition frameworks (Butterfoss & Kegler, 2012)

Community coalitions (Butterfoss & Kegler, 2009); Coalition progress example (Kegler & Swan, 2011)

Social capital and community Community capacity building (Wendel et al., 2009); Linking individuals to capacity (Szreter & social networks and community Woolcock, 2004) organizations (Steckler et al., 2002)

Social norms theories (Smelser, 1998)

Conscientization (Freire, 1968, 1974; Wiggins, 2012)

Social norms, social rules, and socialization (Smelser, 1998); Behavioral journalism example (Van Empelen et al., 2003)

Participatory problem-solving (Cummings & Worley, 2014); Technical assistance (Flaspohler et al., 2008); Forming coalitions (Clavier & de Leeuw, 2013) Community development, participatory problem-solving (Wallerstein et al., 2015); Technical assistance (Flaspohler et al., 2008); Forming coalitions (Clavier & de Leeuw, 2013) Entertainment education (Bouman, 2017); Behavioral journalism (Reininger, 2010); Mobilizing social networks (Valente, 2015); Mass media role modeling (Rogers, 2003) Problem-posing education (Wallerstein et al., 2004)

Critical consciousness (Gadotti, 1994); Reflection-actionreflection (Wallerstein et al., 2015); Consciousness-raising associations example (Minkler, 2004) Participatory problem-solving Community organization (De Community-building vs. (Wallerstein et al., 2015); Forming Leeuw & Simos, 2017) empowerment-oriented social action coalitions (Clavier & De Leeuw, (Minkler & Wallerstein, 2012); 2013); Peer education (Tolli, Advocacy (Galer-Unti et al., 2004; 2012); Media advocacy (Dorfman Wallack et al., 2008); Community & Krasnow, 2014); Framing to influence example (El-Askari & shift perspectives (Snow et al., Walton, 2008) 2004); Social planning, community assessment (Rothman, 2008) Continued

256

ROBERT A. C. RUITER, RIK CRUTZEN, EVELYNE DE LEEUW, AND GERJO KOK

Table 18.1 (cont.) Level and Theory

Key Readings/Examples

Method for Change

Societal level Agenda setting, creating and enforMultiple streams (Kingdon, 2003); Theories of the policy procing laws and regulations (Clavier Advocacy coalition framework cess (Clavier & de Leeuw, & de Leeuw, 2013); Timing to (Sabatier 1988); Punctuated equili2013; Sabatier & Weible, coincide with policy windows brium theory (Baumgartner & Jones, 2014) (Zahariadis, 2007); Stakeholder 2010); Policy advocacy examples politics (Breton et al., 2008) (Breton et al., 2008; Kübler, 2001) Policy network theory (Börzel, 1998); Rhetorical and narrative devices Multiplex policy network applied to policy networks Policy frame analysis (Rein & analysis (De Leeuw et al., (Davidson, 2017; De Leeuw, Schön, 1996); Effects of different 2018) 2018) policy frames (Oliver & Cairney, 2019)

18.3 Planned Behavior Change at Higher Ecological Levels In intervention planning, different frameworks, such as intervention mapping (Bartholomew Eldredge et al., 2016), can be used to provide guidance to planners from the initial stages of problem definition all the way through to problem solution (Chapter 21, this volume). Planners need to address questions relevant to understanding problematic or risky behavior and design effective solutions using expert knowledge, theory, and evidence. The processes involved in answering such questions can be complex and timeconsuming; sometimes intervention planners do not persevere in working through these difficulties. Consequently, the understanding of a problem is often incomplete and attempts to solve the problem may be based on an incorrect or inappropriate conceptualization of it (known as a “type III error”; De Leeuw, 2009). Every decision-making process in the development and planning of behavioral interventions should start with posing the correct or appropriate question. This should then be followed by a brainstorming process to identify possible solutions, an assessment of – and access to – existing knowledge and

expertise within the project team, a review of the empirical literature in order to learn from previous behavior change interventions or initiatives (being applied within the same or similar context), and a problem-driven identification of appropriate theories that can be used to support the general behavior change processes that may operate in this context. This decision-making process should then be concluded by identifying and addressing the need for new research and ultimately by providing a list of answers for which the theoretical and empirical evidence has been evaluated to be sufficient. Using these “core processes” from the intervention mapping approach (Ruiter & Crutzen, 2019) increases the likelihood of obtaining a more complete understanding of the issue and of selecting effective solutions.

18.4 The Need for a Multilevel and Systematic Approach It is important to recognize that the behavior of individuals as well as agents in the environment (e.g., family members and peers, organizational managers, opinion leaders, and policy makers) is part of a “system.” Understanding and changing behavior requires an understanding of this system

Changing Behavior Using Theories at the Interpersonal, Organizational, Community, and Societal Levels

and the development of appropriate strategies to change it. Reviews of theories of behavior change (e.g., Conner & Norman, 2016) indicate that individual-level theories tend to be prominently represented in the development of approaches to behavior change, while environment-level theories are less well represented (see also Chapter 1, this volume). Theories about stigma, power, policy processes, and political contexts seem to have been deemed extraneous to the disciplinary expertise on behavior change. Yet these frameworks are crucial for the development of a socioecological approach to planning effective strategies that can promote behavior change. Changing behavior requires an understanding of the causes and effects of that behavior (Bartholomew Eldredge et al., 2016; Kok et al., 2017; see Chapters 19 and 21, this volume). Consistent with the logic model presented in Figure 18.2, outcomes are determined by changes in the individual behavior of the “atrisk” group and by changes in the behaviors of the “agents.” Environmental agents are the people within the environment at various levels (organizational, community, and policy) that, together, determine the environmental conditions that affect individual behavior. Once the agents and their behaviors have been identified, determinants and methods to change them can be selected (see Chapters 19 and 20, this volume). Translating methods into practical applications demands an understanding of the theory behind the method, especially the theoretical parameters that determine whether the method is effective or not (Bartholomew Eldredge et al., 2016; Kok et al., 2016; Peters, de Bruin, & Crutzen, 2015). These parameters are defined as the conditions that must be satisfied in practical applications for the method to be effective. Change methods at the individual level can be directed toward agents at higher levels (e.g., consciousness-raising; Chapter 45, this volume), in combination with change methods appropriate for higher levels (e.g., agenda setting; Chapter 24,

257

this volume). This follows the logic that changes in the environmental conditions that precede behavior change necessitate the application of theories at both the individual and the environmental levels in order to be able to understand and change behavior. Sidebar 18.1 provides an example of an intervention promoting the fuel-saving behavior of individual van drivers delivering the mail and its effects on improving energy conservation and quality of life. In this example, the relevant agent in the environmental condition is the mail organization, that is, the manager. Determinants of the individual driver’s driving behavior (e.g., attitude, social norms, and perceived skills; see Chapters 2 and 3, this volume) were influenced by behavior change methods such as education, facilitation, and control exerted by the manager, whereas task assignment, reinforcement, and feedback were introduced to strengthen organizational control over the driver’s behavior.

18.5 Looking at Environments As Outcomes Figure 18.2 outlines environmental conditions as they relate to outcomes – either directly or as a result of changing the actual behavior of the target group. In the health domain, for example, researchers and practitioners have also looked at creating potentially healthier environments as a desired outcome, irrespective of individual health behavior change or specific health outcomes. For example, following the 1986 Ottawa Charter for Health Promotion, the World Health Organization (WHO) has focused on the creation of healthy settings where people can “learn, work, play, and love” (WHO, 1986, p. 4). Diverse examples of healthy settings include nations, islands, cities, communities, workplaces, hospitals, prisons, schools, airports, and universities. In order to create healthy settings, researchers must explicitly address health values such as equity,

258

ROBERT A. C. RUITER, RIK CRUTZEN, EVELYNE DE LEEUW, AND GERJO KOK

Sidebar 18.1 Case study: Planned change at the organizational level – promoting a fuel-saving driving style among van drivers delivering mail In this study, Siero and colleagues (1989) promoted a fuel-saving driving style among van drivers of a mail company, applying both individual-level theories and theories at the organizational level, with managers as key environmental agents. Important energy-wasting behaviors among van drivers were identified and transformed into performance objectives (i.e., sub-behaviors that lead to fuelsaving). Next, determinants of the performance objectives were examined. The intervention consisted of several practical applications of methods derived from individual and environmental theories (see Figure 18.2, upper and lower rows, respectively). At the individual level, methods applied included information, model demonstrations, and feedback. At the organizational level, methods applied included facilitation (based on social cognitive theory; Kelder, Hoelscher, & Perry, 2015; Chapter 3, this volume), task assignment (based on organizational development theories; Cummings & Worley, 2014), and control (based on an application of the “coercion” method on power differences and theories of power; Turner, 2005). These methods and their accompanying approaches are outlined in Table 18.1. Fuel-saving was presented as a task for the drivers that would be systematically monitored by their local supervisor. The general director asked the local management to oversee the local energy-savings program and try to achieve the energy-savings goal. In a field experiment, the experimental group was found to have achieved an 8 percent reduction in fuel use as compared to the control group. There was a striking relationship between the weekly feedback on gasoline consumption and the control-related behavior of the management. In almost all offices, intervention by the management (in the form of instructions and pep talks) resulted in decreased fuel consumption. Therefore, changes in the behavior of management played an important role in changing the driving behavior of the van drivers.

diversity, and sustainability and adopt programmatic individual and institutional behavior change approaches to deal with proximal and distal determinants of health (Farrington, Faskunger, & Mackiewicz, 2015; Poland, Krupa, & McCall, 2009). Sidebar 18.2 provides an example of how community action can successfully influence policy change. Distal determinants of behavior change may also be found at the global political level. The need to reduce emissions to address climate

change through, for example, the implementation of recommendations from the Kyoto Protocol and Paris Agreement sets a behavioral agenda at the institutional level. In Australia, community groups recently won a court case about a proposed new coal mine, with the result that it was not allowed to proceed (Hannam, 2019) – echoing a Dutch court case (Verschuuren, 2019). Such contexts are important for individual and population health and facilitate the modification of, for example, sustainability behaviors.

Changing Behavior Using Theories at the Interpersonal, Organizational, Community, and Societal Levels

Behavior change methods

Determinants

Behavior of at-risk group

Environmental condition

Behavior change methods

Determinants

259

Desired outcomes

Behaviors of environmental agents

Figure 18.2 Logic model for relationships among behavior change methods, behavioral

determinants, behaviors, environmental conditions, and desired outcomes such as health, safety, and quality of life (Bartholomew Eldredge et al., 2016)

Sidebar 18.2 Case study: Planned change at the societal level – Greenpeace activists recruit powerful allies At 2 p.m. EST, on April 17, 2001, Greenpeace activists with Halloween masks storm Trader Joe’s stores in nine [US] states, seize cornbread mix from the shelves, and wheel it out in shopping carts. Like a scene in Puritan New England with the public stocks on the village commons, the cornbread boxes are piled in the parking lot of this upscale grocery store chain and slung with warnings for management to heed: We have removed these products from your shelves because they contain unsafe genetically engineered ingredients … now it’s your turn. Management, stunned and angry, lashes back, calling the behavior of the activists “hysteria”. They make it clear that Greenpeace has not seen the last of Trader Joe’s cornbread. Just 6 months later, though, a consumer boycott engineered by Greenpeace is having an effect. For weeks, store managers have been pulling cornbread mix – and other products targeted by environmentalists – from shelves and tossing it into dumpsters behind the stores as it runs past its expiration date. Management calls a press conference and announces that it has agreed to force its suppliers to stop using genetically engineered (GE) ingredients. (Frooman & Murrell, 2005, pp. 3–4)

Stakeholder theory may help behavior change professionals make changes at the organizational and societal levels in order to achieve their goals and improve people’s lives. A stakeholder is any individual, group, or entity that can influence an organization. Behavior change professionals may use direct and indirect methods to increase their influence: compromise, credibility building, coercion, coalition building, communication, and deinstitutionalization (Kok et al., 2015). In the above example, Greenpeace applied coercion, which can only be used effectively when the target organization is dependent on the entity applying coercion (or its allies) for resources – in this case, their consumers.

18.6 Conclusion Taking a socioecological approach to behavior change means taking into account the contexts within which people live, work, and play – and

making these the object of inquiry and intervention. The needs and capacities of people in different settings should also be taken into account. The use of this approach can increase the likelihood of

260

ROBERT A. C. RUITER, RIK CRUTZEN, EVELYNE DE LEEUW, AND GERJO KOK

success because it offers opportunities to situate practice in its context. Members of the setting can optimize interventions for specific contextual contingencies, target crucial factors in the organizational context that influence behavior, and render settings more conducive to target behavior. It is important to note that most behavior change interventions need to be implemented in real-life settings, for example teachers that implement an intervention consisting of a series of lessons on environmental care or on traffic safety should ensure that these take place in the relevant settings (Chapters 19 and 35, this volume). The implementation of behavior change interventions that have been shown to be potentially effective in changing

behavior is a further example of agency in the environment. The agents in this setting are the implementers of the intervention itself, for example teachers (Bartholomew Eldredge et al., 2016; Kok et al., 2017). The intervention planners therefore also create a higher-level intervention to promote implementation (Nilsen et al., 2013). Consideration of program implementation actually begins in the needs assessment phase and continues throughout the process of program development (see Chapter 23, this volume). Sidebar 18.3 provides an example of a failed attempt to implement a behavior change intervention at the organizational level (i.e., the promotion of an influenza vaccination in health care

Sidebar 18.3 Case study: Planned change at the organizational level – influenza vaccination Influenza vaccination uptake among health care workers is the most effective way of preventing transmission to patients, but vaccination coverage rates are low among these workers. Lehmann et al. (2016) tested an opt-out strategy in promoting uptake among health care workers in two health care institutions. Workers were randomly assigned to one of two conditions. In the opt-out condition, nurses and doctors received an email with a prescheduled appointment for an influenza vaccination, which could be changed or canceled. In the opt-in condition, participants received an e-mail explaining that they had to schedule an appointment if they wanted to get vaccinated. Findings in one institute showed that health care workers in the opt-out condition were more likely to have an appointment for influenza vaccination, which, in turn, increased the probability of actually getting vaccinated by an 11.5 percent absolute difference compared with the opt-in condition. In this example, the organization was relatively small and the intervention was implemented by a very motivated “program champion” (Peterson et al., 2007). However, in the other health care institute, the planned intervention was never implemented. Following their original adoption of the intervention, the management team had to cancel the intervention because of the resistance they met from the employee council when announcing the new vaccination policy. Employee council members felt that the new policy would restrict the decision autonomy of employees to such an extent that control over “their right to be sick” would be put in the hands of the management rather than the employees themselves (note that having a professional work attitude and patient protection were not major arguments in the discussion). Organizational change failed in this particular instance because stakeholder participation – adopting a participatory problem-solving approach with representatives of the employees – was not secured in the early stages of intervention planning. Moreover, the management was not willing to exert control in a setting where there were already enough “issues” between management and personnel.

Changing Behavior Using Theories at the Interpersonal, Organizational, Community, and Societal Levels

workers). This example illustrates how the success or failure of the implementation process ultimately depends on careful planning and the involvement of relevant stakeholders from the start of the planning process. In this chapter, a socioecological approach has been adopted within the context of problem-driven applied psychology (see Chapter 17, this volume) and it has been argued that planned behavior change programs not only should target the individual but should also target environmental influences on behavior. Behaviors and desired outcomes – such as health, sustainability, and diversity – are determined by changes in individual behavior as well as by changes in environmental conditions at different levels of the socioecological model (see Figure 18.1), in this case in the behavior of relevant decision makers or agents. Behavior change programs targeting individuals need to be implemented systematically, which also means that they are accompanied by an environmental change regarding relevant adopters and implementers.

References Aiyer, S. M., Zimmerman, M. A., Morrel-Samuels, S., & Reischl, T. M. (2014). From broken windows to busy streets: A community empowerment perspective. Health Education and Behavior, 42, 137–147. https://doi.org/10.1177/1090198114558590 Anderson, R. A., Plowman, D., Corazzini, K. et al. (2013). Participation in decision making as a property of complex adaptive systems: Developing and testing a measure. Nursing Research and Practice, 706842. https://doi.org/ 10.1155/2013/706842 Bandura, A. (1986). Social Foundations of Thought and Action: A Social Cognitive Theory. Englewood Cliffs, NJ: Prentice-Hall. Bandura, A. (1997). Self-Efficacy: The Exercise of Control. New York: W. H. Freeman. Bartholomew Eldredge, L. K., Markham, C. M., Ruiter, R. A. C., Fernández, M. E., Kok, G., & Parcel, G. S. (2016). Planning Health Promotion Programs:

261

An Intervention Mapping Approach (4th ed.). San Francisco, CA: Jossey-Bass. Baumgartner, F. R., & Jones, B. D. (2010). Agendas and Instability in American Politics. Chicago: University of Chicago Press. Berry, G. R. (2003). Organizing against multinational corporate power in cancer alley. The activist community as primary stakeholder. Organization and Environment, 16, 3–33. https://doi.org/10 .1177/1086026602250213 Best, A. (2011). Systems thinking and health promotion. American Journal of Health Promotion, 25, eix–ex. https://doi.org/10.4278/ajhp.25.4.eix Best, A., Greenhalgh, T., Lewis, S., Saul, J. E., Carroll, S., & Bitz, J. (2012). Large system transformation in health care: A realist review. Milbank Quarterly, 90, 421–456. https://doi.org/10.1111/ j.1468-0009.2012.00670.x Bos, A. E. R., Pryor, J. B., Reeder, G. D., & Stutterheim, S. E. (2013). Stigma: Advances in theory and research. Basic and Applied Social Psychology, 35, 1–9. https://doi.org/10.1080/ 01973533.2012.746147 Börzel, T. A. (1998). Organizing Babylon: On the different conceptions of policy networks. Public Administration, 76, 253–273. https://doi.org/ 10.1111/1467-9299.00100 Bouman, M. (2017). Storytelling makes public health statistics more accessible. European Journal of Public Health, 27. https://doi.org/10.1093/eurpub/ ckx187.524 Breton, E., Richard, L., Gagnon, F., Jacques, M., & Bergeron, P. (2008). Health promotion research and practice require sound policy analysis models: The case of Quebec’s Tobacco Act. Social Science and Medicine, 67, 1679–1689. https://doi.org/ 10.1016/j.socscimed.2008.07.028 Brown, A. D., Colville, I., & Pye, A. (2014). Making sense of sensemaking in organization studies. Organization Studies, 36, 265–277. https://doi .org/10.1177/0170840614559259 Brownson, R., Glanz, K., Tabak, R., & Stamatakis, K. (2015). Implementation, dissemination, and diffusion of public health interventions. In K. Glanz, B. K. Rimer, & K. Viswanath (Eds.), Health Behavior: Theory, Research, and Practice (5th ed., pp. 544–592). San Francisco, CA: Jossey-Bass.

262

ROBERT A. C. RUITER, RIK CRUTZEN, EVELYNE DE LEEUW, AND GERJO KOK

Butterfoss, F. D., Kegler, M. C., & Francisco, V. T. (2008). Mobilizing organizations for health promotion: Theories of organizational change. In K. Glanz, B. K. Rimer, & K. Viswanath (Eds.), Health Behavior and Health Education: Theory, Research, and Practice (4th ed., pp. 335–362). San Francisco, CA: Jossey-Bass. Butterfoss, F. D., & Kegler, M. C. (2009). The community coalition action theory. In R. J. DiClemente, R. A. Crosby, & M. C. Kegler (Eds.), Emerging Theories in Health Promotion Practice and Research (2nd ed., pp. 237–276). San Francisco, CA: Jossey-Bass. Butterfoss, F., & Kegler, M. (2012). A coalition model for community action. In M. Minkler (Ed.), Community Organizing and Community Building for Health and Welfare (3rd ed., pp. 309–328). New Brunswick, NJ: Rutgers University Press. Clavier, C., & De Leeuw, E. (Eds.). (2013). Health Promotion and the Policy Process. Oxford: Oxford University Press. Commers, M. J., Gottlieb, N., & Kok, G. (2006). How to change environmental conditions for health. Health Promotion International, 22, 80–87. https://doi.org/10.1093/heapro/dal038 Conner, M. & Norman, P. (Eds.). (2016). Predicting and Changing Health Behaviour: Research and Practice with Social Cognition Models. Maidenhead: Open University Press. Corrigan, P. W., & Kosyluk, K. A. (2013). Erasing the stigma: Where science meets advocacy. Basic and Applied Social Psychology, 35, 131–140. https:// doi.org/10.1080/01973533.2012.746598 Cummings, T. G., & Worley, C. G. (2014). Organization Development and Change (10th ed.). Mason, OH: South-Western Cengage Learning. Davidson, B. (2017). Storytelling and evidence-based policy: Lessons from the grey literature. Palgrave Communications, 3, p 17093. https://doi.org/ 10.1057/palcomms.2017.93 De Leeuw, E. (2009). Evidence for Healthy Cities: Reflections on practice, method and theory. Health Promotion International, 24, i19–i36. https://doi .org/https://doi.org/10.1093/heapro/dap052 De Leeuw, E. (2018). Policy, theory, and evaluation: Stop mixing the fruit salad: Comment on “Developing a framework for a program theory-based approach to

evaluating policy processes and outcomes: Health in all policies in South Australia.” International Journal of Health Policy and Management, 7, 763. https://doi.org/10.15171/ijhpm.2018.35 De Leeuw, E., Browne, J., & Gleeson, D. (2018). Overlaying structure and frames in policy networks to enable effective boundary spanning. Evidence and Policy: A Journal of Research, Debate and Practice, 14, 537–547. https://doi.org/ 10.1332/174426418X15299595767891 De Leeuw, E., & J. Simos (2017). Healthy cities move to maturity. In E. De Leeuw & J. Simos (Eds.), Healthy Cities: The Theory, Policy, and Practice of ValueBased Urban Planning (pp. 75–86). New York: Springer. Dorfman, L., & Krasnow, I. D. (2014). Public health and media advocacy. Annual Review of Public Health, 35, 293–306. https://doi.org/10.1146/ annurev-publhealth-032013-182503 El-Askari, G., & Walton, S. (2008). Local government and resident collaboration to improve health: A case study in capacity building and cultural humility. In M. Minkler (Ed.), Community Organizing and Community Building for Health (2nd ed., pp. 254– 271). New Brunswick, NJ: Rutgers University Press. Farrington, J. L., Faskunger, J., & Mackiewicz, K. (2015). Evaluation of risk factor reduction in a European City Network. Health Promotion International, 30, i86–i98. https://doi.org/10.1093/ heapro/dav038 Fernandez, M. E., ten Hoor, G. A., van Lieshout, S., Rodriguez, S. A., Beidas, R. S., Parcel, G., Ruiter, R. A. C., Markham, C. Mz., & Kok, G., (2019). Implementation mapping: Using intervention mapping to develop implementation strategies. Frontiers in Public Health, 7, 158. https://doi.org/ 10.3389/fpubh.2019.00158 Flaspohler, P., Duffy, J., Wandersman, A., Stillman, L., & Maras, M. A. (2008). Unpacking prevention capacity: An intersection of research-to-practice models and community-centered models. American Journal of Community Psychology, 41, 182–196. https://doi.org/10.1007/s10464-008-9162-3 Freudenberg, N., & Tsui, E. (2014). Evidence, power, and policy change in community-based participatory research. American Journal of Public Health, 104, 11–14. https://doi.org/10.2105/AJPH.2013.301471

Changing Behavior Using Theories at the Interpersonal, Organizational, Community, and Societal Levels

Freire, P. (1968). Pedagogy of the Oppressed. New York: Seabury Press. Freire, P. (1974). Education for Critical Consciousness. New York: Seabury Press. Frooman, J., & Murrell, A. J., (2005). Stakeholder influence strategies: The roles of structural and demographic determinants. Business and Society, 44, 3–31. https://doi.org/10.1177/000765030 4273434 Gadotti, M. (1994). Reading Paulo Freire: His Life and Work (trans. J. Milton). New York: State University of New York Press. Galer-Unti, R. A., Tappe, M. K., & Lachenmayr, S. (2004). Advocacy 101: Getting started in health education advocacy. Health Promotion Practice, 5, 280–288. https://doi.org/10.1177/152483990 3257697 Golden, S. D., & Earp, J. A. (2012). Social ecological approaches to individuals and their contexts: Twenty years of health education & behavior health promotion interventions. Health Education and Behavior, 39, 364–372. https://doi.org/ 10.1177/1090198111418634 Green, L. W., & Kreuter, M. W. (2005). Health Program Planning: An Educational and Ecological Approach (4th ed.). New York: McGraw-Hill Professional. Hannam, P. (2019). “We won”: Landmark climate ruling as NSW court rejects coal mine. Sydney Morning Herald, February 8. www.smh.com.au/ environment/climate-change/we-won-landmarkclimate-ruling-as-nsw-court-rejects-coal-mine20190207-p50wer.html Holt-Lunstad, J., & Uchino, B. N. (2015). Social support and health behavior. In K. Glanz, B. K. Rimer, & K. Viswanath (Eds.), Health Behavior: Theory, Research, and Practice (5th ed., pp. 183– 204). San Francisco, CA: Jossey-Bass. Juvonen, J., & Graham, S. (2014). Bullying in schools: The power of bullies and the plight of victims. Annual Review of Psychology, 65, 159–185. https:// doi.org/10.1146/annurev-psych-010213-115030 Kegler, M. C., & Swan, D. W. (2011). An initial attempt at operationalizing and testing the Community Coalition Action Theory. Health Education and Behavior, 38, 261–270. https://doi.org/10.1177/ 1090198110372875 Kelder, S., Hoelscher, D., & Perry, C. L. (2015). How individuals, environments, and health behaviour

263

interact: Social cognitive theory. In K. Glanz, B. K. Rimer, & K. Viswanath (Eds.), Health Behavior: Theory, Research, and Practice (5th ed., pp. 285– 325). San Francisco, CA: Jossey-Bass. Kingdon, J. W. (2003). Agenda, Alternatives, and Public Policies (2nd ed.). New York: Longman. Kok, G., Gottlieb, N. H., Peters, G.-J. Y. et al. (2016). A taxonomy of behavior change methods: An intervention mapping approach. Health Psychology Review, 10, 297–312. https://doi.org/ 10.1080/17437199.2015.1077155 Kok, G., Gurabardhi, Z., Gottlieb, N. H., & Zijlstra, F. R. H. (2015). Influencing organizations to promote health: Applying stakeholder theory. Health Education & Behavior, 42, 123S–132S. https://doi.org/10.1177/109019811 5571363 Kok, G., Peters, L. W. H., & Ruiter, R. A. C. (2017). Planning theory- and evidence-based behavior change interventions: A conceptual review of the intervention mapping protocol. Psicologia: Reflexão e Crítica, 30, 19. https://doi.org/10.1186/ s41155-017-0072-x Kok, G., Schaalma, H., De Vries, H., Parcel, G., & Paulussen, T. (1996). Social psychology and health education. In W. Stroebe & M. Hewstone (Eds.). European Review of Social Psychology, Vol. 7 (pp. 241–282). London: Wiley. Kübler, D. (2001). Understanding policy change with the Advocacy Coalition Framework: An application to Swiss drug policy. Journal of European Public Policy, 8, 623–641. https://doi .org/10.1080/13501760110064429 Lasswell, H. D. (1936). Politics: Who Gets What, When, How. New York: McGraw-Hill. Lehmann, B. A., Chapman, G. B., Franssen, F. M. E., Kok, G., & Ruiter, R. A. C. (2016). Changing the default to promote influenza vaccination among health care workers. Vaccine, 34, 1389–1392. https://doi.org/10.1016/j.vaccine.2016.01.046 Lewin, K. (1947). Quasi-stationary social equilibria and the problem of social change. In T. M. Newcomb & E. L. Hartley (Eds.), Readings on Social Psychology (pp. 73–117). New York: Holt, Rinehart & Winston. Link, B. G., & Phelan, J. C. (2001). Conceptualizing stigma. Annual Review of Sociology, 27, 363–385. https://doi.org/10.1146/annurev.soc.27.1.363

264

ROBERT A. C. RUITER, RIK CRUTZEN, EVELYNE DE LEEUW, AND GERJO KOK

McCullum, C., Pelletier, D., Barr, D., Wilkins, J., & Habicht, J. P. (2004). Mechanisms of power within a community-based food security planning process. Health Education & Behavior, 31, 206– 222. https://doi.org/10.1177/1090198103259163 McLeroy, K. R., Gottlieb, N. H., & Heaney, C. A. (2001). Social health in the workplace. In M. P. O’Donnell (Ed.), Health Promotion in the Workplace (3rd ed., pp. 459–486). Albany, NY: Delmar. Minkler, M. (2004). Community organizing with the elderly poor in San Francisco’s Tenderloin district. In M. Minkler (Ed.), Community Organizing and Community Building for Health (2nd ed., pp. 272– 288). Piscataway, NJ: Rutgers University Press. Minkler, M., & Wallerstein, N. (2012). Improving health through community organization and community building: Perspectives from health education and social work. In M. Minkler (Ed.), Community Organizing and Community Building for Health and Welfare (3rd ed., pp. 37–58). New Brunswick, NJ: Rutgers University Press. National Cancer Institute. (2007). Greater than the Sum: Systems Thinking in Tobacco Control (Publication No. 06–6085, Tobacco Control Monograph No. 18). Bethesda, MD: US Department of Health and Human Services, National Institutes of Health, National Cancer Institute. Nilsen, P., Ståhl, C., Roback, K., & Cairney, P. (2013). Never the twain shall meet? A comparison of implementation science and policy implementation research. Implementation Science, 8, 1. https://doi .org/10.1186/1748-5908-8-63 Oliver, K., & Cairney, P. (2019). The dos and don’ts of influencing policy: a systematic review of advice to academics. Palgrave Communications, 5, 21. https://doi.org/10.1057/s41599-019-0232-y Peters, G.-J. Y., De Bruin, M. & Crutzen, R. Everything should be as simple as possible, but no simpler: Towards a protocol for accumulating evidence regarding the active content of health behavior change interventions. Health Psychology Review, 9, 1–14. https://doi.org/10.1080/17437199.2013 .848409 Peterson, J. C., Rogers, E. M., Cunningham-Sabo, L., & Davis, S. M. (2007). A framework for research utilization applied to seven case studies. American

Journal of Preventive Medicine, 33(1), S21–S34. https://doi.org/10.1016/j.amepre.2007.03.009 Poland, B., Krupa, G., & McCall, D. (2009). Settings for health promotion: an analytic framework to guide intervention design and implementation. Health Promotion Practice, 10, 505–516. https:// doi.org/10.1177/1524839909341025 Rein, M., & Schön, D. (1996). Frame-critical policy analysis and frame-reflective policy practice. Knowledge and Policy, 9, 85–104. https://doi.org/ 10.1007/BF02832235 Reininger, B. M., Barroso, C. S., Mitchell-Bennett, L. et al. (2010). Process evaluation and participatory methods in an obesity-prevention media campaign for Mexican Americans. Health Promotion Practice, 11, 347–357. https://doi.org/10.1177/ 1524839908321486 Rogers, E. M. (2003). Diffusion of Innovations (5th ed.). New York: Free Press. Rothman, J. (2008). Multi modes of community intervention. In J. Rothman, J. L. Ehrlich, & J. E. Tropman (Eds.), Strategies of Community Intervention (7th ed., pp. 141–170). Peosta, IA: Eddie Bowers. Ruiter, R. A. C., & Crutzen, R. (2019). Core processes: how to use evidence, theories and research in planning behavior change interventions. PsyArXiv. https://doi.org/10.31234/osf.io/j4ftz Ruiter, R. A. C., Massar, K., Van Vugt, M., & Kok, G. (2013). Applying social psychology to understanding social problems. In A. Golec de Zavala & A. Cichocka (Eds.), Social Psychology of Social Problems: The Intergroup Context (pp. 337–362). New York: Palgrave Macmillan. Sabatier, P. A. (1988). An advocacy coalition framework of policy change and the role of policyoriented learning therein. Policy Sciences, 21, 129–168. https://doi.org/10.1007/BF00136406 Sabatier, P. A., & Weible, C. M. (Eds.). (2014). Theories of the Policy Process. Boulder, CO: Westview Press. Schein, E. H. (2010). Organizational Culture and Leadership (4th ed.). San Francisco, CA: JosseyBass. Schutte, L., Meertens, R. M., Mevissen, F. E., Schaalma, H., Meijer, S., & Kok, G. (2014). Long

Changing Behavior Using Theories at the Interpersonal, Organizational, Community, and Societal Levels

Live Love. The implementation of a school-based sex-education program in the Netherlands. Health Education Research, 29: 583–597. https://doi.org/ 10.1093/her/cyu021 Shen, F., & Han, J. (2014). Effectiveness of entertainment education in communicating health information: A systematic review. Asian Journal of Communication, 24, 605–616. https://doi.org/ 10.1080/01292986.2014.927895 Siero, S., Boon, M. E., Kok, G., & Siero, F. W. (1989). Modification of driving behavior in a large transport organization: a field experiment. Journal of Applied Psychology, 74, 417–423. https://doi .org/10.1037/0021-9010.74.3.417 Slater, M. J., & Barker, J. B. (2019). Doing social identity leadership: Exploring the efficacy of an identity leadership intervention on perceived leadership and mobilization in elite disability soccer. Journal of Applied Sport Psychology, 31, 65–86. https://doi .org/10.1080/10413200.2017.1410255 Smedley, B. D., Syme, S. L., & the Committee on Capitalizing on Social Science and Behavioural Research to Improve the Public’s Health. (2000). Promoting Health: Intervention Strategies from Social and Behavioral Research. Washington, DC: Division of Health Promotion and Disease Prevention, Institute of Medicine. Smelser, N. J. (1998). Social structure. In N. J. Smelser (Ed.), Handbook of Sociology (pp. 103–130). Newbury Park, CA: SAGE. Smelser, N. J. (2011). Sociological Theory – A Contemporary View: How to Read, Criticize and Do Theory. New Orleans, LA: Quid Pro Books. Snow, D. A., Soule, S. A., & Kriesi, H. (2004). Mapping the terrain. In D. A. Snow, S. A. Soule, & H. Kriesi (Eds.), Blackwell Companion to Social Movements (pp. 3–16). Oxford: Blackwell. Steckler, A., Goodman, R. M., & Kegler, M. C. (2002). Mobilizing organizations for health enhancement. In K. Glanz, B. K. Rimer, & F. M. Lewis (Eds.). Health Behavior and Health Education: Theory, Research and Practice (3rd ed., pp. 335–360). San Francisco, CA: Jossey-Bass. Szreter, S., & Woolcock, M. (2004). Health by association? Social capital, social theory, and the political economy of public health. International

265

Journal of Epidemiology, 33, 650–667. https://doi .org/10.1093/ije/dyh013 Tolli, M. V. (2012). Effectiveness of peer education interventions for HIV prevention, adolescent pregnancy prevention and sexual health promotion for young people: A systematic review of European studies. Health Education Research, 27, 904–913. https://doi.org/10.1093/her/cys055 Turner, J. C. (2005). Explaining the nature of power: A three-process theory. European Journal of Social Psychology, 35, 1–22. https://doi.org/10.1002/ ejsp.244 Valente, T. W. (2015). Social networks and health behavior. In K. Glanz., B. K. Rimer, & K. Viswanath (Eds.), Health Behavior: Theory, Research, and Practice (5th ed., pp. 205–222). San Francisco, CA: Jossey-Bass. Van Beurden, E. K., Kia, A. M., Zask, A., Dietrich, U., & Rose, L. (2013). Making sense in a complex landscape: How the Cynefin Framework from Complex Adaptive Systems Theory can inform health promotion practice. Health Promotion International, 28, 73–83. https://doi.org/10.1093/ heapro/dar089 Van Empelen, P., Kok, G., Schaalma, H. P., & Bartholomew, L. K. (2003). An AIDS risk reduction program for Dutch drug users: An intervention mapping approach to planning. Health Promotion Practice, 4, 402–412. https:// doi.org/10.1177/1524839903255421 Verschuuren, J. (2019). The State of the Netherlands v Urgenda Foundation: The Hague Court of Appeal upholds judgment requiring the Netherlands to further reduce its greenhouse gas emissions. Review of European, Comparative and International Environmental Law, 28, 94–98. https://doi.org/10.1111/reel.12280 Wallack, L., Dorfman, L., Jernigan, D., & Themba, M. (1993). Media Advocacy and Public Health: Power for Prevention. Newbury Park, CA: SAGE. Wallerstein, N., Minkler, M., Carter-Edwards, L., Avila, M., & Sánchez, V. (2015). Improving health through community engagement, community organization, and community building. In K. Glanz, B. K. Rimer, & K. Viswanath (Eds.), Health Behavior: Theory, Research, and Practice (5th ed., pp. 277–300). San Francisco, CA: Jossey-Bass.

266

ROBERT A. C. RUITER, RIK CRUTZEN, EVELYNE DE LEEUW, AND GERJO KOK

Wallerstein, N., Sanchez, V., &Velarde, L. (2004). Freirian praxis in health education and community organizing: A case study of an adolescent prevention program. In M. Minkler (Ed.), Community Organizing and Community Building for Health (2nd ed., pp. 218–236). New Brunswick, NJ: Rutgers University Press. Wandersman, A., Chien, V. H., & Katz, J. (2012). Toward an evidence-based system for innovation support for implementing innovations with quality: Tools, training, technical assistance, and quality assurance/quality improvement. American Journal of Community Psychology, 50, 445–459. https://doi.org/10.1007/s10464-012-9509-7 Weber, M. (1947). The Theory of Social and Economic Organization. New York: Oxford University Press. Weible, C. M. (2008). Expert-based information and policy subsystems: A review and synthesis. Policy Studies Journal, 36, 615–635. https://doi.org/ 10.1111/j.1541-0072.2008.00287.x Weible, C. M., Sabatier, P. A., & McQueen, K. (2009). Themes and variations: Taking stock of the Advocacy Coalition Framework. Policy Studies Journal, 37, 121–140. https://doi.org/10.1111/ j.1541-0072.2008.00299.x Weick, K. E., & Quinn, R. E. (1999). Organizational change and development. Annual Review of Psychology, 50, 361–386. https://doi.org/10.1146/ annurev.psych.50.1.361 Wendel, M. L., Burdine, J. N., McLeroy, K. R., Alaniz, A., Norton, B., & Felix, M. R. J. (2009). Community capacity: Theory and application. In

R. J. DiClemente, R. A. Crosby, & M. C. Kegler (Eds.), Emerging Theories in Health Promotion Practice and Research (2nd ed., pp. 277–302). San Francisco, CA: Jossey-Bass. WHO (World Health Organization). (1986). Ottawa charter for health promotion. Presented at the International Conference on Health Promotion, Ottawa, Canada. www.euro.who.int/__data/ assets/pdf_file/0004/129532/Ottawa_Charter. pdf?ua=1 Wiecha, J. L., El Ayadi, A. M., Fuemmeler, B. F. et al. (2004). Diffusion of an integrated health education program in an urban school system: Planet Health. Journal of Pediatric Psychology, 29, 467–474. https://doi.org/10.1093/jpepsy/ jsh050 Wiggins, N. (2012). Popular education for health promotion and community empowerment: A review of the literature. Health Promotion International, 27, 356–371. https://doi.org/ 10.1093/heapro/dar046 Yip, J. Y., Lee, R. W., & Tsui, E. (2015). Examining knowledge audit for structured and unstructured business processes: A comparative study in two Hong Kong companies. Journal of Knowledge Management, 19, 514–529. https://doi.org/ 10.1108/JKM-10-2014-0420 Zahariadis, N. (2007). The multiple streams framework: Structure, limitations, prospects. In P. A. Sabatier (Ed.), Theories of the Policy Process (2nd ed., pp. 65–92). Boulder, CO: Westview Press.

Part II Methods and Processes of Behavior Change: Intervention Development, Application, and Translation

19 Design, Implementation, and Evaluation of Behavior Change Interventions: A Ten-Task Guide Charles Abraham and Sarah Denford

Practical Summary Behavior change interventions should meet real needs and be designed so that they can have real-world impact. This chapter highlights ten tasks that need to be completed to develop, implement, and test evidence-based behavior change interventions. This means analyzing behavior patterns in context and investigating whether the best method of change is persuading individuals or changing environments or policies and legislation. It means applying psychological theory to understand the mechanisms that maintain behavior patterns for each targeted group. This may involve application of many theories and identification of the relevant intrapersonal, interpersonal, and social processes regulating a targeted behavior pattern. Once mechanisms have been mapped into a logic model, evidence-based change techniques can be selected. Collaborative design facilitates engagement and successful implementation, optimizing the chances of effectiveness.

19.1 Introduction Resolution or amelioration of problems often requires a person, a group, an organization, a community, a policy-making body, or a government to change what they are doing. So, behavior change interventions have been developed to target a wide variety of people and organizations and have been evaluated to assess their effectiveness in changing many behavior patterns. Targeted individual behavior patterns include improved diet (Sacerdote et. al., 2005), increased physical activity (Sheeran et. al., 2019), improved sleep (Capezuti et al., 2018), reduced gambling (Rodda et al., 2018), and reduced energy-use patterns in the context of global warming (Kurz et al., 2014). Organizational interventions include changing policies and practices to increase productivity or improve the safety,

health, or quality of life of their employees (Nielsen et al., 2010), as well as attempts to change individual behavior through organizational change, including worksite interventions (Abraham & Graham-Rowe, 2009) and school-based interventions (Denford et al. 2016; Pearson et al., 2015). Targeted community practices include helping farmers preserve water quality by altering crop planting, promoting harvesting practices to protect water quality (Blackstock et al., 2010), and reducing health risks (O’Loughlin et al., 1999). Behavior change is often dependent on changes to people’s surroundings and/or how they interact with them, including changes to the environments within which they make decisions, whether these are restaurants, supermarkets, or online https://doi.org/10.1017/9781108677318.019

270

CHARLES ABRAHAM AND SARAH DENFORD

environments (Thaler & Sunstein, 2008; VicHealth, 2018; see Chapters 14 and 42, this volume). For example, one might want to persuade a government to legislate in relation to the content of food products or the information describing that content (Lobstein & Davies, 2008; House of Lords, 2011) or, alternatively, to raise taxes on harmful products. Regulation at organizational, city, and national levels is especially important when widespread behavior patterns result in high health, wellbeing, and economic costs. Interventions may also be structural. For example, urban planning can ensure shade in cities to reduce temperatures and skin cancer risks (Barton & Grant, 2006; Municipal Association of Victoria, 2019). Alternatively, they may involve changes to legislative and policy environments (Glanz et al., 2005). Behavior change interventions can be effective. For example, the UK’s National Institute for Health and Care Excellence (NICE) commissioned a review that included data from 103 systematic reviews of interventions targeting one of six behavior patterns (cigarette smoking, alcohol consumption, physical activity, healthy eating, drug use, and sexual risk-taking). The review found that, although the degree of effectiveness varied across populations and intervention

characteristics, interventions were, in general, effective in changing behavior patterns (Jepson et al., 2010). Moreover, when effective, preventive behavior change interventions can be very cost-effective because they save health care costs. Many useful context-specific and generic frameworks have been developed to guide practitioners, policy makers, and scientists through the intervention design and evaluation process (Milstein et al., 2000; Chapter 22, this volume). This chapter will draw on the intervention mapping framework (Bartholomew et al., 2016; Chapter 18, this volume) to highlight essential tasks involved. Whatever the behavior change pattern targeted, whatever the level (individual to population), whatever intervention development guidance is employed, and whatever models of behavioral regulation are identified, effectiveness will be optimized by working carefully through these tasks.

19.2 Ten Tasks The ten tasks listed in Table 19.1 may seem reassuringly obvious because they follow scientific method. In practice, however, each of these tasks can generate multiple ways forward, not all of which will generate useful research or helpful

Table 19.1 Ten tasks involved in the design, development, implementation, and evaluation of behavior change interventions 1. Understand and define the problem (including assessing the needs of target groups). 2. Clarify how behavior change can ameliorate or resolve the problem. 3. Identify which group/s of people need to change which behavior, or behavior patterns, and at what level. 4. Understand behavioral antecedents, that is, the contexts, cues, and underlying mechanisms that maintain the targeted behavior patterns for each targeted group. 5. Design interventions/intervention components that can alter some, or all, of these behavioral antecedents and regulatory mechanisms. 6. Pilot, or pretest, intervention prototypes to discover whether they are acceptable, feasible, and affordable. 7. Refine and develop the intervention with those who will implement and experience it to optimize fidelity of implementation and effectiveness. 8. Implement the intervention and identify and minimize embedding problems. 9. Evaluate efficacy by investigating whether the intervention shows evidence of changing targeted antecedents and behaviors. 10. Evaluate effectiveness by testing the intervention in new contexts and scaling up to target new groups or populations.

Design, Implementation, and Evaluation of Behavior Change Interventions

outcomes. So, as recommended in the intervention mapping framework (Bartholomew et al., 2016), it is important to think through what is involved in each of these tasks, even when designing something as simple as a health-promotion leaflet (Lake et al., 2018). Designers may need to cyclically return to earlier tasks as intervention development proceeds. In the next section, some of the challenges and pitfalls that may be encountered are considered, with a particular focus on tasks 4 and 5. Working through these tasks necessitates careful planning, consultation, and pretesting, followed by design refinement and gradual upscaling toward widespread use. Evidence is critical to optimizing intervention effectiveness. Many types of research can generate the evidence needed at each stage. Exploratory research, sometimes called elicitation research, is required at the beginning when needs are being assessed and the key challenges defined. For example, it is important to know what the target audience believes and what they want to do. Epidemiological and medical research may also be important when prioritizing change targets (see tasks 1 and 3). Experimental psychological research is essential to tasks 4 and 5, and this chapter will focus most closely on these tasks (see also Chapter 20, this volume). Detailed qualitative and observational research is likely to facilitate intervention pretesting, refinement efficacy, and effectiveness assessment (tasks 6–8; Chapters 21 and 30, this volume). Finally, efficacy and effectiveness trials are needed to assess whether the implemented intervention changes behavior patterns as planned (tasks 9–10; Chapter 23, this volume).

19.3 Understand the Problem and Identify Behavior Change Targets (Tasks 1–3) The increasing prevalence of obesity is a global public health priority because being overweight is associated with numerous health problems,

271

including Type 2 diabetes, coronary heart disease, hypertension, osteoarthritis, sleep apnea, hormonal abnormalities, and some cancers (WHO, 2016). Consequently, those who are overweight have increased morbidity and mortality risks, which generates large increases in health service and community costs. It has been estimated that, in 2005, the costs of obesity to Australia was AU $21 billion (Colagiuri et al., 2010) and that the cost to the United States is US$147 billion annually (CDC, 2017). Even higher costs are incurred with driving behavior. For example, the National Highway Traffic Safety Administration (2008) has estimated that road traffic accidents in the United States killed 10,497 people in 2016 and cost more than $830 billion annually. Changing behavior at the population level can save lives and save money. Obesity is caused by the consumption of energy-dense diets, high in fat and sugars, which, combined with low levels of physical activity, prompt energy storage in the form of fat cells. The underlying behavior change challenges are clear. There is a need to assist people in eating healthier diets and engaging in more physical activity. So, tasks 1 and 2 are relatively easy to define, but who should be targeted? It might be pertinent to prioritize those most in need (e.g., those who are most obese) or those with high blood glucose levels who, through sustained behavior change, can avoid diabetes; or perhaps it might be prudent to target children (or their parents) who may not yet have established unhealthy dietary and sedentary routines; or all of these groups. In each case, different challenges arise. Should governments be “lobbied” to change the content of food and drink products or ensure improved labeling? Should communities be empowered to change how takeaway outlets prepare food products (Hillier-Brown et al., 2014, 2017), or should restaurants and bars be helped to make some foods and drinks more prominent or available (VicHealth, 2018)? Should schools be

272

CHARLES ABRAHAM AND SARAH DENFORD

helped to promote new dietary and physical activity habits among children (Wyatt et al., 2017)? Perhaps greater participation in physical activity should be promoted through workplace changes (Abraham & Graham-Rowe, 2009), or the establishment of weight-loss groups in the community (Borek et al., 2018), or the provision of guidance to general practitioners on how to help their patients adopt healthier diets (Sacerdote et al., 2005), or even food rehabilitation programs (Kessler, 2009). All these approaches have been tried, and many trials have been conducted to evaluate such interventions. These examples highlight that an important element of tasks 1–3 is deciding at what level or levels it is most appropriate to intervene. Individuals need to change but how can that change best be facilitated, especially across populations? Should efforts be focused on individuals themselves through education, or through the organizations that may have the capacity to change normative patterns of behavior, or community practices, or legislation at the governmental level (see Chapters 18 and 28, this volume). In the case of widespread problems that prove difficult to resolve, the most effective response is likely to be interventions at multiple levels (Glanz et al., 2005). Even potentially effective interventions may not have population-level effects unless they are part of a comprehensive multilevel strategy (WHO, 2016; Wyatt et al., 2017). Of course, all such interventions, whether implemented in organizations, communities, or through legislation need to be evaluated and, depending on effectiveness, abandoned, replaced, or refined (see tasks 9 and 10; Bowen et al., 2015; Grimshaw et al., 2004; House of Lords, 2011; Chapter 22, this volume). This section has highlighted consideration of different types of interventions, targeting different groups at different levels, and emphasized the importance of multilevel, multi-intervention strategies to tackle population change. In the next section, the theory-based strategies and

mechanisms that have been applied to evoke behavior change are reviewed.

19.4 Understanding Mechanisms and Including Evidence-Based Change Techniques (Tasks 4 and 5) Carefully designed interventions may fail because interventionists do not understand, or do not appropriately target, the processes that generate behavior patterns (Kelly & Barker, 2016). Developing accurate descriptions of mechanism (that is, theories or a set of theories) and using these descriptions to identify and assemble precisely engineered, target-bespoke change techniques are central and prerequisite to effective behavior change intervention. So, what mechanisms are important? Unfortunately, there is no easy answer to this question; mechanisms must be considered on a behavior-by-behavior and target-group-to-target-group basis. The academic disciplines of psychology, sociology, and behavioral economics provide a wide range of explanations of behavioral consistency and variability across contexts and groups. Precise articulation of these processes, in task 4, is critical to success. Interventionists need to (1) select theories of behavioral regulation that have a persuasive evidence base and are directly relevant to the mechanisms underpinning the behavior change challenge defined when completing task 3 and, as in task 5, (2) translate those theories into the design of behavior change techniques that, collectively, constitute a potentially effective intervention. Detailed guidance on how to manage this challenge is provided by the intervention mapping framework (Bartholomew et al., 2016). The outcome of tasks 4 and 5 is a logic model (Moore et al., 2014; Chapter 20 and Chapter 46, this volume). A logic model is not a theory but a bespoke map of what an intervention is designed to change and how it will work. There are many components to a logic model (W. K. Kellogg

Design, Implementation, and Evaluation of Behavior Change Interventions

Foundation, 2004; McLaughlin & Jordan, 1999). Here the focus is on the intended mechanism or process changes, that is, the interim changes that will precede the targeted behavior change (see also Hagger et al., 2020; Suls et al., 2020; Chapters 20 and 46, this volume). Consider, for example, an intervention designed to prompt people to test concentrations of radon gas in their houses. How would it work? It might inform people of the sources of radon gas in their area, explain their susceptibility to negative consequences of exposure, and offer easily accessible, low-cost test kits. In this case, the logic model of the intervention postulates increased (1) knowledge, (2) perceived susceptibility, (3) perceived severity, (4) testing confidence, and (5) test intention. These are five distinct, measurable psychological outcomes and their assessment is critical to the process evaluation that needs to be undertaken as part of task 9. Not only are these changes expected to be observed but they are also expected to predict uptake of test kits, the primary outcome against which the intervention’s effectiveness will be evaluated during task 9 (Weinstein et al., 1991). Articulating and measuring interim mechanistic outcomes in the logic model is critical to understanding why interventions do, or do not, work. For example, if this intervention failed, the intervention designers could identify whether that uptake failure was attributable to a failure to increase one or more of the five expected psychological changes. This is crucial to refining and improving the intervention when, after failure, the interventionists return to tasks 6 and 7. Consider a second public health problem. People failing to turn up for health care appointments (“no-show” appointments) create considerable costs for health care systems. For example, no-shows have been estimated to cost US health care systems US$150 billion every year (Gier, 2017). Younger, poorer patients living further from health care facilities, with no private health insurance and who made appointments further in advance, are more likely to miss appointments

273

(Dantas et al., 2018). No-show patients are likely to receive poorer care, including preventive cancer screening, to have poorer outcomes for long-term conditions, and are more likely to be hospitalized. Moreover, for some patients, such as those with long-term illnesses, including mental health problems, greater numbers of missed appointments predict increased mortality risk, including suicide risk (McQueenie et al., 2019). Thus, no-show patients also represent a high-risk patient group who can be identified by health services. Consideration of tasks 1–3 is more manageable here than in relation to obesity prevalence. Research highlights which patients are most likely to suffer the consequences of no-shows and, also, which patient groups are more likely to miss appointments. This immediately helps us think about priority targets. Moreover, while societal level factors such as trust in the national or local health services may shape no-show patterns and, for some patients, with long-term illnesses, their medical appointments are an ongoing part of their everyday lives, no-shows are more easily conceptualized as discrete action sequences than, for example, eating. Consequently, a series of intervention types can be derived from the task 4 analysis of antecedents of attending a medical appointment. For any appointment, it seems likely that (1) how salient attendance is, (2) how important it is compared to other priorities, (3) how easy it is to attend, and (4) what the perceived outcomes of attendance and nonattendance are will be important predictors of attendance (Kaplan-Lewis & PercacLima, 2013). Therefore, reminding patients, and particularly those from groups least likely to attend, in an easy-to-access manner, providing persuasive information that can change perceived outcomes, changing the rewards experienced by patients during and after appointments, and making contact easy by, for example, providing transport or parking or conducting appointments virtually are all candidate components for interventions to reduce no-shows (Molfenter, 2013).

274

CHARLES ABRAHAM AND SARAH DENFORD

The intervention is, therefore, likely to focus on communication with individual patients and groups of patients and on the practices of health care providers, both practitioners and organizations. This initial analysis provides helpful parameters for task 3 and task 4 design. Next, some important psychological ideas that may be useful to such analyses are considered.

19.4.1 Modeling Prerequisite Motivation and Skills The information–motivation–behavioral skills model (Figure 19.1; Fisher & Fisher, 1992) proposes that behavior change occurs when individuals are well informed, highly motivated, and have the skills necessary to perform the behavior. The model has been used successfully to design effective health behavior change interventions (Fisher et al., 2002). This model poses a useful set of questions when developing a logic model; are people informed, are they motivated, and do they have the skills needed to act? Informed people may not act because they are not motivated (other things are more important) or because they lack the capacity to act (the change is, or seems to be, too difficult). When an action sequence is easy and important, an uninformed person may change in response to information provision. However, research may be needed to establish what information is most effective and how it is best presented. For

example, does calorie information help people regulate their food intake or would traffic light guidance (red = unhealthy, amber = not too often, and green = heathy) be more useful (House of Lords, 2011)? What information does a newly diagnosed Type 2 diabetic need to attend a retinopathy screening test (Lake et al., 2018)? These questions can be answered by elicitation research, including surveys and interviews, prior to intervention design. Fishbein (2001; see also Chapter 2, this volume) proposed that motivation is influenced by (1) an individual’s beliefs about the advantages of changing their behavior outweighing the disadvantages (their attitudes; “it’s a good thing to do”); (2) their anticipation that changing behavior will lead to a positive emotional reaction (anticipated affect; “it’ll feel good”); (3) their beliefs that others want them to change their behavior and that others are adopting the target behavior (normative beliefs); (4) their belief that the behavior is consistent with their own self-image and corresponds to who they think they are; and (5) their belief that they can successfully change the behavior in question (self-efficacy; Bandura, 1977; Chapter 3, this volume). This checklist of antecedents can help inform the design of elicitation research and, subsequently, the mechanistic targets specified in the logic model for the intervention. It is important to know which elements of motivation are missing if the intervention is to target increases in motivation. For example, in thinking about radon testing,

Health behavior information

Health behavior skills

Health behavior

Health behavior motivation

Figure 19.1 The information–motivation–behavioral skills model (Fisher & Fisher, 1992)

Design, Implementation, and Evaluation of Behavior Change Interventions

should there have been a consideration of normative beliefs, perhaps suggesting that many neighbors had already tested their houses? Normative beliefs may be especially important when targeting young people (Albarracín et al., 2005). A wide range of techniques can be used to increase motivation. For example, fear appeals may increase motivation by presenting graphic portrayals of frightening consequences to alter beliefs about consequences and anticipated affect (Witte & Allen, 2000; see also Chapter 34, this volume). This type of behavior change technique is most likely to be effective when the target audience lacks motivation but has the necessary skills to perform the behavior (Ruiter et al., 2001). If individuals are lacking in self-efficacy or behavioral skills, fear may increase anxiety (because the person is faced with a potentially severe threat) but leave recipients disempowered. One way to escape from this uncomfortable state is to deny or undermine the relevance of the threat (“it won’t happen to me”; “it’s all exaggerated”; etc.). Such defensive responses undermine motivational change (Rogers, 1975; Witte & Allen, 2000). It is important, therefore, to carefully assess the target population before using fear appeals. This analysis also emphasizes that certain types of change techniques, such as fear appeals, are only likely to be effective if certain prerequisites are already in place, such as sufficient levels of self-efficacy (Peters, de Bruin, & Crutzen, 2015), thereby identifying objectives for elicitation research prior to completion of task 5. People are motivated to do many things and have moment-to-moment, multiple and hierarchically organized goals that imply sequences of subgoals and incorporate the potential for goal conflict (Austin & Vancouver, 1996). For example, Carver and Scheier’s (1982) control theory posits self-related goals or system concepts (e.g., “be a successful person”) at the top of the hierarchy, abstract action goals (principles; e.g., “work hard at my job”) in the middle, and courses of action (programs; e.g., “work at weekends”) at

275

the bottom (see Powers, 1973; Chapter 9, this volume). This then needs to be translated into even more specific goals (e.g., “finish this chapter on Sunday”). At all levels in this hierarchy, goals may be thwarted by higher priorities. For example, being a “good” (rather than successful) person may mean that other family priorities become more important on a Sunday when a book chapter was meant to be finished the following day. Even when motivation is established, goal priority may be critical to the translation of intentions into action (Conner et al., 2016), so the development of goal rescheduling and prioritization techniques during task 5 is recommended. Motivated people may fail to change if they lack confidence and skills. People who believe they can succeed (i.e., have high self-efficacy) set themselves more challenging goals, exert more effort, use more flexible problem-solving strategies, and are more persistent (Bandura, 1997) than those with low self-efficacy. Bandura (1997) proposed that self-efficacy can be enhanced through four types of change technique: mastery experiences in which the individual successfully performs the behavior, which implies learning in graded steps; vicarious experience in which the individual observes others successfully enacting the behavior, particularly if that person is rewarded or experiences positive outcomes; verbal persuasion, especially by highlighting past successes and the success of others who have no greater skills; and perception and regulation of physiological and affective states, that is, recognizing and managing emotional arousal (see Chapters 3 and 32, this volume). Self-efficacy is a prerequisite to sustained motivation and action and so needs to be considered during task 4 and may necessitate inclusion of self-efficacy–enhancing techniques during task 5, if elicitation research has not confirmed self-efficacy to be high. Most behavior changes require enhancement or development of skills. For example, when trying to lose weight, new cooking skills may be needed, or new exercise skills (e.g., a popular

276

CHARLES ABRAHAM AND SARAH DENFORD

mobile device app promises to help the user to develop running skills that will take them from “couch to 5 K”), or, crucially, the social skills needed to manage interactions involving food (e.g., refusing a slice of cake a friend has baked and brought to work or avoiding a Saturday night takeaway with friends). The development of behavioral skills requires engaging, easy-tounderstand, graded instruction, along with practice and feedback, whether this is in printed form, using an app or online intervention, or in a faceto-face setting. Without preparatory skills, a person trying to change a behavior pattern may fail because of low self-efficacy or poor proficiency. This is a key consideration for task 5.

19.4.2 Changing Habits Often behavior change is conceptualized as helping people make the right choices. Choice implies rational consideration of the “pros” and “cons” of alternative courses of action. Yet many problematic behavior patterns may be enacted without deliberation; they may be automatic or habitual in context. So, changing the context may be crucial to changing the behavior. A series of dual-process models have been developed to explain the difference between choices and habits. These models characterize two distinct systems that regulate everyday behavior patterns (see Strack & Deutsch, 2004; Borland, 2014; Chapters 12 and 15, this volume). One system regulates deliberative control of action (choice), while the other operates automatic action sequences involving low levels of conscious monitoring (habit). The same behavioral sequence may involve one or both systems in different contexts. Consider, for example, leaving home and locking the outer door. If you have lived in the same place for a while you may do this while thinking of other things. The behavioral sequence can be initiated and completed automatically with little conscious monitoring and, consequently may be difficult to remember later. Of course, you are conscious

while locking the door and so can engage in conscious control. For example, if, while you are leaving, a friend says, “Don’t forget to lock the door and take your key,” then you may more carefully monitor your actions and make changes to your usual routine such as checking the door a second time. Such activation of conscious control facilitates changes in habitual behavior patterns and enhances action recall (Chapter 37, this volume). So, task 4 involves consideration of the habitual nature of the target behavior change and task 5 may need to focus on habit-making and habit-breaking change techniques (Dean, 2013). Habit-breaking interventions have been designed and evaluated (e.g., Chapter 41, this volume). For example, Dean (2013) describes habit reversal therapy for people with Tourette’s syndrome. This involves becoming aware of the tics and the situations in which they occur and then practicing a competing behavior. The intervention is effective but intensive, involving more than ten hours of therapeutic contact (McGuire et al., 2014). This has interesting parallels with Kessler’s (2009) “food rehabilitation,” which also begins with building awareness, both of everyday cues that prompt overeating and of “premonitionary urges” or feelings that signal the activation of a habitual response. By training conscious intervention just before responding to cues, individuals can create “mental space” for reflective initiation of competing behaviors, just as a friend’s reminder can change how you lock your front door. The next stage of food rehabilitation is learning competing behaviors, such as deliberately walking by the doughnut shop or taking another route to work. Kessler emphasizes the need for cognitive and emotional reconstruction, that is, learning to reflectively nurture new thoughts and emotions in relation to conditioned cues. Over time, this results in attitude change that bolsters motivation to change (Chapter 31, this volume). Habit-breaking involves practice and usually depends on social support. Similarly, habit-making requires sustained practice and effort even in the face of setbacks.

Design, Implementation, and Evaluation of Behavior Change Interventions

19.4.3 If-Then Plans and Precision of Mechanism Specification A review of types of change techniques that have been found to be effective in changing, for example, habitual eating highlighted if-then planning as a promising type of technique (van Beurden et al., 2016). Such techniques have proved effective in helping motivated people translate their intentions into action (Gollwitzer & Sheeran, 2006). Luszczynska, Sobczyk, and Abraham (2007) found that a single lesson in if-then planning helped overweight women attending a weightloss class double their weight loss at two-month follow-up. Sheeran and Orbell (2000) included an if-then planning prompt in a questionnaire administered to woman who should attend cervical cancer screening. Recipients were provided with space to write “when, where and how you will make an appointment.” In this simple intervention, motivated people were prompted to make specific action plans (e.g., specifying a time) to later prompt a response that they had the skills to enact (making an appointment). This was effective; 92 percent of those offered this planning opportunity attended for screening compared to 67 percent in the no–ifthen control group, despite equivalent levels of motivation before intervention. Sheeran, Aubrey, and Kellett (2007) aimed to decrease no-shows at psychotherapy appointments utilizing a questionnaire. Recipients were advised to use a self-talk technique to terminate advance worrying about their appointment: “As soon as I feel concerned about attending my appointment I will ignore that feeling and tell myself this is perfectly understandable.” Unlike the if-then technique employed by Sheeran and Orbell (2000), this technique does not prompt “action” planning – no behavioral response is specified. This is a thought-regulation technique designed to reduce cognitive elaboration of appointment-related anxieties (Chapter 34, this volume). This intervention was also successful with 75 percent attending compared to 63 percent in the

277

control group. These two interventions highlight a critical point regarding intervention design. Both aimed to change appointment attendance, both targeted motivated people, both used questionnaires as the delivery format, both advocated if-then planning, but they targeted different mechanisms of change using different techniques. The logic models and intervention design were different. Clarifying mechanisms at this specificity, during task 4, combined with an application of correspondingly specific, tailored change techniques during task 5 is a prerequisite to effective behavior change intervention design. Transferring previously successful techniques, or categories of techniques (such as if-then plans), to a new behavior challenge may prove ineffective if the mapping of mechanisms during task 4 is mistaken or imprecise or the conditions of operation of selected techniques are not present in the new situation (Peters et al., 2015).

19.5 Cocreate, Pretest in Context, and Refine (Tasks 6, 7, and 8) Tasks 3–5 are likely to optimize intervention effectiveness and generate better results if undertaken jointly with experienced representatives of those who will deliver or host the intervention, including health care practitioners, and purposively chosen representatives of those who will receive the intervention, for example those who need to lose weight or attend appointments more frequently (see Chapters 23 and 24, this volume). The combination of behavior change interventionists with expertise in specifying change mechanisms and selecting corresponding change techniques, who can lead tasks 3 and 4, with those who understand how the intervention is likely to be experienced and used in context, perhaps at particular “teachable” moments, is critical to the precision design required by task 5. Pretesting is listed after design in Table 19.1 but, in reality, testing begins as soon as potentially effective techniques emerge from the design process (Bartholomew et al., 2016). Is

278

CHARLES ABRAHAM AND SARAH DENFORD

the intervention idea acceptable, comprehensible, engaging, and likely to be used as designed? These are critical questions that need to be answered positively before investment in expensive design and evaluation begins. They need to be considered from task 1 onward and need to be interrogated in detail once an intervention design emerges from task 5 (Chapter 21, this volume). An intervention that is unattractive, unaffordable, or unsuitable in context will not be adopted or, if adopted, will not be implemented with fidelity and so is likely to be ineffective because it is not delivered as designed. Thus, implementation (task 8) needs to be considered from the outset and considerable pretesting and refinement of the intervention in context may be required before a usable intervention is tested in context. Hence the importance of pretesting intervention prototypes (task 6). Reviews that examine the characteristics of effective interventions in particular contexts can provide guidelines on change techniques and delivery formats that may be most likely to succeed once change mechanisms are understood (Johnson & May, 2015; Pearson et al., 2015). For example, ensuring the motivation of key champions and opinion leaders and, thereby, explicit norms and guidelines, is likely to be important in changing practitioners’ behavior in organizations. Ensuring good training and feedback for those delivering intervention is also likely to be crucial. The Reach, Effectiveness, Adoption, Implementation, and Maintenance (RE-AIM) framework sets out key criteria that should be assessed to indicate whether or not an implemented intervention is useful in real-world settings (Glasgow Vogt, & Boles, 1999; Chapter 23, this volume). Successful implementation depends not only on effectiveness but also on reaching the target population through adoption and fidelity of delivery. This will depend on how compatible the intervention is with the goals of the organization or community in which it is to be delivered. Moreover, if an intervention is time-consuming and expensive then, even if it is implemented successfully as part of a trial, it may

not be maintained beyond the evaluation. Thus, optimizing the intervention’s performance in relation to the RE-AIM criteria is a key outcome of tasks 6 and 7.

19.6 Evaluation Efficacy and Effectiveness (Tasks 9 and 10) The gold standard test of intervention effectiveness is a randomized controlled trial (RCT), and this can be the main element of task 10 (Chapter 22, this volume). Individuals, or organizations in the case of a cluster RCT, are randomly assigned to receive the intervention or not (e.g., in no-intervention or usual-care treatments). The primary outcome measure of a trial evaluating a behavior change intervention is the target behavior change. Did people change their diet, take more physical activity, or turn up for appointments? This may be measured immediately after the intervention or after a longer follow-up. In health-related interventions, researchers also want to assess the effect of any observed behavior change. For example, was a weight difference between control and intervention groups maintained after two years (Borek et al., 2018)? A key result for any such trial is the effect size. This is often used to compare the outcomes of trials to identify more, or less, effective interventions. Many excellence guidelines for the design, ethics, and conduct of RCTs are available (see National Institutes of Health, 2017). It is not always possible to randomly assign intervention participants, so other evaluation methods may need to be employed during task 10, including matched sample trials, before and after studies, case studies, and a series of case studies. The reliability and validity of evidence generated by such evaluation methods have been ranked in hierarchies of evidence (e.g., CDC, 2017). Evans (2003) reviews a series of such hierarchies and discusses how the feasibility, appropriateness, and effectiveness of evaluation methods must be considered in selecting the best evaluation method or methods.

Design, Implementation, and Evaluation of Behavior Change Interventions

The evaluation method adopted during task 10 may be expensive. Conducting large RCTs is very expensive. So small-scale evaluations should be conducted initially to identify efficacy or potential effectiveness. For example, an intervention might be tested on a small sample of people or a small set of organizations (e.g., sets of twenty participants or two organizations allocated to each of the intervention and control groups) to discover whether the intervention shows signs of benefit. This is task 9 and it prepares for the final evaluation (Flay, 1986). Again, a series of methods may be employed, including, for example, case studies and postintervention interviews (Chapter 30, this volume). When different methods confirm a potential intervention benefit, this provides a good rationale for a larger-scale effectiveness evaluation during task 10. As noted, when considering task 4, evaluation of an intervention needs to assess the effects on interim outcomes representing mechanism. For example, did the intervention increase selfefficacy, did it increase motivation, or did it prompt if-then planning? Assessing the mechanism of impact is crucial to understanding how an intervention works and is especially important in guiding revisions and refinements when an intervention is ineffective or only partially effective in changing targeted behavior patterns. This a key element of process evaluations that should be conducted alongside outcome evaluations in completing tasks 9 and 10. In addition to assessing the mechanism of impact, a process evaluation should assess the process of implementation (including acceptability, sustainability, and fidelity of delivery) and aspects of the implementation context (e.g., the setting, such as rural vs. urban, community vs. health service; variation in targeted populations, such as age and culture) that may influence intervention effectiveness. An intervention may work better with a particular subgroup (e.g., younger or older people) or when delivered in one context rather than another (e.g., community vs.

279

hospital). These assessments are especially useful when planning to scale up intervention delivery to population levels because adaptions may need to be made to ensure fidelity and effectiveness as implementation and context change (for guidance on process evaluations, see Moore et al., 2014). Even if an intervention is effective in changing behavior in a desirable manner, it may not be implemented beyond a trial if it is too expensive or if the behavior change it instigates makes little difference to public practices or, for example, on health outcomes. Therefore, the final element of task 10 is to conduct longer-term evaluations of interventions at the population level. At this stage, the primary outcome may no longer be behavior change but clinically or practically relevant outcomes. Was the prevalence of obesity, diabetes, or cardiovascular disease reduced? Did health services save money because of reduced no-shows? Public health evaluations need to be both long-term and large-scale and assess whether the intervention will save money or cost more. Economic evaluations necessitate careful measurement of intervention delivery components and implementation across tasks 8–10 (Alayli-Goebbels et al., 2014).

19.7 Conclusion and Implications for Research, Practice, and Translation The key to developing effective behavior change intervention requires knowledge of how particular theories describe the mechanisms underpinning motivation, affect, and behavior and how these theories can be applied to behavior change challenges. This handbook includes many chapters focusing on theories that identify the constructs and types of mechanisms that relate to human behavior and describe the types of change techniques that have been used to target particular mechanisms. For example, Chapter 13 (this volume) highlights current understanding of habitual behavior patterns and Chapter 41 (this volume) explores how these patterns can be

280

CHARLES ABRAHAM AND SARAH DENFORD

leveraged to change habits. This chapter has provided an overview of the processes and challenges involved in developing, implementing, and testing evidence-based behavior change interventions. An iterative sequence of ten tasks has been presented that outline the steps interventionists need to address when developing behavior change interventions. The tasks focus especially on identifying relevant regulatory mechanisms and precisely matching these mechanisms to evidence-based change techniques to create bespoke interventions tailored to the context in which they will be delivered. A series of recommendations based on the ten tasks has been proposed. In summary, interventionists should: • •

•

• •

•

•

• •

Ensure that an intervention is needed before spending time and resources on its development. Follow the principles of intervention mapping, working through the ten tasks highlighted in Table 19.1. Consider carefully the behavior pattern that is to be changed, as well as the characteristics, lifestyles, and culture of those the intervention is designed to help. Use relevant evidence to identify mechanisms of behavior regulation. Choose specific change techniques, ensuring a precise match between techniques, mode of delivery, and specified mechanisms. Cocreate interventions with intended users and implementers to maximize adoption, fidelity of implementation, and sustainability. Ensure those delivering the intervention have appropriate training to safeguard fidelity of delivery. Pilot all intervention materials before implementation. Evaluate interventions using the best quality evidence, employing process, outcome, and economic evaluations beginning with lowercost efficacy assessments and moving to effectiveness tests that justify widespread implementation of effective interventions.

References Abraham, C., & Graham-Rowe, E. (2009). Are worksite interventions effective in increasing physical activity? A systematic review and metaanalysis. Health Psychology Review, 3, 108–144. https://doi.org/10.1080/17437190903151096 Alayli-Goebbels, A. F. G., Evers, S. M. A. A., Alexeeva D. et al. (2014). A review of economic evaluations of behavior change interventions: Setting an agenda for research methods and practice. Journal of Public Health, 36, 336–344. https://doi.org/ 10.1093/pubmed/fdt080 Albarracín, D., Gillette, J. C., Earl, A. N., Glasman, L. R., Durantini, M. R., & Ho, M.-H. (2005). A test of major assumptions about behavior change: A comprehensive look at the effects of passive and active HIV-prevention interventions since the beginning of the epidemic. Psychological Bulletin, 131, 856–897. https://doi.org/10.1037/ 0033-2909.131.6.856 Austin, J. T., & Vancouver, J. B. (1996). Goal constructs in psychology: Structure, process, and content. Psychological Review, 120, 338–375. https://doi.org/10.1037//0033-2909.120.3.338 Bartholomew, L. K., Markham, C. M., Ruiter, R., Fernández, M. E., Kok, G., & Parcel, G. S. (2016). Planning Health Promotion Programmes: An Intervention Mapping Approach. San Francisco, CA: Jossey-Bass. Barton, H., & Grant, M. (2006). A health map for the local human habitat. The Journal for the Royal Society for the Promotion of Health, 126, 252–253. https://doi.org/10.1177/ 1466424006070466 Blackstock, K. L., Ingram, J., Burton, R., Brown, K. M., & Slee, B. (2010). Understanding and influencing behaviour change by farmers to improve water quality. Science of the Total Environment, 408, 5631–5638. https://doi.org/ 10.1016/j.scitotenv.2009.04.029 Borek, A., Abraham, C., Greaves, C., & Tarrant, M. (2018). Group-based diet and physical activity weight-loss interventions: A systematic review and meta-analysis of randomised controlled trials. Applied Psychology: Health and Well-Being. 10, 62–86. https://doi.org/10.1111/aphw.12121

Design, Implementation, and Evaluation of Behavior Change Interventions

Borland, R. (2014). Understanding Hard to Maintain Behaviour Change: A Dual Process Approach. Chichester: Wiley & Sons. Bowen, D. J., Barrington, W. E., & Beresford, S. A. (2015). Identifying the effects of environmental and policy change interventions on healthy eating. Annual Review of Public Health, 36, 289–306. https://doi.org/10.1146/annurev-publhealth032013-182516 Capezuti, E., Zadeh, R. S., Pain, K., Basara, A., Jiang, N. Z., & Krieger, A. C. (2018). A systematic review of non-pharmacological interventions to improve night-time sleep among residents of longterm care settings. BMC Geriatrics, 18, 143. https://doi.org/10.1186/s12877-018-0794-3 Carver, C. S., & Scheier, M. F. (1982). Control theory: A useful conceptual framework for personalitysocial, clinical and health psychology. Psychological Bulletin, 92, 111–135. https://doi .org/10.1037//0033-2909.92.1.111 CDC (Centers for Disease Control and Prevention). (2017). Overweight and obesity. CDC. Website. www.cdc.gov/obesity/adult/causes.html Colagiuri, S., Lee, C. M. Y., Colagiuri, R. et al. (2010). The cost of overweight and obesity in Australia. Australian Medical Journal, 192, 260–264. https:// doi.org/10.5694/j.1326-5377.2010.tb03503.x Conner, M., Abraham, C., Prestwich, A. et al. (2016). Impact of goal priority and goal conflict on the intention-health behavior relationship: Tests on physical activity and other health behaviors. Health Psychology, 35, 1017–1026. https://doi .org/10.1037/hea0000340 Dantas, L. F., Fleck J. L., Cyrino Oliveira, F. L., & Hamacher, S. (2018). No-shows in appointment scheduling: A systematic literature review. Health Policy 122, 412–421. https://doi.org/10.1016/j .healthpol.2018.02.002 Dean, J. (2013). Making Habits, Breaking Habits. Boston: De Capo Press. Denford, S., Abraham, C., Campbell, R., & Brusse, H. (2016). A comprehensive review of reviews of school-based interventions to improve sexualhealth. Health Psychology Review, 11, 33–52. https://doi.org/10.1080/17437199.2016.1240625 Evans, D. (2003). Hierarchy of evidence: A framework for ranking evidence evaluating healthcare

281

interventions. Journal of Clinical Nursing, 12, 77– 84. doi/full/10.1046/j.1365-2702.2003.00662.x Fishbein, M., Triandis, H. C., Kanfer, F. H., Becker, M. H., Middlestadt, S. E., & Eichler, A. (2001). Factors influencing behavior and behavior change. In A. Baum, T. R. Revenson, & J. E. Singer (Eds.), Handbook of Health Psychology (pp. 3–17). Hillsdale, NJ: Lawrence Erlbaum Associates. Fisher, J. D., & Fisher, W. A. (1992). Changing AIDSrisk behavior. Psychological Bulletin, 111, 455–471. Fisher, J. D., Fisher, W. A., Bryan, A. D., & Misovich, S. J. (2002). Information-motivation-behavioral skills model-based HIV risk behavior change intervention for inner-city high school youth. Health Psychology, 21, 177–186. https://doi.org/ 10.1037/0278-6133.21.2.177 Flay, B. (1986). Efficacy and effectiveness trials (and other phases of research) in the development of health promotion programs. Preventive Medicine, 15, 451–474. https://doi.org/10.1016/0091-7435 (86)90024-1 Gier, J. (2017). Missed appointments cost the U.S. healthcare system $150B each year. Healthcare Innovation, April 26. www.hcinnovationgroup .com/clinical-it/article/13008175/missed-appoint ments-cost-the-us-healthcare-system-150b-eachyear Glanz, K., Sallis, J. F., Saelens, B. E., & Frank, L. D. (2005). Healthy nutrition environments: Concepts and measures. American Journal of Health Promotion, 19, 330–333. https://doi.org/10.4278/ 0890-1171-19.5.330 Glasgow, R. E., Vogt, T. M., & Boles, S. M. (1999). Evaluating the public health impact of health promotion interventions: The RE-AIM framework. American Journal of Public Health, 89, 1322–7. https://doi.org/10.2105/ajph.89.9.1322 Gollwitzer, P. M., & Sheeran, P. (2006). Implementation intentions and goal achievement: A meta-analysis of effects and processes. Advances in Experimental Social Psychology, 38, 69–121. doi.org/10.1016/ S0065-2601(06)38002-1 Grimshaw, J. M., Thomas, R. E., MacLennan, G. et al. (2004). Effectiveness and efficiency of guideline dissemination and implementation strategies. Health Technology Assessment, 8, 6. https://doi .org/10.3310/hta8060

282

CHARLES ABRAHAM AND SARAH DENFORD

Hagger, M. S., Moyers, S., McAnally, K., & McKinley, L. E., (2020). Known knowns and known unknowns on behavior change interventions and mechanisms of action. Health Psychology Review, 14, 199–212. https://doi.org/10.1080/ 17437199.2020.1719184 Hillier-Brown, F. C., Bambra, C. L., Cairns, J., Kasim, A., Moore, H. J., & Summerbell, C. D. (2014). A systematic review of the effectiveness of individual, community and societal level interventions at reducing socioeconomic inequalities in obesity amongst children. BMC Public Health 14, 834 https://doi.org/10.1186/ 1471-2458-14-834 Hillier-Brown, F. C., Summerbell, C. D., Moore, H. J. et al. (2017). A description of interventions promoting healthier ready-to-eat meals (to eat in, to take away, or to be delivered) sold by specific food outlets in England: a systematic mapping and evidence synthesis. BMC Public Health, 17, 93. https://doi.org/10.1186/s12889-016-3980-2 House of Lords. (2011). Science and Technology Committee – Second Report: Behaviour Change (HL Paper No. 179). London: The Stationery Office. https://publications.parlia ment.uk/pa/ld201012/ldselect/ldsctech/179/ 17902.htm Kaplan-Lewis, E., & Percac-Lima, S. (2013). No-show to primary care appointments: Why patients do not come. Journal of Primary Care and Community Health, 4, 251–255. https://doi.org/10.1177/ 2150131913498513 Kelly, M. P., & Barker, M. (2016). Why is changing health-related behaviour so difficult? Public Health, 136, 109–116. https://doi.org/10.1016/j .puhe.2016.03.030 Kessler, D. A. (2009). The End of Overeating: Taking Control of the Insatiable American Appetite. New York: Macmillan. Kurz, T., Gardner, B., Verplanken, B., & Abraham C. (2014). Habitual behaviours or patterns of practice? Explaining and changing repetitive climate-relevant actions. Wiley Interdisciplinary Reviews: Climate Change, 6, 113–128. https://doi .org/10.1002/wcc.327 Jepson, R. G., Harris, F. M., Platt, S., & Tannahill, C. (2010). The effectiveness of interventions to

change six health behaviours: A review of reviews. BMC public Health, 10, 538. Johnson, M. J., & May, C. R. (2015). Promoting professional behaviour change in healthcare: What interventions work, and why? A theory-led overview of systematic reviews. BMJ Open, 5, e008592. https://doi.org/10.1136/bmjopen-2015-008592 Lake, A. J., Browne, J. L., Abraham, C. et al. (2018). A tailored intervention to promote uptake of retinal screening among young adults with type 2 diabetes: An intervention mapping approach. BMC Health Services Research. 18, 396. https://doi.org/10.1186/s12913-0183188-5 Lobstein, T., & Davies, S. (2008). Defining and labelling “healthy” and “unhealthy” food Public Health Nutrition: 12, 331–340. https://doi.org/ 10.1017/S1368980008002541 Luszczynska, A., Sobczyk, A., & Abraham, C. (2007). Planning to lose weight: RCT of an implementation intention prompt to enhance weight reduction among overweight and obese women. Health Psychology, 26, 507–512. https://doi.org/10.1037/0278-6133.26.4.507 McGuire, J. F., Piacentini, J., Brennan, E. A. et al. (2014). A meta-analysis of behavior therapy for Tourette Syndrome. Journal of Psychiatric Research, 50, 106–112. https://doi.org/10.1016/j .jpsychires.2013.12.009 McLaughlin, J. A., & Jordan, G. B. (1999). Logic models a tool for telling your program’s performance story. Evaluation and Program Planning, 11, 65–72. https://doi.org/10.1016/ s0149-7189(98)00042-1 McQueenie, R., Ellis, D. A., McConnachie, A., Wilson, P., & Williamson A. E. (2019). Morbidity, mortality and missed appointments in healthcare: A national retrospective data linkage study. BMC Medicine, 17, 2. https://doi.org/10.1186/s12916018-1234-0 Milstein, B., Wetterhall, S., & CDC Evaluation Working Group. (2000). A framework featuring steps and standards for program evaluation. Health Promotion Practice, 1, 221–228. https:// doi.org/10.1177/152483990000100304 Molfenter, T. (2013). Reducing appointment no-shows: Going from theory to practice. Substance Use and

Design, Implementation, and Evaluation of Behavior Change Interventions

Misuse, 48, 743–749. https://doi.org/10.3109/ 10826084.2013.787098 Moore, G., Audrey, S., Barker, M. et al. (2014). Process Evaluation of Complex Interventions: Medical Research Council Guidance. London: MRC Population Health Science Research Network. https://doi.org/10.1136/bmj.h1258 Municipal Association of Victoria. (2019). Shade Design for Public Places: Selecting Appropriate, Innovative and Cost Effective Shade Measures. Melbourne: Municipal Association of Victoria. www.mav.asn.au/__data/assets/pdf_file/0018/ 7326/Shade-design-for-public-places.pdf National Highway Traffic Safety Administration. (2018). Summary of Motor Vehicle Crashes: 2006 Data. (Traffic Safety Facts. Report No. DOT HS 812 580). Washington, DC: National Highway Traffic Safety Administration. https://crashstats .nhtsa.dot.gov/Api/Public/ViewPublication/ 812580 National Institutes of Health. (2017). HIH Clinical research trials and you. National Institutes of Health. Website. www.nih.gov/health-informa tion/nih-clinical-research-trials-you/basics Nielsen, K., Randall, R., Holten, A., & González, E. R. (2010). Conducting organizational-level occupational health interventions: What works? Work and Stress, 24, 234–259. https://doi.org/ 10.1080/02678373.2010.515393 O’Loughlin, J. L., Paradis, G., Gray-Donald, K., & Renaud, L. (1999). The impact of a communitybased heart disease prevention program in a lowincome, inner-city neighborhood. American Journal of Public Health, 89, 1819–1826. http:// doi.org/10.2105/AJPH.89.12.1819 Pearson, M., Chilton, R., Wyatt, K., Abraham, C., Ford, T., & Anderson, R. (2015). Implementing health promotion programmes in schools: A realist systematic review of research and experience in the United Kingdom. Implementation Science, 10. https://doi.org/10.1186/s13012-015-0338-6 Peters, G.-J. Y., de Bruin, M., & Crutzen, R. (2015). Everything should be as simple as possible, but no simpler: Towards a protocol for accumulating evidence regarding the active content of health behaviour change interventions. Health

283

Psychology Review, 9, 1–14. https://doi.org/ 10.1080/17437199.2013.848409 Powers, W. T. (1973). Behavior: The Control of Perception. Chicago: Adline Publishing. Rodda, S., Merkouris, S., Abraham, C., Hodgins, D. C., Cowlishaw, S., & Dowling, N. (2018). Therapistdelivered and self-help interventions for gambling problems: A review of contents. Journal of Behavioral Addictions, 7, 211–226. https://doi .org/10.1556/2006.7.2018.44 Rogers, R. W. (1975). A protection motivation theory of fear appeals and attitude change. The Journal of Psychology, 91, 93–114. https://doi.org/abs/ 10.1080/00223980.1975.9915803 Ruiter, R. A. C., Abraham, C., & Kok, G. (2001). Scary warnings and rational precautions: A review of the psychology of fear appeals. Psychology and Health, 16, 613–630. https://doi.org/10.1080/ 08870440108405863 Sacerdote, C., Fiorini, L., Rosato, R., Audenino, M., Valpreda, M., & Vineis, P. (2005). Randomized controlled trial: Effect of nutritional counselling in general practice. International Journal of Epidemiology, 35, 409–415. https://doi.org/ 10.1093/ije/dyi170 Sheeran, P., Abraham, C., Jones, K. et al. (2019). Promoting physical activity among cancer survivors: Meta-analysis and meta-CART analysis of randomized controlled trials. Health Psychology, 38, 467. Sheeran, P., Aubrey, R., & Kellett, S. (2007). Increasing attendance for psychotherapy: Implementation intentions and the self-regulation of attendancerelated negative affect. Journal of Consulting and Clinical Psychology, 75, 853–863. https://doi.org/ 10.1037/0022-006X.75.6.853 Sheeran, P., & Orbell, S. (2000). Using implementation intentions to increase attendance for cervical cancer screening. Health Psychology, 19, 283–289. https://doi.org/10.1037//02786133.19.3.283 Strack, F., & Deutsch, R. (2004). Reflective and impulsive determinants of social behavior. Personality and Social Psychology Review, 8, 220–247. https://doi.org/10.1207/s15327957 pspr0803_1

284

CHARLES ABRAHAM AND SARAH DENFORD

Suls, J., Mogavero, J. N., Falzon, L., Pescatello, L. S., Hennessy, E. A., & Davidson, K. W. (2020). Health behaviour change in cardiovascular disease prevention and management: Meta-review of behavior change techniques to affect selfregulation. Health Psychology Review, 14, 43–65. https://doi.org/10.1080/17437199.2019.1691622 Thaler, R., & Sunstein, C. R. (2008). Nudge: Improving Decisions about Health, Wealth, and Happiness. New Haven, CT: Yale University Press. van Beurden, S. B., Greaves, C. J., Smith, J. R., & Abraham, C. (2016). Techniques for modifying impulsive processes associated with unhealthy eating: A systematic review. Health Psychology, 35, 793–806. https://doi.org/10.1037/hea0000337 VicHealth (2018). Healthy Food and Drink Choices in Community Sport: Building on Success. www .vichealth.vic.gov.au/media-and-resources/publi cations/healthy-choice-food Weinstein, N. D., Sandman, P. M., & Roberts, N. E. (1991). Perceived susceptibility and selfprotective behavior: A field experiment to encourage home radon testing. Health Psychology, 10, 25–33. https://doi.org/10.1037/ 0278-6133.10.1.25

WHO (World Health Organization). (2016). Consideration of the Evidence on Childhood Obesity for the Commission on Ending Childhood Obesity: Report of the Ad Hoc Working Group on Science and Evidence for Ending Childhood Obesity. Geneva: WHO. Witte, K., & Allen, M. (2000). A meta-analysis of fear appeals: Implications for effective public health campaigns. Health Education and Behavior, 27, 591–615. https://doi.org/10.1177/1090198100 02700506 W. K. Kellogg Foundation. (2004). Using Logic Models to Bring Together Planning, Evaluation, and Action: Logic Model Development Guide. Michigan: W. K. Kellogg Foundation. Wyatt, K., Lloyd, J., Creanor, S. et al. (2017). Cluster randomised controlled trial, economic and process evaluation to determine the effectiveness and cost effectiveness of a novel intervention (Healthy Lifestyles Programme, HeLP) to prevent obesity in school children. NIHR Public Health Research, 6, 1. www.ncbi .nlm.nih.gov/pubmed/29356471

20 Moving from Theoretical Principles to Intervention Strategies: Applying the Experimental Medicine Approach Alexander J. Rothman, William M. P. Klein, and Paschal Sheeran

Practical Summary What are effective and efficient ways to change the behavior patterns that underlie the major health, environmental, and social threats found across the globe? Too often, investigators can specify the behavioral goals to be obtained but struggle to identify which among an array of strategies should be used to pursue those goals. Research is needed that can provide investigators with evidence-based guidelines that specify not only whether an intervention works but also how and under what conditions. This chapter describes how the experimental medicine approach can be used to design and test intervention strategies in a manner that will enrich our understanding of how intervention strategies work and the conditions under which they operate effectively. Through the systematic use of this approach, evidence will emerge that addresses practitioners’ prevailing concerns directly – what intervention strategy is the most effective and efficient way to address the problem at hand.

20.1 Introduction: Advancing Behavior Change by Connecting Theoretical Principles to Intervention Design Even a cursory examination of the major health, environmental, and social threats that can be found across the globe reveals the crucial role played by human behavior. Rates of disease and life expectancy are tied to behavioral factors such as poor diet, high rates of inactivity, the use of tobacco, and insufficient access to or use of health care (Adams et al., 2019; Bauer et al., 2014). Environmental damage is similarly tied to patterns of human behavior, in this case rates of renewable and nonrenewable energy sources, modes of transportation, water usage, and recycling, to name but a few (Gifford, Kormos, & McIntyre, 2011; Swim, Clayton, &

Howard, 2011). Thus, to address these challenges, there is little debate about what needs to change – human behavior. Yet how to effectively and efficiently promote and support these changes remains a vexing challenge. More often than not, investigators can articulate the behavioral goals to be obtained (e.g., increase rates of measles vaccination) but are less able to delineate the strategies that should be used to pursue those goals. For example, having identified the need to increase rates of measles vaccination, investigators remain unsure whether to design and implement media campaigns that highlight the dangers of measles, bans that limit access to services such as public school education without vaccination, or health care delivery systems that make it easier for people to obtain a vaccine (see https://doi.org/10.1017/9781108677318.020

286

ALEXANDER J. ROTHMAN, WILLIAM M. P. KLEIN, AND PASCHAL SHEERAN

Sidebar 20.1 Connecting principles and practice: the case of vaccination behavior

The development of vaccinations for infectious disease is one of the leading public health achievements of the past century (CDC, 1999). Yet the benefits afforded by vaccination rest on people’s behavior. Moreover, communities of individuals that rely on the benefits of herd immunity reveal the degree to which people’s health depends on the behavioral decisions of others. The increase in measles cases in the United States in 2019 illustrates the impact of localized drops in vaccination behavior (Patel et al., 2019). Thus, there is a growing need for evidence-based behavior change strategies that can be used to address the challenges posed by insufficient or delayed vaccine uptake. In a comprehensive review of the psychological principles that can be leveraged to improve vaccination behavior, Brewer and colleagues (2017) have identified three broad classes of approaches for investigators to consider. The first approach focuses on strategies that strive to modify people’s vaccine-relevant beliefs (e.g., perceptions of disease risk; vaccine effectiveness). The second approach focuses on strategies designed to create or highlight social norms regarding vaccinations. The third approach focuses on strategies designed to modify people’s vaccination behavior directly by either helping people act on their good intentions or mandating action through regulations. Although there is a rich set of principles that can guide the development of intervention strategies, a more robust evidence base is needed that can enable investigators to determine the most effective and efficient strategy to use and the specific conditions (e.g., vaccination type; population) under which to use them.

Sidebar 20.1). Given an array of potential strategies, investigators may wonder whether they differ in the ease of implementation, whether they produce similar outcomes, and whether there are conditions under which each strategy is more or less effective. Questions such as these are quite common but unfortunately answers are elusive. To address this challenge, there is a need to approach the design and evaluation of behavioral intervention strategies in a manner that affords the creation of evidence-based guidelines that specify not only how interventions work but also the conditions under which they can effectively and efficiently help people initiate and maintain new patterns of behavior. In this chapter, the design and testing of interventions is situated within the experimental medicine approach (Bernard, 1957; Nielsen et al., 2018; Sheeran, Klein, & Rothman, 2017). As indicated by its title, the experimental medicine approach leverages the strength of the

experimental method to specify and test how intervention strategies work and to identify the conditions under which they operate effectively. Moreover, it organizes how investigators specify the questions that underlie the study of behavior change interventions. In particular, it requires investigators to articulate precisely what intervention strategy they are using, how they think the strategy operates, and the outcomes it generates. Through the systematic use of this approach, evidence will emerge that addresses practitioners’ prevailing concerns directly – what intervention strategy is the most effective and efficient way to address the problem at hand. This chapter provides an overview of how to implement the experimental medicine approach, describes its key features, and offers an illustrative example of how it can be used. Next, the chapter addresses why investigators need to be precise in how they describe and operationalize the outcomes they prioritize, the

Moving from Theoretical Principles to Intervention Strategies

constructs they hope to change, and the intervention strategies they propose to use. Finally, the experimental medicine approach is considered within a broader set of initiatives that have emerged to support a programmatic approach to the design, evaluation, and implementation of behavior change interventions.

20.2 The Experimental Medicine Approach to Intervention Design and Testing Research on behavior change is typically motivated by the identification of a practical problem – the need to reduce people’s home energy consumption, to increase medication adherence, to promote civic engagement – and the desire to evaluate an intervention strategy designed to address it. Traditionally, this approach has led investigators to conduct efficacy trials that focus primarily on determining whether a strategy can elicit changes in the targeted outcome, with limited attention to the processes that might underlie its effectiveness. For example, residents in a community might be randomly assigned to a new treatment program in which they receive a $20 incentive payment each month if their energy consumption falls below a set threshold or a control condition in which they receive informational tips each month about how to reduce their energy consumption below a set threshold. Home energy use might then be monitored over the next twelve months. What might be gleaned from this trial? In the bestcase scenario, the incentive-based program is shown to be more effective. This finding is useful but, on its own, leaves investigators with a limited understanding of why the program worked or the conditions under which it is most likely to work. In the absence of this information, subsequent decisions about how to disseminate this program and what features are critical to its success are difficult. Must the incentive payment be $20 or could it be set at a lower value and remain

287

effective? Are there contexts in which it should be raised? Answers to questions such as these require an understanding of how incentives affect people’s energy usage. What if the trial reveals that the incentive-based program was no more effective than the control treatment? Now, even more questions emerge – ones similarly predicated on understanding how incentives operate. The incentive was put in place because it was thought to alter how people think about their energy use. Perhaps the incentive provided by the program was ineffective because it was insufficient to change people’s beliefs about their energy use. If so, modifying the program to provide a larger incentive might render it effective. Alternatively, it could be that the incentive program changed people’s beliefs about their energy use but these thoughts were unrelated to their behavioral decisions. If this is this case, the conclusion might be that incentives are not an appropriate strategy for this behavioral domain. More often than not, investigators grapple with whether to modify or abandon a particular behavior change strategy and do so without the evidence needed to make informed decisions. Given this state of affairs, there is a need for approaches that encourage investigators to focus on the processes that underlie the design and evaluation of intervention strategies and, thus, generate an evidence base that can address these key questions (Collins, 2018; Czajkowski et al., 2015; Nielsen et al., 2018; Sheeran et al., 2017). The experimental medicine approach provides investigators with a structure for building a program of research that can generate the evidence base needed to make informed decisions regarding the design and dissemination of effective behavioral interventions. Experimental medicine is comprised of four key steps (see Figure 20.1), which are introduced here and then discussed more fully in the next sections of the chapter. Before starting the experimental medicine approach, investigators must specify the primary outcome of interest – this could be a single

288

ALEXANDER J. ROTHMAN, WILLIAM M. P. KLEIN, AND PASCHAL SHEERAN

Target Identification (Path A)

Target (Mechanism of Action)

l de

o M of D) t es th lT a ul (P

Ta

rg e (P t Va at li h da B ti ) on

ll Te (P st at of h M D od ) e

l

Ta

rg

et E (P ng at ag h C) eme nt

F

Fu

Intervention Strategy

Outcome (Behavior)

Figure 20.1 Mapping the paths within the experimental medicine approach

behavior (e.g., measles vaccination), a pattern of behavior (e.g., regular physical activity), or an outcome afforded by the behavior (e.g., weight). With the primary outcome identified, the first step of the experimental medicine approach involves the identification of factors that are related to the outcome of interest and are potentially modifiable (path A, Figure 20.1). These two conditions indicate that the identified factor could serve as the mechanism through which the intervention works (i.e., a mechanism of action) and, thus, qualifies as a potential target for an intervention strategy. The second step involves the validation of the potential target by assessing when and to what degree variation in the identified target reliably elicits change in the outcome of interest (path B, Figure 20.1). Evidence of validation provides the basis for moving to the third step, engagement, which involves testing the ability of different intervention strategies to engage the identified target and selecting a strategy that can elicit sufficient change in the target (path C, Figure 20.1). This step is critical as a target can operate as an explanatory mechanism only if investigators can identify a strategy that

can engage it. Having specified a target (or a set of targets) that can be measured and an intervention strategy (or a set of intervention strategies) that can engage the target, investigators can then turn to conducting a full test of the proposed model to determine whether the intervention strategy (or set of strategies) can not only elicit a meaningful change in the primary outcome of interest but do so through its ability to effect change in the specified target(s) (path D, Figure 20.1). Causal chain analysis (Spencer, Zanna, & Fong, 2005) and statistical tests of mediation (MacKinnon, 2008) can be used to quantify the magnitude of the underlying process. Through its comprehensive focus on specifying the underlying linkages between intervention strategies, targets (i.e., mechanisms of action), and outcomes, the experimental medicine approach is consistent with a range of approaches that have emerged that encourage investigators to dedicate time and effort to thinking through the linkages between intervention strategies and their hypothesized mechanisms of action (e.g., science of behavior change [SOBC], Nielsen et al., 2018; theories

Moving from Theoretical Principles to Intervention Strategies

and techniques of behavior change [TAT], Michie et al., 2018) and to developing strategies that can elicit the necessary change in the mechanisms of action (e.g., obesity-related behavioral intervention trials [ORBIT], Czajkowski et al., 2015; multiphase optimization strategy [MOST], Collins, 2018). A key principle that underlies all of these approaches is the importance of precision. Specifically, the strength of the evidence base that emerges from the experimental medicine approach depends on investigators articulating precisely the purpose for which they are intervening (e.g., what is the primary outcome?), the intervention strategy that are using (e.g., what behavior change technique?), the reason they have selected that strategy (e.g., what mechanism of action is targeted?), and the conditions under which they are pursuing these goals (e.g., for whom, when, and/or where is this intervention being tested?). Through its emphasis on providing answers to these questions, the experimental medicine approach offers investigators the opportunity to generate an evidence base that is not only robust but also broadly applicable, as insights regarding the effect of an intervention strategy on one behavior are structured in a manner that can inform decisions regarding its applicability to other behaviors.

20.2.1 Situating the Experimental Medicine Approach Within a Practical Problem The experimental medicine approach provides investigators with a structure to guide the development and testing of a behavior change strategy. To illustrate this, consider the example introduced in Section 20.1, the emerging need to develop and test strategies to promote adherence to measles vaccination guidelines in the United States. The Centers for Disease Control (CDC) recommend that parents have their children receive the MMR vaccine (which provides protection against measles, mumps, and rubella) in a series of two shots – one between the ages of twelve and fifteen months and

289

the second between the ages of four and six years. Although the incidence of measles dropped dramatically with the introduction of the vaccine in the 1960s, over the past several years there has been a spike in measles cases that is paralleled by a small but growing drop in vaccination rates (CDC, 2019). Thus, adherence to the recommended vaccine schedule is the primary outcome of interest. At the same time the primary outcome is identified, it is also important to determine the population of interest and, when relevant, the broader social and structural conditions in which they are situated. Given the present example, one might choose to focus on whom in the family serves as the primary caregiver. Moreover, one would need to decide whether to focus on adults with currently unvaccinated children or those who are expecting their first child. Depending on the interests of the investigators, consideration might be given to other factors such as cultural background, socioeconomic status, geography, insurance status, to name but a few. These decisions set the context within which the investigators implement the experimental medicine approach. There is an array of social, psychological, and structural factors that affect vaccination rates (see Brewer et al., 2017) but, for the purposes of this example, the focus will be on people’s beliefs about measles and, specifically, the premise that with the near eradication of measles – and the associated absence of disease-related morbidity and mortality – people underestimate the threat posed by measles. Thus, investigators might propose that the perceived threat of measles is the target. Given evidence that perceptions of threat can be manipulated experimentally, and that doing so can lead to increases in precautionary behavior (Sheeran, Harris, & Epton, 2014), investigators can conclude that these perceptions are modifiable and a potentially viable target. However, as a first step, investigators would want to confirm through either their own work or evidence in the literature that there is a meaningful relation between the perceived threat of measles and vaccination

290

ALEXANDER J. ROTHMAN, WILLIAM M. P. KLEIN, AND PASCHAL SHEERAN

behavior. With this evidence in hand, the next step would be to validate the target. Validation focuses on providing experimental evidence that manipulating the perceived threat of measles (e.g., by showing people a brief video illustrating the health consequence associated with contracting measles) increases adherence with vaccine guidelines. This will typically involve smaller-scale experiments that focus on demonstrating proof of concept and provide investigators with assurance that the core proposition regarding the relation between perceived threat and behavior is viable. This approach is similar to the early stages of intervention development identified in the ORBIT model (Czajkowski et al., 2015). Having demonstrated that experimentally manipulating perceived threat can enhance adherence, investigators turn to target engagement. Target engagement challenges investigators to refrain from merely using the intervention strategy that is readily available and instead consider the relative merits of different strategies. Although the strategy used to validate threat as a target might prove to be the optimal choice, investigators are unlikely to know this a priori. Thus, target engagement involves identifying a set of potential strategies (e.g., video presentations from health professionals; personal testimonials from people who have had measles or from parents whose children have had measles) and systematically comparing their ability to increase the perceived threat of measles. A key aspect of this step is that investigators need to specify the criteria by which they will evaluate the effectiveness of these strategies and, thus, articulate what the intervention strategy needs to accomplish in order to be effective. For example, investigators might prioritize the magnitude of the change elicited by the strategy (e.g., perceived threat must be elevated above a given threshold), the durability of the observed change (e.g., the change in perceived threat is maintained over a specified time period), and/or the resources required to elicit the observed change (e.g., the time and

effort needed to deliver the strategy need to be below a specified threshold). On completion, the tests of target engagement will enable investigators to make an active choice regarding the most effective, and perhaps efficient, way to manipulate the perceived threat of measles. Given preliminary evidence that manipulating the perceived threat of measles can increase adherence and the selection of an intervention strategy that has been shown to engage perceived threat effectively, investigators are now in a position to undertake a full scale, well-powered randomized controlled trial (RCT) that can provide an informative test of the theoretical premise that motivated these efforts. Specifically, the RCT can assess not only the effect of the selected intervention strategy on adherence to measles vaccination guidelines but also whether this effect can be attributed to increased perceptions of threat. Although completing the steps outlined in the experimental medicine approach should provide investigators with cautious optimism regarding the effectiveness of their intervention, it remains, in the end, an empirical question. However, regardless of the result, investigators will find they can contribute to our understanding of both the intervention’s impact and the processes hypothesized to underlie it.

20.3 Implementing the Experimental Medicine Approach: The Importance of Specificity The experimental medicine approach provides investigators with a structured series of steps to guide the development and testing of a behavior change intervention. However, to successfully navigate through these steps investigators need to be precise in their specification of the key constituent elements of the experimental medicine approach: the primary behavioral outcome, the target (i.e., mechanism of action), and the intervention strategy to be implemented. Specifying each of these elements not only

Moving from Theoretical Principles to Intervention Strategies

structures the decisions investigators must make but also facilitates their ability to communicate clearly what they have done and found. The sections that follow outline the issues investigators should be mindful of as they strive to specify each of these elements and how being precise provides valuable structure to their decisions.

20.3.1 Specifying the Behavior: What Is the Primary Outcome? More often than not investigators know what they want to accomplish – they want to increase people’s physical activity, promote the use of renewable energy sources, maximize savings for retirement, to name a few outcomes. Yet the transition from a general goal to a specific outcome of interest can be a challenge. Investigators need to articulate what is the particular behavioral outcome they want to elicit, as it guides deliberations about potential intervention targets and what behavioral outcome needs to be assessed. Is it the performance of a single behavior (e.g., obtaining a flu shot in 2019)? Is it a behavior within a specified time window (e.g., obtaining a flu shot between October 1 and November 30, 2019)? Or is it a pattern of behavior (e.g., obtaining a flu shot each year over a specified time period)? The greater the precision with which the outcome is specified enables investigators to be more precise in the logic model that underlies their intervention approach (see also Chapter 19, this volume). The factors that enable a person to perform a behavior once may be meaningfully distinct from those that enable someone to perform a behavior repeatedly over time (Kwasnicka et al., 2016; Rothman et al., 2011). Thus, the targets that are identified and the intervention strategies needed to engage those targets may differ depending on the nature of the outcome. Greater precision regarding the outcome of interest may also help investigators specify and test more precise predictions, which, in turn, will afford more informative results (cf. Meehl, 1978).

291

One challenge for investigators is that there is not yet an organizing typology to structure how people conceptualize behavior. Within the health domain, discussions have emerged regarding an array of potentially useful distinctions between types or classes of behavior. For instance, does the behavior need to occur only once or is it repeated? And, if it is repeated, on what time scale? Patterns of regular physical activity are comprised of repeated behavior across days, whereas screening behaviors might repeat but at much longer time intervals (e.g., once every year). One broad distinction that has received attention is whether the behavior in question needs to be changed or maintained (Rothman et al., 2011). Moreover, in contexts that involve changing behavior, finer distinctions can be made between new behaviors that are initiated (e.g., starting a new treatment), behaviors that need to be increased (e.g., eating more servings of fruit and vegetables), behaviors that need to be reduced (e.g., spending less time sedentary each day), and behaviors that need to be eliminated (e.g., stopping smoking) (Sheeran et al., 2019). Similarly, in contexts that involve maintaining behaviors, distinctions can be made between patterns of behaviors that continue to require active deliberation and those that have become more automatic and are considered habits (Gardner, 2015; see Chapters 13 and 41, this volume). Although these distinctions are conceptually meaningful and of interest to investigators, the implications they have for intervention efforts are not yet fully formulated.

20.3.2 Specifying the Appropriate Target: What Is the Mechanism of Action? The target or mechanism of action sits at the heart of the experimental medicine approach. It is the construct that is hypothesized to guide changes in behavior and, in turn, what the selected intervention strategy will try to engage. Thus, the care and precision with which the investigator specifies the

292

ALEXANDER J. ROTHMAN, WILLIAM M. P. KLEIN, AND PASCHAL SHEERAN

target of interest is critical. To help investigators navigate through these issues, initiatives have emerged to provide guidance regarding best practices for identifying and measuring constructs. For example, SOBC has supported efforts to not only develop evidence-based measures for constructs that underlie behavior change but also provide an accessible repository of measures (Nielsen et al., 2018). Complementary sources of guidance regarding measures include the patient-reported outcomes measurement information system (PROMIS) (Carle et al., 2015), the phenotype measurement (PhenX) project (Hamilton et al., 2011), and the accumulating data to optimally predict obesity treatment (ADOPT) core measures project (MacLean et al., 2018). Because a key aspect of the experimental medicine approach is identifying a specific target, investigators need to be able to both specify the construct of interest and delineate how it is distinct from other potential targets. The growing availability of rigorous, valid measures is an important step in this effort as it provides investigators with the tools to obtain preliminary evidence for a hypothesized target (see Sidebar 20.2). To date, evidence regarding the validity of a potential target primarily rests on findings from observational studies that track the correlation between measures of the target and measures of the primary outcome, either at the same or at a later point in time. Although the research literature is replete with evidence syntheses revealing robust relations between constructs and behavior (e.g., Brewer et al., 2007; McEachan et al., 2011), investigators need to be mindful of the limitations placed on inferences that can be drawn from correlational data (Weinstein, 2007). In particular, most reported associations between targets and outcomes are bivariate and do not take into account the influence of alternative targets (e.g., do perceptions of risk account for variability in vaccination behavior after controlling for the effect of perceptions of worry?). Thus,

experimental tests of target validity are key sources of evidence. Fortunately, evidence syntheses within the health domain have demonstrated that experimentally modifying constructs such as perceived severity, selfefficacy, social norms, attitudes, and intentions does lead to meaningful changes in behavior (e.g., Sheeran et al., 2014; Sheeran et al., 2016; Webb & Sheeran, 2006). Findings such as these are encouraging but, more often than not, investigators will find they are focusing on a hypothesized link between a target and an outcome that has yet to be examined experimentally. In this case, an initial experimental test should be conducted to affirm that manipulating the target of interest can lead to a change in the outcome. Although the discussion regarding potential targets has focused on the identification of a single target, investigators may find themselves pursuing research questions that involve multiple targets. For example, investigators might hypothesize that people’s willingness to engage in a precautionary behavior is largely a function of their perceptions of personal risk and confidence in their ability to behaviorally address the risk (i.e., self-efficacy; Tannenbaum et al., 2015). In cases such as these, investigators need to think carefully about whether they expect the targets – personal risk and self-efficacy – to affect behavior independently or jointly, as this will shape the evidence needed to validate the targets and what needs to be done to obtain that evidence. In particular, investigators must be able to measure each construct reliably and to identify distinct strategies that can manipulate people’s standing on each construct.

20.3.3 Specifying the Appropriate Intervention Strategy: How to Engage the Target? Behavioral scientists have spent decades theorizing about constructs and their interplay with behavior.

Moving from Theoretical Principles to Intervention Strategies

293

Sidebar 20.2 Resources for making informed decisions about constructs and measures

Investigators must navigate through key decisions about which specific constructs to target and, once identified, how best to measure them. Over the past decade, there has been a dramatic increase in the resources available to help support investigators. Several initiatives have focused on identifying constructs that are potential mechanisms of action (i.e., targets). For example, the science of behavior change initiative (Nielsen et al., 2018) has developed an online repository of potential targets along with information about measurement approaches and evidence for target engagement.1 Complementary information regarding hypothesized linkages between potential mechanisms of action and intervention strategies can be obtained from the theory and techniques tool,2 which is based on a triangulation of evidence from expert judgments and literature reviews (Johnston et al., 2019). Several initiatives have worked to improve the quality of measures available for key constructs and to facilitate the use of these higher-quality measures. For example, PROMIS (Carle et al., 2015)3 and the Phenotype Measurement (PhenX) project (Hamilton et al., 2011)4 provide online archives of reliable, well-validated behavioral measures (e.g., depression, anxiety, pain) reviewed and recommended by panels of experts. The greater accessibility of these measures reduces the likelihood that investigators will develop de novo measures for a given construct that may not have strong psychometric properties and, thus, be unable to detect potential effects of an intervention. Finally, resources are also emerging that focus on recommendations regarding constructs and measures within specific behavioral domains. The ADOPT core measures project (MacLean et al., 2018) focuses on recommendations regarding weight-loss interventions for adults. Across each of four domains (behavioral, biological, environmental, and psychosocial), ADOPT provides investigators with expert-based recommendations regarding which constructs to focus on and recommendations regarding best practices for measurement.5

On the other hand, the design of interventions has been construed as more of a practical than a theoretical concern. One consequence of this approach is that descriptions of interventions have focused on operational characteristics (e.g., duration, intensity, delivery mode, setting; Moher et al., 2001; see Chapter 21, this volume). Although the specific strategies embedded within the intervention are characterized, the level of description provided has traditionally been vague (e.g., counseling) and inconsistent across reports (Abraham & Michie, 2008; see Chapter 19, this volume). This lack of clarity reflected the absence of a shared language to

describe the different intervention strategies contained within an investigator’s toolbox, which impaired their ability to articulate both to themselves and to their colleagues what they had implemented.

1 2

3

4 5

See https://scienceofbehaviorchange.org/measures/. See https://theoryandtechniquetool.humanbehaviourch ange.org/. See www.healthmeasures.net/explore-measurement-sys tems/promis. See www.phenxtoolkit.org/. See www.gem-measures.org/workspaces/ADOPT.

294

ALEXANDER J. ROTHMAN, WILLIAM M. P. KLEIN, AND PASCHAL SHEERAN

To address this gap, formalized taxonomies of behavior change techniques have emerged, providing distinct labels and definitions for techniques that could be reliably applied to intervention efforts retrospectively and prospectively (Abraham & Michie, 2008; Kok et al., 2016; Michie et al., 2013). The prevailing behavior change technique taxonomy, version 1 (BCTTv1), developed by Michie and colleagues (2013), provides investigators with a set of ninety-three distinct behavior change techniques that are organized within nineteen clusters (e.g., goals and planning; feedback and monitoring; reward and threat). With the shared language provided by the taxonomy, investigators are able to communicate effectively and precisely about the intervention strategies they are using. This is particularly valuable for implementing the experimental medicine approach as it provides a structure for investigators to leverage as they work to identify the intervention strategy or technique that can effectively engage the proposed target. How do investigators specify the appropriate intervention strategy? Given the primary goal is to identify a strategy or set of strategies that can modify a specified target, investigators need theoretical and empirical guidance regarding the linkages between intervention strategies and targets. One source of guidance comes from reviews of prior interventions whose components have been identified and coded and that describe the relative strength of the association between the use of specific strategies and observed changes in a specified target (e.g., Gillison et al., 2018; Knittle et al., 2018). A second, complementary source of guidance comes from an effort to map hypothesized linkages between intervention techniques and mechanisms of action (i.e., targets). Based on a synthesis of evidence from a review of published interventions and expert judgment (Carey et 6

See https://theoryandtechniquetool.humanbehaviourch ange.org/.

al., 2019; Connell et al., 2019), an interactive database has been launched that enables investigators to examine the status of hypothesized links between techniques and targets.6 In both cases, the available evidence should be taken as generative rather than confirmatory regarding the potential link between techniques and targets. Although it offers a way to prioritize which intervention strategies to consider, it is essential that investigators obtain experimental evidence of target engagement – that implementing the specified strategy elicits a meaningful change in the target. Through its emphasis on target engagement, the experimental medicine approach encourages investigators to think strategically about not only what is needed to engage a proposed target but also what is the minimum acceptable magnitude of change in the target. Too often, investigators select an intervention strategy that, at best, has been shown to elicit some change in the target. It might be a strategy they or a colleague used previously. Little effort has been directed toward delineating the most effective and efficient way to elicit change in targets that have been shown to affect changes in behavior (e.g., intentions, perceived risk). Given that the degree of change elicited in a target is a key determinant of change in the primary outcome, the benefits afforded by optimizing target engagement are likely to be considerable. Conducting competitive tests of target engagement should become a research priority; these efforts would augment not only the viability of an investigator’s own intervention efforts but also an evidence base that would be of value to investigators broadly.

20.4 Putting the Pieces Together: Moving Forward with the Experimental Medicine Approach The experimental medicine approach provides investigators with a series of steps that afford a rigorous and robust test of a behavioral intervention

Moving from Theoretical Principles to Intervention Strategies

strategy. Although implementing this approach does not ensure that a proposed intervention strategy will prove effective, it does ensure that efforts undertaken will be informative regarding each facet of the intervention model. The tools and empirical evidence that emerge from efforts to identify potential targets, validate their association with a primary outcome, and demonstrate that strategies can effectively engage a target provide a foundation of knowledge that can inform future efforts both within a given research group and more broadly across the investigative community. Given the informational value of this work, investigators should be encouraged to embed their efforts within structures that support a more open and accessible scientific process (Munafò et al., 2017; Nosek et al., 2018). This might include, but is not limited to, preregistering the hypotheses and research questions, allowing measures and intervention materials to be publicly accessible, and making data and findings publicly accessible in a manner that is usable and time-sensitive. Taken together, these actions should facilitate a more productive, comprehensive evidence base that, in the hands of the research community, will stimulate advances in our ability to understand and promote behavior change. The experimental medicine approach is structured to focus investigators’ attention on the role of specific targets and the intervention strategies that can modify those targets. Yet behavior change interventions are typically comprised of multiple intervention strategies, each designed to engage with a particular target. How should investigators navigate the complexity afforded by this arrangement and, in particular, specify the contribution of each proposed strategy? The multiphase optimization strategy (MOST; Collins, 2018) provides an innovative framework designed to assess the relative contribution of distinct intervention strategies toward change of a specified outcome. Although a full discussion of this approach is beyond the scope of this chapter, a key step within MOST is the use of factorial experiments to develop an optimized version of an intervention (for a full

295

discussion of this approach, see Collins, 2018). In particular, investigators can use this step to assess the relative impact of intervention strategies designed to engage either the same target (generating evidence regarding which strategy is the most effective way to engage a target) or different targets (generating evidence regarding which target pathway is the most effective way to modify behavior). There is a growing recognition that greater attention needs to be paid to the considerable unexplained heterogeneity in intervention effects across both studies (e.g., Jachimowicz et al., 2019) and participants within an intervention (e.g., MacLean et al., 2018). The experimental medicine approach offers a useful framework for thinking systematically about the boundary conditions that regulate the effectiveness of an intervention strategy. Specifically, the observation that an intervention strategy is more (or less) effective for some people, specific behaviors, or in certain contexts can be unpacked into two distinct but complementary questions. First, investigators may examine whether the boundary condition regulates the link between the target and the primary outcome. If analyses reveal the impact of a specified target on an outcome is inconsistent across conditions, it would indicate the need to identify a new target (and associated intervention strategy). Second, investigators may examine whether the boundary condition regulates an intervention strategy’s ability to engage the target. If this proves to be the case, investigators can maintain their focus on the target but pursue an alternative strategy for engaging it. By specifying where the boundary condition disrupts the causal chain underlying the intervention, investigators can make more informed decisions regarding when a particular strategy should and should not be used. Moreover, with the accumulation of publicly available evidence across studies, investigators should be able to examine whether evidence for moderating effects is robust across a large set of samples, behaviors, and settings.

296

ALEXANDER J. ROTHMAN, WILLIAM M. P. KLEIN, AND PASCHAL SHEERAN

Finally, a possible broader limiting condition concerns the applicability of the experimental medicine approach to intervention design and testing conducted outside of behavioral medicine. The experimental medicine approach and, more generally, the series of initiatives that have emerged to improve intervention design and testing (e.g., SOBC, ORBIT, MOST) have all focused on addressing health outcomes. Can the experimental medicine approach be used to structure efforts concerned with changes in behavior in the environmental, civic, or educational domains? The overarching structure of the approach is designed to be broadly applicable. However, efforts to implement it within a given domain need to be mindful of three potential challenges. First, are there theoretical models and empirical data available to facilitate the identification of potential, modifiable targets? And are there reliable and valid measures to assess the identified targets? Within the health domain, the experimental medicine approach capitalizes on a long-standing focus on mechanistic models and measure development. If similar resources are not available within a domain, to what extent can the models and measures developed within health be applied? Second, is there a shared language to describe the intervention strategies utilized within the domain? If not, can taxonomies that have been utilized in the health domain be applied to the techniques used in a different domain? Third, is there a willingness to use experimental methodologies within the domain? Within the health domain, there is a long-standing tradition of using experimental methods to evaluate interventions. To the extent that this is not the case in other domains, implementing the experimental medicine approach may prove challenging. Although all of these issues deserve careful consideration, it is also the case that the benefits afforded by adopting the experimental medicine approach are broadly applicable across domains. There is good reason to be confident that, regardless of behavioral domain, the experimental medicine approach will help

investigators maximize the practical and theoretical knowledge that can come from efforts to design and test behavior change interventions.

20.5 Summary and Conclusion Given that human behavior is a critical determinant of the major health, environmental, and social threats found across the globe, there is a critical need to identify intervention strategies that elicit changes in people’s behavior effectively and efficiently. Moreover, investigators need evidencebased guidelines that can enable them to make informed decisions about which strategy to use under specific conditions. To date, the informational value of evidence that has emerged from efforts to develop and test behavioral interventions has been limited. The experimental medicine approach provides investigators with a structured series of steps that can guide a program of research on behavioral intervention strategies. Moreover, the design of the experimental medicine approach allows it to operate synergistically with both the theoretical and the practical principles specified in the chapters across this volume. Through the careful specification of the primary outcome, the mechanism of action targeted, and the intervention technique implemented, the experimental medicine approach will help investigators maximize the informational value of the evidence they generate, providing the broader investigative community with the guidance that they need when deciding which strategy to use to address the problem at hand.

References Abraham, C., & Michie, S. (2008). A taxonomy of behavior change techniques used in interventions. Health Psychology, 27, 379–387. https://doi.org/ 10.1037/0278-6133.27.3.379 Adams, M. L., Grandpre, J., Katz, D. L., & Shenson, D. (2019). The impact of key modifiable risk factors on chronic conditions. Preventive Medicine, 120, 113– 118. https://doi.org/10.1016/j.ypmed.2019.01.006

Moving from Theoretical Principles to Intervention Strategies

Bauer, U. E., Briss, P. A., Goodman, R. A., & Bowman, B. A. (2014). Prevention of chronic disease in the 21st century: Elimination of the leading preventable causes of premature death and disability in the USA. The Lancet, 284, 45–52. https://doi.org/10.1016/S0140-6736(14)60648-6 Bernard, C. (1957). An Introduction to the Study of Experimental Medicine. New York: Dover Publications. (Originally published in 1865.) Brewer, N. T., Chapman, G. B., Gibbons, F. X., Gerrard, M., McCaul, K. D., & Weinstein, N. D. (2007). Meta-analysis of the relationship between risk perception and health behavior: The example of vaccination. Health Psychology, 26, 136–145. https://doi.org/10.1037/0278-6133.26.2.136 Brewer, N. T., Chapman, G. B., Rothman, A. J., Leask, J., & Kempe, A. (2017). Understanding and increasing vaccination behaviors: Putting psychology into action. Psychological Science in the Public Interest, 18, 149–207. https://doi.org/ 10.1177/1529100618760521 Carey, R. N., Connell, L. E., Johnston, M. et al. (2019). Behaviour change techniques and mechanisms of action: A synthesis of links hypothesised in published intervention literature. Annals of Behavioral Medicine. https://doi.org/10.31234/ osf.io/x5372 Carle, A. C., Riley, W., Hays, R. D., & Cella, D. (2015). Confirmatory factor analysis of the Patient Reported Outcomes Measurement Information System (PROMIS) adult domain framework using item response theory scores. Medical Care, 53, 894–900. https://doi.org/10.1097/mlr.000000000 0000413 CDC (Centers for Disease Control and Prevention). (1999). Ten great public health achievements: United States, 1900–1999. MMWR. Morbidity and Mortality Weekly Report, 48, 241–243. CDC (Centers for Disease Control and Prevention). (2019). Measles Cases and Outbreaks: Measles Cases in 2019. Atlanta, GA: US Department of Health and Human Services. www.cdc.gov/ measles/cases-outbreaks.html Collins, L. M. (2018). Optimization of Behavioral, Biobehavioral, and Biomedical Interventions: The Multiphase Optimization Strategy (MOST).

297

New York: Springer. https://doi.org/10.1007/97 8-3-319-72206-1_1 Connell, L. E, Carey, R. N., de Bruin, M. et al. (2019). Links between behaviour change techniques and mechanisms of action: An expert consensus study. Annals of Behavioral Medicine. https://doi.org/ 10.31234/osf.io/fge86 Czajkowski, S. M., Powell, L. H., Adler, N. et al. (2015). From ideas to efficacy: The ORBIT model for developing behavioral treatments for chronic diseases. Health Psychology, 34, 971–982. https:// doi.org/10.1037/hea0000161 Gardner, B. (2015). A review and analysis of the use of “habit” in understanding, predicting, and influencing health-related behaviour. Health Psychology Review, 9, 277–295. https://doi.org/ 10.1080/17437199.2013.876238 Gifford, R., Kormos, C., & McIntyre, A. (2011). Behavioral dimensions of climate change: Drivers, responses, barriers, and interventions. WIREs Climate Change, 2, 801–827. https://doi .org/10.1002/wcc.143 Gillison, F. B., Rouse, P., Standage, M., Sebire, S. J., & Ryan, R. M. (2019). A meta-analysis of techniques to promote motivation for health behaviour change from a self-determination theory perspective. Health Psychology Review, 13, 110–130. http://doi .org/10.1080/17437199.2018.1534071 Hamilton, C. M., Strader, L. C., Pratt, J. G. et al. (2011). The PhenX Toolkit: Get the most from your measures. American Journal of Epidemiology, 174, 253–260. https://doi.org/10.1093/aje/kwr193 Jachimowicz, J. M., Duncan, S., Weber, E. U., & Johnson, E. J. (2019). When and why defaults influence decisions: A meta-analysis of default effects. Behavioural Public Policy, 3, 159–186. https://doi.org/10.1017/bpp.2018.43 Johnston, M., Carey, R. N., Connell Bohlen, L. et al. (2019). Development of an online tool for linking behavior change techniques and mechanisms of action based on triangulation of findings from literature synthesis and expert consensus. PsyArXiv. https://doi.org/10.31234/osf.io/ur6kz Knittle, K., Nurmi, J., Crutzen, R., Hankonen, N., Beattie, M., & Dombrowski, S. U. (2018). How can interventions increase motivation for physical activity? A systematic review and meta-analysis.

298

ALEXANDER J. ROTHMAN, WILLIAM M. P. KLEIN, AND PASCHAL SHEERAN

Health Psychology Review, 12, 211–230. https:// doi.org/10.1080/17437199.2018.1435299 Kok, G., Gottlieb, N. H., Peters, G.-J. Y. et al. (2016). A taxonomy of behaviour change methods: An intervention mapping approach. Health Psychology Review, 10, 297–312, https://doi.org/ 10.1080/17437199.2015.1077155 Kwasnicka, D., Dombrowski, S. U., White, M., & Sniehotta, F. (2016). Theoretical explanations for maintenance of behavior change: A systematic review of behavior theories. Health Psychology Review, 10, 277–296. https://doi.org/10.1080/ 17437199.2016.1151372 MacKinnon, D. P. (2008). Introduction to Statistical Mediation Analysis. Mahwah, NJ: Lawrence Erlbaum Associates. MacLean, P. S., Rothman, A. J., Nicastro, H. L. et al. (2018). The accumulating data to optimally predict obesity treatment (ADOPT) core measures project: Rationale and approach. Obesity, 26, S6– S15. https://doi.org/10.1002/oby.22154 McEachan, R. R., Conner, M., Taylor, N. J., & Lawton, R. J. (2011). Prospective prediction of healthrelated behaviours with the theory of planned behaviour: A meta-analysis. Health Psychology Review, 5, 97–144. https://doi.org/10.1080/ 17437199.2010.521684 Meehl, P. E. (1978). Theoretical risks and tabular asterisks: Sir Karl, Sir Ronald, and the slow progress of soft psychology. Journal of Consulting and Clinical Psychology, 46, 806–834. https://doi .org/10.1037//0022-006x.46.4.806 Michie, S., Carey, R. N., Johnston, M. et al. (2018). From theory-inspired to theory-based interventions: A protocol for developing and testing a methodology for linking behaviour change techniques to theoretical mechanisms of action. Annals of Behavioral Medicine, 52, 501–512. https://doi.org/10.1007/s12160-0169816-6 Michie, S., Richardson. M., Johnston, M. et al. (2013). The Behavior Change Technique Taxonomy (v1) of 93 hierarchically clustered techniques: Building an international consensus for the reporting of behavior change interventions. Annals of Behavioral Medicine, 46, 81–95. https:// doi.org/10.1007/s12160-013-9486-6

Moher, D., Schultz, K. F., Altman, D. G., & the CONSORT Group. (2001). The CONSORT statement: Revised recommendations for improving the quality of reports of parallel-group randomized trials. Lancet, 357, 1191–1194. https://doi.org/10.1016/s0140-6736(00)04337-3 Munafò, M. R., Nosek, B. A., Bishop, D. V. M. et al. (2017). A manifesto for reproducible science. Nature Human Behavior, 1 (21). https://doi.org/ 10.1038/s41562-016-0021 Nielsen, L., Riddle, M., King, J. W., & the NIH Science of Behavior Change Implementation Team. (2018). The NIH science of behavior change program: Transforming the science through a focus on mechanisms of change. Behaviour Research and Therapy, 101, 3–11. https://doi.org/ 10.1016/j.brat.2017.07.002 Nosek, B. A., Ebersole, C. R., DeHaven, A. C., & Mellor, D. T. (2018). The preregistration revolution. Proceedings of the National Academy of Sciences, 115, 2600–2606. https://doi.org/ 10.1073/pnas.1708274114 Patel, M., Lee, A. D., Redd, S. B. et al. (2019). Increase in measles cases – United States, January 1–April 26, 2019. MMWR. Morbidity and Mortality Weekly Report, 68, 402–404. http://dx.doi.org/ 10.15585/mmwr.mm6817e1 Rothman, A. J., Baldwin, A. S., Hertel, A. W., & Fuglestad, P. (2011). Self-regulation and behavior change: Disentangling behavioral initiation and behavioral maintenance. In K. D. Vohs & R. F. Baumeister (Eds.), Handbook of Self-Regulation: Research, Theory, and Applications (2nd ed., pp. 106–122). New York: Guilford Press. Sheeran, P., Avishai, A., Jones, K., Villegas, M., Wright, C. E., & Brewer, N. T. (2019). Health behavior tasks: Conceptualizing actions to promote health and prevent disease. Unpublished manuscript, University of North Carolina at Chapel Hill. Sheeran, P., Harris, P. R., & Epton, T. (2014). Does heightening risk appraisals change people’s intentions and behavior? A meta-analysis of experimental studies. Psychological Bulletin, 140, 511–543. https://doi.org/10.1037/a0033065 Sheeran, P. S., Klein, W. M. P., & Rothman, A. J. (2017). Health behavior change: Moving from observation to intervention. Annual Review of

Moving from Theoretical Principles to Intervention Strategies

Psychology, 68, 573–600. https://doi.org/10.1146/ annurev-psych-010416-044007 Sheeran, P., Maki, A., Montanaro, E. et al. (2016). The impact of changing attitudes, norms, and selfefficacy on health-related intentions and behavior: A meta-analysis. Health Psychology, 35, 1178– 1188. https://doi.org/10.1037/hea0000387 Spencer, S. J., Zanna, M. P., & Fong, G. T. (2005). Establishing a causal chain: Why experiments are often more effective than mediational analyses in examining psychological processes. Journal of Personality and Social Psychology, 89, 845–851. http://doi.org/10.1037/0022-3514.89.6.845 Swim, J. K., Clayton, S., & Howard, G. S. (2011). Human behavioral contributions to climate

299

change. American Psychologist, 66, 251–264. https://doi.org/10.1037/a0023472 Tannenbaum, M. B., Hepler, J., Zimmerman, R. S. et al. (2015). Appealing to fear: A meta-analysis of fear appeal effectiveness and theories. Psychological Bulletin, 141, 1178–1204. http://dx.doi.org/ 10.1037/a0039729 Webb, T. L., & Sheeran, P. (2006). Does changing behavioral intentions engender behavior change? A meta-analysis of the experimental evidence. Psychological Bulletin, 132, 249–268. https://doi .org/10.1037/0033-2909.132.2.249 Weinstein, N. D. (2007). Misleading tests of health behavior theories. Annals of Behavioral Medicine, 33, 1–10.

21 Developing Behavior Change Interventions Nelli Hankonen and Wendy Hardeman

Practical Summary When trying to help people change behavior, it is important that intervention designers consider carefully what actions might bring about the desired outcomes and why. A systematic approach to intervention development aids this process. Key tasks are (1) identify and analyze the problem addressed by the intervention; (2) identify how the intervention will achieve the desired outcomes, decide on its content and delivery mode (s), and design a logic model or program theory; (3) develop intervention materials or prototypes (e.g., interface); and (4) test or pilot the intervention iteratively in an early stage. This approach can be adopted to develop new interventions and to optimize existing interventions. It is useful to build an explicit model of assumed influences of an intervention during its design, including influences on behavior, contextual influences on implementation, and the causal pathway involved. The decisions about intervention content and delivery modes are informed by an understanding of the target group, behaviors, context, and working mechanisms of behavior change techniques.

21.1 Introduction Intervention development is an essential process in which several parties work together to produce an intervention that is fit-for-purpose and likely to be effective in changing the behavior of the target population. Parties involved are likely to include practitioners, researchers, the target group, and other stakeholders (see also Chapters 24 and 25, this volume). The scope of development work is influenced by the available time and resources. The aim of this chapter is to provide an overview of the basic steps in the intervention development process. The chapter shows how careful development helps ensure that the intervention is (1) informed by evidence from multiple sources (e.g., the research literature, the target group or population, stakeholders, relevant theory); (2) that the intervention is an appropriate solution to a problem; (3) that

context and any challenges in the implementation of the intervention have been carefully considered and addressed; (4) that feasibility, acceptability, and resource use (value for money) have been considered at an early stage; (5) that the intervention is optimized prior to evaluation; and (6) that the evaluation will focus on key uncertainties about the intervention. A carefully reported development process is also crucial in understanding issues arising from intervention implementation, such as providing insight into what may have gone wrong if the intervention does not achieve stipulated goals or is labeled as “ineffective.” It also provides necessary detail to contribute to the evidence base of behavioral interventions and behavior change theories. The chapter will focus on the development of behavior change interventions across disciplines and https://doi.org/10.1017/9781108677318.021

Developing Behavior Change Interventions

behavioral domains and cover interventions that focus on changing the behavior of individuals as well as larger-scale interventions that change the behavior of groups all the way up to populationlevel interventions. Key frameworks that have been applied to inform development of behavior change interventions are the intervention mapping approach (Bartholomew Eldredge et al., 2016), the behavior change wheel (Michie, Atkins, & West, 2014; Michie, van Stralen, & West, 2011), and the UK Medical Research Council (MRC) framework for complex interventions. Readers are directed to these key frameworks for further reading and for examples of the application of the various development steps (for a comprehensive list, the reader is directed to Araújo-Soares et al., 2018; O’Cathain et al., 2019; Chapter 19, this volume). The steps suggested by the frameworks are shown in Table 21.1. Intervention mapping explains a careful, stepwise process in translating theories of behavior change (see the chapters in Part I) into practical health promotion programs. The behavior change wheel (Michie et al., 2011; Michie et al., 2014) integrates theory and evidence in informing decisions about intervention design together with stakeholders. It is based on a comprehensive review and synthesis of existing frameworks for intervention development. By contrast, the UK Medical Research Council (MRC) framework (Craig et al., 2008) offers more general guidance to intervention development, focusing on identifying the evidence base, identifying or developing theory, and modeling process and outcomes.

21.2 Overarching Principles in Developing Behavior Change Interventions This section describes some key general principles relevant to the tasks involved in intervention development (see Chapters 19 and 20, this volume). Commonly used frameworks during the development of behavior change

301

interventions share the following key tasks (see Table 21.1): Task 1: Identifying and analyzing the problem addressed in behavioral terms and developing intervention objectives. Task 2: Identifying intervention mechanisms, content, and delivery mode, including the design of a logic model or program theory. Task 3: Developing materials and/or technology. Task 4: Early, iterative testing of the intervention and empirical optimization. In research, these phases are usually followed by a feasibility or pilot study, a substantive evaluation study (e.g., a randomized controlled trial), and an implementation phase (see Chapter 22, this volume). The following sections describe some overarching principles that apply to all four tasks.

21.2.1 Flexibility: An Iterative Rather Than Linear Process Although the key tasks involved in intervention development are presented in a specific order, intervention development should be considered an iterative rather than linear process. For example, pretesting of an intervention (task 4) may reveal the need to revise intervention materials (task 3). The above steps are applicable to developing new interventions and to optimizing existing interventions. For example, an intervention in the real world may fail to reach its intended target group and is, thus, in need of optimization. If the intervention has not been specified or described well, development of a logic model (task 2) may help reveal weak links that could be optimized, in this example an asset-based approach might be adopted, which brings the intervention to the target population using existing resources (e.g., community groups). Alternatively, the tasks may be completed by different stakeholders, for example practitioners and decision makers may have done a problem analysis and a multidisciplinary

Table 21.1 Intervention development tasks mapped onto phases and steps of two influential frameworks and two reviews of frameworks

Task Problem

Tasks in Intervention Development (this chapter)

Systematic Review (O’Cathain et al., 2019)

Integrative Review (Araújo-Soares et al., 2018)

Task 1: What is the problem to be addressed?

1. Conception 2. Planning

• Analyze the problem in

• Establish a group,

A. Analyzing the problem and developing an intervention objective

behavioral terms (needs assessment)

• Develop a preliminary logic model or program theory

• Consider systems of beha viors and ecological system that the problem and behaviors are part of

including stakeholders

Behavior Change Wheel (Michie et al., 2014)

Step 1: Logic model of the problem

Stage 1: Understanding the behavior

• Establish and work with a planning group

• Understand the problems

• Conduct a needs assessment

• Step 1: Define the problem in behavioral terms

• Step 2: Select to create a logic model of the

or issues to be addressed

• Decide/define behaviors • Identify possible ways of

problem

• Describe the context for the intervention including the population, setting and community

making changes

• Specify who will change, how, and when

• Define the target behavior(s) • Consider costs and and target groups

Intervention Mapping (Bartholomew Eldredge et al., 2016)

target behavior

• Step 3: Specify the target behavior

• Step 4: Identify what needs to change

• State program goals

delivery

• Consider whether it Intervention

is worthwhile 3. Designing

Task 2: What are the hypothesized mechanisms of effect on behavior and intervention components?

• Ideas about solutions,

• Understand the target

• Revisit decisions about

behavior(s) informed by theory and evidence

• Select key modifiable determinants or influencing factors to be targeted

• Define and develop intervention content and delivery modes

and components and features where to intervene

• Decisions about the content, format, and delivery of the intervention

• Implementation plan

B. Defining the scientific core of the intervention

• Causal modeling • Defining intervention features

• Developing a logic model of change

Stage 2: Identify intervention Step 2: Program outcomes options and objectives – Logic model of change • Step 5: Identify intervention functions • State expected outcomes for behavior and • Step 6: Identify environment policy categories • Specify performance objectives for behavioral and Stage 3: Identifying content and environmental outcomes implementation options • Select determinants Step 7: Identify behavior for behavioral and change techniques environmental outcomes Step 8: Identify mode • Construct matrices of of delivery (MOD) change objectives

Continued

• Bringing it all together: A

• Create a logic model

well-defined logic model or program theory

of change Step 3: Program design

• Generate program themes, components, scope, and sequence

• Choose theory- and evidence-based change methods

• Select or design practical

Materials

Task 3: Development of intervention materials and technology

4.Creating

• Prototypes or mock-ups

applications to deliver change methods C. Development of material and Step 4: Program production interface • Refine program structure and organization

• Prepare plans for program materials

• Draft messages, materials, and protocols

• Pretest, refine, and produce materials

Testing and Task 4: Empirical optimization optimiof the intervention zation

5. Refining

D. Empirical optimization

Step 5: Implementation plan (see Step 4)

• Test on small samples • Test on a more diverse population • – Optimize

Continued

Table 21.1 (Cont.)

Task

Tasks in Intervention Development (this chapter)

Systematic Review (O’Cathain et al., 2019)

Integrative Review (Araújo-Soares et al., 2018)

Intervention Mapping (Bartholomew Eldredge et al., 2016)

6. Documenting

• Document the interven-

Evaluation

See Chapter 22, this volume

tion, describing the intervention so others can use it 7. Planning for future evaluation

E. Evaluating the intervention F. Process evaluation

Step 6: Evaluation plan

• Develop the objectives

Implementation

See Chapter 23, this volume

of the outcome and process evaluations (see 3., Designing – implementation plan)

G. Implementation: real-world application

(Step 5: Implementation plan)

Behavior Change Wheel (Michie et al., 2014)

Developing Behavior Change Interventions

team may do the remaining tasks for the intervention development. The key tasks may differ across interventions according to (1) the context of intervention development, that is, whether the intervention is in a research, policy, or practice context, and (2) the mode of delivery. Focusing on context, the time frame and scope of intervention development work need to be matched to resources, expertise, and time. Intervention development as a part of a research study is often more comprehensive, based on theory and evidence, and takes longer than when this is done by real-world organizations (e.g., a health protection agency that needs to act quickly in response to emerging issues such as a new flu outbreak). A lengthy, comprehensive, empirically based intervention development is rarely the case in practice. With time and resource constraints, one may conduct rapid reviews of the evidence and a quick succession of stakeholder consultations, rather than systematic reviews and comprehensive qualitative research, and focus on key uncertainties surrounding the intervention. Reporting the process carefully (see Sidebar 21.1) is useful for any intervention to inform long-term monitoring, audit, or evaluation.

21.2.2 Consider Constraints in Intervention Development Intervention developers rarely have all the options for intervention development available. Commissioners or funders often set constraints in terms of a specific mode of delivery (e.g., a mobile phone application), context, (e.g., school-based intervention), or providers (e.g., teachers or nurses). A priori practical constraints may dictate maximum dose (e.g., maximum feasible contact time with target group of five minutes for a health care consultation or three hours in school curriculum) or a cost threshold for the intervention. Any constraints need to be made explicit as they influence decisions at all stages (see also Sidebar 21.1) and may limit the

305

available candidate intervention components. All constraints need to be monitored throughout the intervention development. Common constraints are costs. Economic modeling and estimation of resource use during intervention development can help ensure that the resulting intervention is likely to be cost-effective, and many intervention development teams now include economists who build an economic model to test assumptions (see Chapter 26, this volume). Time and financial resources may also affect the scope of intervention development work, especially for interventions developed or optimized in the real world. In these instances, it can be helpful to draw a logic model and/or to define the key uncertainties about the proposed intervention and focus limited development resources on addressing these uncertainties.

21.2.3 Including Stakeholders and End Users in the Intervention Development Team Across all tasks, intervention development should be informed by relevant expertise and evidence. Academic disciplines may include psychology, policy, sociology, economics, business, computer science, and service design. Key stakeholders could include members of the public, representatives of the target group(s), intervention providers/professionals, commissioners, policy makers, and funders. Involvement of experts in the relevant disciplines and key stakeholders aids understanding of the intervention context and informs strategic decisions that reflect scientific and practical expertise, experiential evidence, and the preferences and views of the end users and those whose involvement is critical for the adoption and implementation of the intervention (e.g., Bartholomew Eldredge et al., 2016). It also assists with the recruitment and engagement of the target group and the cost-effectiveness, coownership, and codesign of a comprehensive

306

NELLI HANKONEN AND WENDY HARDEMAN

logic model or program theory. Ideally, key stakeholders should be consulted in the earliest stages of the initial problem formulation (see Chapters 24 and 25, this volume). The selection of stakeholder groups is influenced by the skills needed to develop the intervention, those involved in the adoption and implementation of the intervention, commissioning the intervention, and those who will help increase impact. For example, decision makers who can ensure implementation of the behavioral intervention in the long term should ideally be included in the intervention development team. Their involvement will facilitate reach, adoption, and implementation of the intervention once the research evaluating the intervention has been concluded (Chapter 23, this volume).

21.2.4 Using a “Complex Systems” Approach During Intervention Development Recently, calls have been made to develop and study interventions with explicit complexity science perspective (e.g., Hawe, 2015; Heino et al., 2019; Skivington et al., 2018). The target setting and behavior of an intervention, as well as the intervention itself, can be considered a complex, adaptive, and dynamic system, which is more than the sum of its parts (see Gomersall, 2018; Resnicow & Page, 2008). Complex systems share, for example, the following key features: (1) interconnections in the system, that is, relationships and interconnections between different parts or components are important rather than the individual parts separately; (2) selforganization and emergence, that is, order is created in a system without explicit hierarchical direction or central planning; and (3) coevolution of the system and the environment, that is, the system (e.g., intervention) and the environment influence each other’s development (for an overview, see Heino et al., 2019). There is some evidence that interventions that make use of

these features are more effective than more simplistic interventions (Leykum et al., 2007). Although many interventions and policies incorporate some principles of the complex systems approach, traditional intervention development frameworks have not explicitly drawn on these fully. Systems thinking could be more explicitly integrated and adopted in both such frameworks and behavior change theories. Additionally, this approach can encourage awareness of real-world uncertainties that the proposed intervention could address or consider. Therefore, intervention developers are encouraged not only to consider how the intervention is expected to work (i.e., the internal intervention logic and complexity therein) but to consider the overall system, what parts of the system could influence the intervention, and how the intervention could lead to wider system change.

21.3 Key Tasks in Intervention Development The following sections report key tasks in intervention development that help to iteratively identify, develop, and refine the content and mode(s) of delivery of a theory- and evidence-based intervention, including consideration of context, implementation, and hypothesized mechanism of effect (see Table 21.1). Documenting the sequence of decisions during intervention development can be helpful. The important background features and decisions involved may include, but are not limited to describing, (1) issues at the beginning (e.g., preparatory work to describe the team and planned development process); (2) the time used and available for intervention development process (e.g., length of design period, frequency of design meetings); (3) the possible commissioner demands, limitations, or requests of the intervention or the development process (e.g., future use, use of technology, limited financial resources, quick timeline for development); and

Developing Behavior Change Interventions

(4) details of decisions during the process, including considered alternative options, leading to choices about the intervention (e.g., program components/activities; intermediate targets; behavior change techniques to target predictors/mechanisms, including to what extent various combinations of techniques were explicitly considered and left out; see also Sidebar 21.1).

21.3.1 Task 1: What Is the Problem to Be Addressed? A new intervention is essentially a solution to a perceived and defined problem. If the problem is not well defined, the intervention may not achieve the desired outcomes, such as increasing patient safety, reducing costs in the health service, improving literary levels, or reducing inequalities. Therefore, the first step is to clearly define the problem to be addressed by the intervention and justify why the problem matters. The problem may be defined at several levels (e.g., individual, organizational, and societal). Examples of problems are the burden on society, such as sickness absences from work or demands on health care services due to the rising prevalence of Type 2 diabetes (societal-level problems); health care professionals lacking the time for promoting healthy behaviors (an organizational-level problem); or parents deciding not to vaccinate their children (an individual-level problem). The importance of a health problem is often quantified as its cost to health and social care systems and society and the burden of disease for patients. It is important to identify and define the target population at an early stage: who is affected by the problem; who might benefit most from the proposed intervention; how they can be reached by the intervention; and how these considerations influence choices about the content and mode of delivery of the intervention. Again, this involves a careful review of context and evidence, including research, policy documents, and stakeholder consultations. The next sections outline some of

307

the specific tasks to be completed during the definition of the problem to be addressed.

21.3.1.1 Analyze the Problem in Behavioral Terms A thorough analysis of the problem includes an understanding of the context of the proposed or existing intervention. Identification of the current behavioral status of the target population (e.g., are people doing what they should be doing to achieve desired outcomes?) and understanding and defining what needs to change, who needs to change, and when and where the change might occur are all important questions to address. This analysis will inform the objectives or goals of the intervention. Specifying these also allows a detailed understanding of potential constraints, linked to context (e.g., school), intervention duration (e.g., five minutes maximum), or cost (e.g., cost per participant). It helps avoid inappropriate and ineffective solutions to the problem. As an example, when the delivery of evidence-based health care is suboptimal (e.g., hospital nurses do not wash their hands in between seeing patients or family doctors overprescribe antibiotics), common interventions are guidelines and training. These tend to focus on the individual level, whereas a detailed problem analysis might have revealed that the problem is not at the individual level but at the organizational or team level (e.g., lack of alcohol rub, incentivization of care, perceived team roles). In intervention mapping, needs assessment involves assessing the problem (e.g., air pollution in inner cities) and its likely behavioral, social, and environmental causes. This involves the identification and definition of the sequence of behaviors needed to achieve desired outcomes, based on existing evidence (e.g., people using more sustainable alternatives, such as public transport, cycling, or walking rather than driving a car; city councils building the infrastructure to promote sustainable travel) and the identification

Sidebar 21.1 Recording and reporting the intervention development process Guidance such as TIDieR (Hoffmann et al., 2014) has improved the transparent reporting of interventions. Furthermore, the process of and decisions during intervention development can be reported. For example, Araújo-Soares et al. (2018) suggest reporting several items in their step 2, “defining the scientific core of the intervention,” which corresponds to the task 2 of this chapter. The checklist (adapted) includes the following: (1) Understand causal/contextual factors (causal modeling) Describe: a. formal (behavioral) theories used in understanding the predictors of the target behavior b. how key uncertainties were identified to select the aim(s) of evidence synthesis c. literature search and review process d. the rationale/aims and the process of (potential) original empirical research e. rating of influencing factors (psychological, social, predictors/mechanisms) for changeability and relevance (2) Develop a logic model or program theory Describe: a. the process of developing the model (if possible, include early and later versions) b. key explicit criteria (e.g., acceptability, cost-effectiveness) in making decisions about the model c. whether and which other similar existing interventions were used in developing the logic model or program theory or whether an existing intervention was used as core basis and retrofitted d. key uncertainties left in the logic model or program theory and the possible “weak links” the development team considers merit further investigation e. assessment of evaluability potential of the intervention f. (and develop) a dark logic model that describes considerations made around potential unintended consequences and steps made to avoid them (3) Define intervention features Describe: a. decision processes (including considered alternative options) leading to decisions about i. program components/activities ii. intermediate targets iii. behavior change techniques or methods to target predictors/mechanisms, e.g., to what extent various combinations of techniques were explicitly considered and left out iv. dose/intensity/frequency/duration v. delivery channel(s) vi. providers (expertise/background/training) vii. location/infrastructure b. whether and how anticipated acceptability was investigated c. the decision processes related to scope for local adaptation and extent of fidelity assessment

Developing Behavior Change Interventions

of intermediate outcomes relevant for the hypothesized mechanisms of the intervention (Bartholomew Eldredge et al., 2016).

21.3.1.2 Develop a Preliminary Logic Model or Program Theory A clearly defined problem to be addressed by the proposed intervention and specification of the intervention’s context provides a helpful starting point for an early draft of a logic model or program theory. A logic model is a graphical depiction of “if-then” relationships between the resources needed for the intervention, the activities or components of the intervention, and the hypothesized short-term, medium-term, and long-term outcomes and impact of the intervention (W. K. Kellogg Foundation, 2004). Logic models focus on the big ideas, not the details of your intervention, and are depicted on one page. A logic model reads like a series of “if-then” statements that connect the components of the intervention. Logic models vary in content. A logic model facilitates a shared understanding among key stakeholders about the problem addressed, the justification for the intervention, its components, and how it is hypothesized to work. It can help identify weak links in need of development work to resolve. Different templates for logic models are available (University of Wisconsin, 2008; University of Wisconsin-Extension, 2003; W. K. Kellogg Foundation, 2004). Readers are directed to various examples of logic models and program theories (Davidoff et al., 2015; Evans et al., 2018; Morgan et al., 2015; Tully et al., 2019; Appendix 21.1, supplemental materials). The literature uses multiple terms – logic models, program theories, intervention theories, and theories of change – and consensus is lacking about their similarities and differences. Logic models tend to depict a temporal sequence, describing inputs/ resources, outputs (activities and participation), outcomes, and impact, whereas program theories and theories of change tend to depict how the intervention is expected to work in its context. However, the

309

distinctions are blurred: The W. K. Kellogg Foundation (2004) states the theory approach as one of the three approaches to their logic models: the theory of change that influenced intervention design and plan. Davidoff and colleagues (2015) report that a program theory specifies (1) the components, expected outcomes, and the methods for assessing the outcomes of an intervention, often in the form of a logic model, and (2) the “hypotheses” of the intervention, that is, specification of the rationale behind, and underlying assumptions of, the mechanisms that link the processes and inputs of the intended and unintended outcomes of the intervention, as well as the conditions and context necessary for effectiveness (Davidoff et al., 2015). Realist program theories consist of context, mechanism, and outcome configurations (CMOCs), which are generated in consultation with stakeholders and allow one to understand “what works for whom in what circumstances” (Pawson, 2006). A pragmatic approach can be recommended: In selecting a suitable format, for example logic model or program theory, consider which format fits the proposed intervention and purpose best, for instance who will use the logic model or program theory, how will it be used, and whether it is used during the development, implementation, or evaluation phase. If the key purpose is to tell the “big story” of the intervention and its effects over time to present to a wide range of stakeholders, or details of implementation, then a logic model may be best. If the key purpose is to depict the causal pathways, then a program theory may be suitable. If the purpose is to depict the problem, one could use the intervention mapping guidance. Finally, one can combine components of logic models and program theories (e.g., describe the problem, intervention activities, and causal pathways over time). In the later tasks, the logic model or program theory can be extended to include determinants of the target behavior or behaviors, the behavior change techniques that will be used to change them (e.g., goal setting and incentives), other

310

NELLI HANKONEN AND WENDY HARDEMAN

intervention components (e.g., such as a face-toface meetings, leaflets, or a smartphone app), and measures of outcomes and impact in the short, medium, and long term. The task of the development team is then to refine this initial logic model, together with stakeholders. It should be noted that a logic model of the problem can be separated from the logic model of the proposed solution (see Bartholomew Eldredge et al., 2016) or combined.

21.3.1.3 Consider Systems of Behaviors and an Ecological System of the Target Problem and/or Behavior Problems addressed and behaviors targeted by a proposed intervention do not exist in a vacuum but are dependent on other behaviors and contexts, that is, they are part of a wider system. Other behaviors may facilitate the target behavior or conflict with it. Therefore, intervention developers need to consider the target behavior in the context of other behaviors (e.g., clustering of unhealthy behaviors and consumer behaviors). Behavior results from a system of influences, including proximal individual cognitive and emotional factors, social and community influences, and distal factors such as living and working conditions (environment, housing, education) and socioeconomic, cultural, and environmental conditions (see Chapters 17, 18, and 28, this volume). Further, creating a map of social agents (which groups of people may influence the problem and/or the target group’s behavior) can be helpful. Consideration of the functionalities of the system, including interactions and feedback loops, is also important.

21.3.1.4 Defining the Target Behavior(s) and Target Group(s) Influential frameworks for intervention development stress the importance of clearly defining and specifying the target behavior(s) for the intervention. The selection of the target behavior depends on various considerations, such as being an important cause of the problem, modifiability, and

acceptability by the target group. For example, in an intervention study to reduce sedentary behavior among adolescents, empirical studies showed that an early selection of “reducing screen time” as the target behavior was perceived as highly unacceptable by the target group. As a consequence, the target behavior was changed to reducing sitting time in the school context (Hankonen et al., 2016). Once the target behavior has been identified, it can be added to the logic model along with the intervention objectives. Also, at this point it is important to check that the definition of the target groups is clear and agreed on, as well as possible subpopulations/segments.

21.3.2 Task 2: What Are the Hypothesized Mechanisms of Effect on Behavior and Intervention Components? In this phase, developers will further populate the logic model by including the hypothesized mechanisms of effect, informed by an understanding of the behavior, and intervention components to the logic model. The developers will need to source information or evidence on key components, or “links in the chain,” of the proposed mechanism (i.e., how the intervention will achieve its effect). The team decides on and records: • Key modifiable influences of the target behavior(s) • Hypothesized causal mechanisms of intervention effects • The nature of the intervention (intervention function) • Behavior change technique(s) • Mode(s) of delivery, e.g., face-to-face or digital • Intervention provider(s) (if relevant) • Personalization and tailoring of intervention content (especially relevant for digital interventions) • Setting(s) • Intervention intensity, such as timing and dose

Developing Behavior Change Interventions

• Strategies to optimize reach, (cost-)effectiveness, adoption, implementation, and long-term maintenance of the intervention This is not an exhaustive list as the key issues that need to be considered depend on the nature of the intervention and its context.

21.3.2.1 Understand the Target Behavior(s) Informed by Theory and Evidence A wide range of theories and frameworks have been applied to predict and understand behavior (for examples, see the chapters in Part I of this handbook). The capability, opportunity, motivation – behavior (COM-B) model can be used as a starting point (Michie et al., 2011), as it is an overarching framework of behavior, as well as the theoretical domains framework as a framework to achieve a more fine-grained understanding of behavioral influences. Understanding the behavior also involves a consideration of preparatory behaviors, which may be influenced by unique determinants. For example, for someone to take medication as prescribed, they will need to obtain the medication from a pharmacy or, in order to use a condom, both partners need to negotiate its use. As a result, behavioral targets may extend beyond the single behavioral outcome of the intervention (see Section 21.1.3.5). Preparatory behaviors can be added to the logic model along with intervention objectives.

21.3.2.2 Select Key Modifiable Determinants to Be Targeted Target behaviors and determinants of behavior (or influences on behavior) targeted by the proposed intervention need to be prioritized and selected. Appendix 21.2 (supplemental materials) provides a list of criteria that are commonly used in intervention development: acceptability, practicability, effectiveness/relevance, affordability, possible side effects, and equity. A key consideration is changeability: the extent that determinants can be changed based on current evidence and theory and

311

the impact of those changes on key outcomes (e.g., Araújo-Soares et al., 2018; Bartholomew Eldredge et al., 2016). Changeable, modifiable factors that have a strong relationship to the target behavior are potential targets for interventions (e.g., Michie et al., 2011). Identifying the key modifiable influences on behavior often requires a range of quantitative and qualitative methods, data, and sources (see Appendixes 21.3 and 21.4, supplemental materials).

21.3.2.3 Define and Develop Intervention Content and Delivery Modes Once the key modifiable determinants of the target behavior have been identified, the next task is to select intervention techniques, that is, the methods or strategies that will affect a change in behavior by changing the identified determinants such as cognitions or environmental variables. This selection should be based on formative evidence on the effectiveness of the techniques in changing the determinants and the subsequent change in the target behavior or the plausibility of such links, based on evidence from pathways of similar interventions, if such evidence is not available. Often behavior change interventions have had heterogeneous effects, which can make it challenging to select intervention components. In this case, it may be possible to retrospectively “code” the behavior change techniques and other intervention features (e.g., modes of delivery) used in previous intervention work during evidence synthesis as part of the development work in order to provide the requisite evidence on which to base the selection of determinants and matched techniques. Such an approach would enable the intervention designer to explore, narratively or quantitatively, the extent to which intervention techniques and other features are associated with intervention effectiveness (e.g., Dombrowski et al., 2012). This approach is not without its limitations – for example, studies rarely report

312

NELLI HANKONEN AND WENDY HARDEMAN

the fidelity of delivery of the techniques as well as other features and whether participants used them in their daily lives (“enactment”), both of which impact intervention effectiveness (see Toomey et al., 2019). Other relevant sources of evidence to inform the selection of intervention components include evidence regarding the setting, target behavior, behavior change techniques, and mode of delivery. This evidence can come from evidence synthesis or existing reviews, qualitative research, and consultation with stakeholders (see Appendix 21.4, supplemental materials). Acceptability is a further important factor that should be accounted for when selecting techniques and other intervention features (e.g., modes of delivery) for inclusion in an intervention. Acceptability of techniques and other intervention features, as perceived by those delivering or receiving the intervention, is defined as a “multifaceted construct that reflects the extent to which people delivering or receiving a healthcare intervention consider it to be appropriate, based on anticipated or experienced cognitive and emotional responses to the intervention” (Sekhon, Cartwright, & Francis, 2017, p. 4). Perceptions about acceptability are linked to ethicality, perceived effectiveness, likability, burden, and coherence of the intervention. Acceptability of intervention components necessitates consultation with stakeholders as part of the coproduction of the intervention, because feedback from those groups is pivotal for the adaptation of the intervention content to the specific context and target group or population. Empirically, acceptability can be investigated before the intervention takes place through discussion with the appropriate stakeholders, who report on the anticipated acceptability of an intervention scenario, but should also be part of the evaluation of the intervention during its delivery or retrospectively after the intervention is complete during follow-up measurement (Sekhon et al., 2017; see Chapter 22, this volume).

Finally, other criteria often need to be taken into consideration. Stakeholders or commissioners of the intervention may have strong preferences about certain features such as the mode of delivery. For example, the team developing the intervention may have been commissioned by an organization to develop an intervention delivered via a preferred mode of delivery. For instance, the organization may specify that an intervention aiming to change the behavior of employees or children in a school setting should be delivered by a mobile phone app or a web-based platform. In addition, feasibility considerations play an important role in the selection of intervention content. For example, the duration or intensity of an intervention can be constrained by its context, and cost-effectiveness considerations may set upper limits on intervention costs. For example, if the selected behavior change techniques to be used in a planned intervention have tended to be delivered via a face-to-face, practitioner-client mode of delivery, the costs associated with the practitioner delivery need to be taken into account in feasibility considerations and it may be decided that such costs are prohibitive, necessitating a rethink of the mode of delivery and, perhaps as a consequence, the techniques selected. For instance, a feasibility study into five-minute physical activity advice in primary care evaluated action planning, an evidence-based BCT. However, no participant defined an action plan, so it did not prove feasible in clinical practice (Pears et al., 2015).

21.3.2.4 Bringing It All Together: A WellDefined Logic Model or Program Theory Section 21.3.1.2 introduced logic models as a graphic representation of the problem, intervention components, causal pathways, process and outcome measures, and expected impact. It is recommended that a logic model or program theory is developed in an iterative manner, starting with an initial draft that includes the context and

Developing Behavior Change Interventions

the problem. The model or theory is then populated once decisions are made about how the intervention is expected to achieve its outcomes (causal pathways), about its behavior change techniques, and about the modes of delivery. A carefully constructed logic model provides intervention designers, intervention deliverers, and all stakeholders with a clear “visual map” of the proposed intervention (Appendix 21.1, supplemental materials). The presumed causal mechanisms can also be expressed verbally, for example, in a series of “if -then,” “so that” statements (Davidoff et al., 2015) or as intervention mapping matrices (Bartholomew Eldredge et al., 2016). It is important to specify not only the desired outcomes and impact of the intervention but also any unintended, negative, and harmful consequences. A “dark” logic model refers to a careful elaboration of potential pathways by which the intervention could lead to negative or harmful consequences (Bonell et al., 2015), such as increasing health inequalities (see Chapter 27, this volume). Instead of only identifying potential harms, this process clearly outlines the mechanisms through which such harms may take place – for example, people who are most in need of the intervention are not reached due to recruitment challenges or a lack of resources (e.g., travel expenses and childcare) required to take part in the intervention (for an example, see Cook et al., 2018).

21.3.3 Task 3: Development of Intervention Materials and Technology If a planned intervention is not attractive and easy to use, the target group will not try it out or continue to use it. The intervention designer wants to maximize the reach and engagement of the target population but for behavioral interventions delivered via digital modes (see Chapter 29, this volume), for example, engagement is a key challenge (Perski et al., 2017). Promoting engagement

313

of key stakeholders with the intervention is another critical design task (see Chapter 24, this volume). Design decisions on the “look” and “feel” of intervention materials, such as posters or leaflets, depend on the target audience, behavior, and the chosen mode of delivery. Successful engagement with the intervention relies on effective coproducing of the intervention with stakeholders, particularly with members of the target group. Adopting a multidisciplinary approach when selecting the team employed to design the intervention is also important to enable input from design-related and creative disciplines such as computer science, environmental design, or educational sciences. Final program material production (e.g., posters, videos, smartphone apps) may involve creative consultants, artists, or graphic designers. It can be beneficial to work with advertising professionals, graphical artists, and website and app developers in the production of intervention materials and technology. However, it is important to bear in mind that approaches to changing consumer preferences for products are different to changing other behaviors, and interventions with a host of features may not necessarily increase engagement. Researchers, the target group, practitioners, and any other stakeholders need to work closely with those designing the intervention materials. Writing design documents to guide the creation and review of intervention materials and technology can help in ensuring that behavioral science insights and intervention strategies are adequately transferred into the production of materials (Bartholomew Eldredge et al., 2016). The importance of behavior change expertise input is particularly important to note, as advertising professionals may have limited knowledge and experience of the techniques used to change behavior and their effectiveness (see Chapters 19 and 20, this volume). To make the intervention attractive, clear, and relevant to end users, it is good practice to engage

314

NELLI HANKONEN AND WENDY HARDEMAN

them at an early stage (see Chapters 24 and 25, this volume) using consensus conferences, codesign workshops, and user-centered design (O’Cathain et al., 2019). It is crucial to coproduce early prototypes of intervention materials with the end users and conduct early user testing in an iterative manner (agile development). In practice, this task often happens in parallel with task 2.

21.3.4 Task 4: Empirical Optimization of the Intervention In this task, the intervention designer pilot tests the intervention in a small group of end users from the target group or population of interest in order to identify and solve any problems or issues before a feasibility study or, in real-world situations, to conduct a full-scale pilot test of the intervention. This can involve initial testing using research methods such as a small-scale experiment followed up with surveys, interviews (e.g., data-prompted interviews; Kwasnicka et al., 2015), or focus groups with the participants to collect evidence on issues such as feasibility, acceptability, and fidelity. The feedback may help refine the intervention content and/or mode of delivery. Formal feasibility testing of the intervention, in which the full intervention is tested prior to a fullscale evaluation, is common and can take many forms (see Chapter 22, this volume). Most intervention development frameworks recommend pilot or feasibility testing of a “beta” version of the intervention, in a small-scale study, prior to a wider-scale evaluation such as a fully powered study and rollout. A key aim is to provide important feedback on the intervention from the target population that will enhance its acceptability, feasibility, and fidelity. One approach to refining behavioral interventions is the multiphase optimization strategy (MOST), a framework to rigorously test and select the best options for intervention components (Collins, Murphy, & Strecher, 2007).

21.4 Implications for Research, Practice, and Translation The series of tasks required to design behavioral interventions can, and should, be addressed systematically and reported transparently, within the limits of resources. A detailed understanding of the problem, behavior, and context will facilitate the adoption and implementation of interventions. While acknowledging the complexity and dynamic nature of human and social systems – including the design of behavior change interventions – a systematic approach to intervention development is important if the design team is to develop behavior change interventions that are feasible, acceptable, and, above all, optimally effective. Numerous approaches and frameworks for the development of behavior change interventions share many common features. This chapter has identified those commonalities but there are also a number of unique features of each. It would be helpful to develop evidence-based guidance for researchers, practitioners, and policy makers on which approach is best suited to a particular problem, setting, and intervention. This would necessitate the systematic development of an evidence base in which different approaches were applied to intervention development in like contexts, behaviors, and populations. The generation of such evidence would be useful for all groups involved in intervention design but especially for those working in practice and policy. A further important advance would be to systematically review the content of different approaches to behavior change intervention design (e.g., O’Cathain et al., 2019), with a view to consolidate common features and identify and incorporate useful unique features in order to arrive at a comprehensive and optimally effective approach, perhaps via expert consensus. Another improvement would be the development of a decision tree that helps researchers and intervention developers in practice decide which framework is most applicable for their proposed intervention.

Developing Behavior Change Interventions

Several key future developments and methodological challenges in intervention development methods can be identified. First, more focus is needed on developing interventions that can be adopted, implemented, and sustained in the “real world” through early testing of interventions/ components. Second, further work is needed to understand how to best harness, address, model, and plan for characteristics of dynamic complex systems in intervention development. Third, more transparent reporting of intervention development would help users and other intervention developers understand why some choices were made, in addition to just seeing the final intervention, and thereby improve insight in the “art” of intervention development. Sharing details of decisions will help advance the field by clarifying what alternative pathways developers usually choose between and why. This will also help in developing better methods to best adapt and optimize existing interventions.

21.5 Summary and Conclusion This chapter has outlined the key tasks involved in developing interventions and general principles that underpin all the tasks. It has emphasized the iterative nature of intervention development, the importance of coproducing interventions with the target group and stakeholders, making use of theory and evidence, and the need to adapt the development process according to constraints.

References Araújo-Soares, V., Hankonen, N., Presseau, J., Rodrigues, A., & Sniehotta, F. F. (2018). Developing behavior change interventions for selfmanagement in chronic illness: An integrative overview. European Psychologist, 24, 7–25. https://doi.org/10.1027/1016-9040/a000330 Bartholomew Eldredge, L. K., Markham, C. M., Ruiter, R. A. C., Fernández, M. E., Kok, G., & Parcel, G.

315

S. (2016). Planning Health Promotion Programs: An Intervention Mapping Approach. Hoboken, NJ: John Wiley & Sons. Bonell, C., Jamal, F., Melendez-Torres, G. J., & Cummins, S. (2015). “Dark logic”: Theorising the harmful consequences of public health interventions. Journal of Epidemiology and Community Health, 69, 95–98. https://doi.org/ 10.1136/jech-2014-204671 Collins, L. M., Murphy, S. A., & Strecher, V. (2007). The multiphase optimization strategy (MOST) and the sequential multiple assignment randomized trial (SMART): New methods for more potent eHealth interventions. American Journal of Preventive Medicine, 32, S112–S118. https://doi.org/10.1016/j.amepre.2007.01.022 Cook, P. A., Hargreaves, S. C., Burns, E. J. et al. (2018). Communities in charge of alcohol (CICA): A protocol for a stepped-wedge randomised control trial of an alcohol health champions programme. BMC Public Health, 18, 522. https://doi.org/ 10.1186/s12889-018-5410-0 Craig, P., Dieppe, P., Macintyre, S., Michie, S., Nazareth, I., & Petticrew, M. (2008). Developing and evaluating complex interventions: The new Medical Research Council guidance. BMJ, 337, a1655. https://doi.org/10.1136/bmj.a1655 Davidoff, F., Dixon-Woods, M., Leviton, L., & Michie, S. (2015). Demystifying theory and its use in improvement. BMJ Quality and Safety, 24, 228– 238. https://doi.org/10.1136/bmjqs-2014-003627 Dombrowski, S. U., Sniehotta, F. F., Avenell, A., Johnston, M., MacLennan, G., & Araújo-Soares, V. (2012). Identifying active ingredients in complex behavioural interventions for obese adults with obesity-related co-morbidities or additional risk factors for co-morbidities: A systematic review. Health Psychology Review, 6, 7–32. https://doi.org/ 10.1080/17437199.2010.513298 Evans, R., Brockman, R., Grey, J. et al. (2018). A cluster randomised controlled trial of the Wellbeing in Secondary Education (WISE) Project: An intervention to improve the mental health support and training available to secondary school teachers: protocol for an integrated process evaluation. Trials, 19, 270. https://doi.org/ 10.1186/s13063-018-2617-4

316

NELLI HANKONEN AND WENDY HARDEMAN

Gomersall, T. (2018). Complex adaptive systems: A new approach for understanding health practices. Health Psychology Review, 12, 405–418. https:// doi.org/10.1080/17437199.2018.1488603 Hankonen, N., Heino, M. T. J., Araújo-Soares, V. et al. (2016). “Let’s Move It”: A school-based multilevel intervention to increase physical activity and reduce sedentary behaviour among older adolescents in vocational secondary schools: A study protocol for a cluster-randomised trial. BMC Public Health, 16, 451. https://doi.org/ 10.1186/s12889-016-3094-x Hawe, P. (2015). Lessons from complex interventions to improve health. Annual Review of Public Health, 36, 307–323. https://doi.org/10.1146/ annurev-publhealth-031912-114421 Heino, M. T. J., Noonan, C., Knittle, K., & Hankonen, N. (2019). Studying behaviour change mechanisms under complexity. Unpublished manuscript, University of Helsinki. Hoffmann, T. C., Glasziou, P. P., Boutron, I. et al. (2014). Better reporting of interventions: Template for intervention description and replication (TIDieR) checklist and guide. BMJ, 348, g1687. https://doi.org/10.1136/bmj.g1687 Kwasnicka, D., Dombrowski, S. U., White, M., & Sniehotta, F. F. (2015). Data-prompted interviews: Using individual ecological data to stimulate narratives and explore meanings. Health Psychology, 34, 1191–1194. https://doi.org/ 10.1037/hea0000234 Leykum, L. K., Pugh, J., Lawrence, V. et al. (2007). Organizational interventions employing principles of complexity science have improved outcomes for patients with Type II diabetes. Implementation Science, 2, 28. https://doi.org/ 10.1186/1748-5908-2-28 Michie, S., Atkins, L., & West, R. (2014). The Behaviour Change Wheel: A Guide to Designing Interventions. London: Silverback Publishing. Michie, S., van Stralen, M. M., & West, R. (2011). The behaviour change wheel: A new method for characterising and designing behaviour change interventions. Implementation Science, 6, 42. https://doi.org/10.1186/1748-5908-6-42 Morgan, H., Hoddinott, P., Thomson, G. et al. (2015). Benefits of Incentives for Breastfeeding and

Smoking Cessation in Pregnancy (BIBS): A Mixed-Methods Study to Inform Trial Design. Southampton: NIHR Journals Library. O’Cathain, A., Croot, L., Sworn, K. et al. (2019). Taxonomy of approaches to developing interventions to improve health: A systematic methods overview. Pilot and Feasibility Studies, 5. https://doi.org/10.1186/s40814-019-0425-6 Pawson, R. (2006). Evidence-Based Policy: A Realist Perspective. London: SAGE. Pears, S., Morton, K., Bijker, M., Sutton, S., Hardeman, W., & VBI Programme Team. (2015). Development and feasibility study of very brief interventions for physical activity in primary care. BMC Public Health, 15, 333. https://doi.org/ 10.1186/s12889-015-1703-8 Perski, O., Blandford, A., West, R., & Michie, S. (2017). Conceptualising engagement with digital behaviour change interventions: A systematic review using principles from critical interpretive synthesis. Translational Behavioral Medicine, 7, 254–267. https://doi.org/10.1007/s13142-016-0453-1 Resnicow, K., & Page, S. E. (2008). Embracing chaos and complexity: A quantum change for public health. American Journal of Public Health, 98, 1382–1389. https://doi.org/10.2105/AJPH .2007.129460 Sekhon, M., Cartwright, M., & Francis, J. (2017). Acceptability of healthcare interventions: An overview of reviews and development of a theoretical framework. BMC Health Services Research, 17, 88. https://doi.org/10.1186/s12913017-2031-8 Skivington, K., Matthews, L., Craig, P., Simpson, S., & Moore, L. (2018). Developing and evaluating complex interventions: Updating Medical Research Council guidance to take account of new methodological and theoretical approaches. The Lancet, 392, S2. Meeting abstract: Public Health Science 2018, Belfast, Northern Ireland, November 23, 2018. http://dx.doi.org/10.1016/ s0140-6736(18)32865-4 Toomey, E., Hardeman, W., Hankonen, N. et al. (2019). Focusing on fidelity: Recommendations for improving intervention fidelity within trials of health behavioral interventions. Unpublished manuscript, National University of Ireland, Galway.

Developing Behavior Change Interventions

Tully, M. A., Cunningham, C., Wright, A. et al. (2019). Peer-led walking programme to increase physical activity in inactive 60-to 70-year-olds: Walk with Me pilot RCT. Public Health Research, 7. https:// doi.org/10.3310/phr07100 University of Wisconsin. (2008). Developing a logic model: Teaching and training guide. Program Development and Evaluation. Website. https://fyi .extension.wisc.edu/programdevelopment/logicmodels/

317

University of Wisconsin-Extension. (2003). Enhancing program performance with logic models. Program Development and Evaluation. Website. https://fyi .extension.wisc.edu/programdevelopment/design ing-programs/ W. K. Kellogg Foundation. (2004). Using Logic Models to Bring Together Planning, Evaluation, and Action: Logic Model Development Guide. Michigan: W. K. Kellogg Foundation.

22 Evaluation of Behavior Change Interventions Lynsay Matthews and Sharon A. Simpson

Practical Summary Undertaking an evaluation of a behavior change intervention that has the potential to impact on policy and practice requires careful consideration of multiple factors. The research questions and overall focus of the evaluation need to be established, that is, what is the purpose of the evaluation? This guides the choice of evaluation design and outcomes. For example, (1) an experimental design (e.g., a randomized controlled trial) may be appropriate to identify the efficacy of an intervention within a relatively narrow setting; (2) a quasi-experimental design (e.g., interrupted time series designs) may be suitable in a setting where randomization is not feasible; and (3) a nonexperimental design (e.g., qualitative methods) may be useful in gathering information to help understand the findings of the main evaluation. Developing a comprehensive program theory involving relevant stakeholders will play a vital role in informing the choice of design and outcomes but also identifying the important contextual factors that interact with the intervention to produce outcomes.

22.1 Introduction Undertaking robust evaluations of behavior change interventions, which advance knowledge and provide useful data to inform decision-making and practice, is a crucial step to producing research with impact. Randomized controlled trials (RCTs) have been considered the “gold standard” method for intervention evaluation, producing unbiased effect estimates. Other methods are generally considered to be at greater risk of bias. This hierarchy of evidence approach has shaped the field of evaluation (Murad et al., 2016). However, there has been a shift toward acknowledging the role of other methods in advancing science and addressing questions useful to policymakers and practitioners. The limitations of RCTs and the importance of other methods, such as natural experiments, have been recognized (Craig et al., 2017).

There have also been calls for a change in focus toward considering the “usefulness of evidence” (Leviton & Melichar, 2016; Public Health England, 2018) rather than a focus on achieving a statistically significant difference on a primary outcome variable in tightly controlled RCTs with limited external validity. Thus, the outcome of an evaluation may be to advance theory or produce evidence that can be used by decision makers and practitioners. The RE-AIM framework (Glasgow et al., 2019) has been influential in emphasizing the importance not only of considering efficacy and effectiveness but also whether the intervention reaches the people targeted, can be adopted in different settings, can be implemented by staff with moderate levels of experience and training, and produce replicable long-lasting effects at a https://doi.org/10.1017/9781108677318.022

Evaluation of Behavior Change Interventions

reasonable cost. The importance of considering context and the wider system in which the intervention sits when developing, evaluating, and rolling out interventions has also been identified as key in recent guidance (Craig et al., 2018), including the updated UK Medical Research Council’s (MRC) guidance on developing and evaluating complex interventions (Skivington et al., 2019). This encourages a shift from traditional evaluation approaches to considering new evaluation perspectives and systems thinking, that is, thinking about the “bigger picture” (Egan, McGill, Penney, et al., 2019). Choosing the most appropriate method for an evaluation should be guided by the research questions a researcher or interventionist wishes to address. Some methods can only answer certain types of questions and some designs may be impractical or impossible to utilize in certain circumstances. Furthermore, there is typically a focus of one or more elements, such as an outcome evaluation, process evaluation, or economic evaluation. Generally, an evaluation seeks to answer three fundamental questions: “(a) has a change occurred; (b) did the change occur as a result of the intervention and not some extraneous factor or cause; and (c) is the degree of change perceived to be significant to important stakeholders?” (Sanson-Fisher et al., 2014, p. 13). It is expected that prior to evaluation the researcher or interventionist will have developed a detailed program theory, which captures intervention prerequisites, inputs, mechanisms, contextual dependencies, and outcomes as well as unintended consequences, sometimes referred to as the “dark logic” (Bonell et al., 2015). This will have been achieved either (1) through detailed development work (O’Cathain et al., 2019; Chapter 21, this volume) or (2) by working with policy makers or practitioners if evaluating an existing intervention (e.g., a new policy). To date, many interventions have not been developed in a sufficiently rigorous way and, sometimes when they have been developed well, they are often evaluated using weak

319

research designs giving biased effect estimates. Inadequate attention has also been given to context and implementation issues, which should be considered even at the development stage. If the above are not addressed there is a risk that evaluations of interventions are undertaken that have little chance of being effective when implemented; or the use of inappropriate research designs may lead to promising interventions potentially being discarded because the findings are not deemed valid or fit for purpose. This chapter provides a concise overview of the key factors to consider when evaluating behavior change interventions. It is intended as a starting point to encourage readers to think about issues relevant to their context and research questions and to signpost relevant literature. Overall, the aims of this chapter are to outline the key terms and concepts related to evaluation; introduce different perspectives to evaluation and the types of data produced; provide a brief overview of a few different evaluation designs; and highlight the implications for research, policy, and practice. Examples are provided that should be relevant to researchers, professionals, practitioners, and policy makers working in all areas of behavior change. A glossary of key terms can be found in Appendix 22.1 (supplemental materials).

22.2 Evaluation Aims and Approaches In its broadest sense, evaluation supports judgments about the value of an intervention. Not all evaluations have the same aims. Evaluators should start with considering the questions they want to answer. Is it to assess, for example, whether the intervention is efficacious within a specific setting; or to assess the mechanisms of the intervention; or what happens when the intervention is delivered within different contexts/settings; or all of these. Clarity regarding the aims, research questions, and purpose of the evaluation will help evaluators choose the most appropriate

320

LYNSAY MATTHEWS AND SHARON A. SIMPSON

methods and evaluation design. Examples of evaluation approaches include efficacy, effectiveness, realist, and systems. These perspectives overlap and there is no clear cut-off point. They do, however, focus on different types of research question: (1) an efficacy perspective asks whether the intervention produces the intended outcome in an experimental or “ideal” setting; (2) an effectiveness perspective asks whether the intervention produces the intended outcome in a realworld setting; (3) a realist perspective asks how the intervention works, for whom, under what circumstances, and why; and (4) a systems perspective asks how the intervention produces change by interacting with elements of the system in which it is situated. These approaches can be considered in the development stages of the work and also in the feasibility stage of evaluation to help identify uncertainties and refine research questions for a full evaluation.

22.3.1 Feasibility or Pilot Study A feasibility study seeks to address areas of uncertainty around the feasibility and acceptability of the intervention and evaluation design, as well as exploring the mechanisms of the intervention. It does not aim to assess efficacy or effectiveness (Cook et al., 2018; L. Moore et al., 2018). Key aspects to consider in a feasibility study include: •

•

•

22.3 Stages of Evaluation Although some early “testing” may take place when developing an intervention, evaluation mainly takes place in two stages of the development and evaluation cycle: first in the feasibility or piloting stage and then at the “full” evaluation stage (O’Cathain et al., 2019; Skivington et al., 2019). It is important to note that, although these are referred to as “stages,” the process is not necessarily linear. For example, depending on the result of the feasibility stage, a researcher or interventionist may not necessarily go on to conduct a full evaluation of the intervention. They may have to make significant changes to the intervention and/or evaluation procedure and thus return to the development stage. For some study designs, there may not be a feasibility and full evaluation stage. There might only be a full evaluation stage, for example in the case of a natural experiment. The research questions addressed and methods used in a feasibility study and a full evaluation, although overlapping, are different.

•

•

•

• •

•

Assessment of the feasibility of the intervention design, including, for example, acceptability of the intervention to participants and providers, mode of delivery, system fit, and fidelity. Assessment of the feasibility of the evaluation design, including, for example, recruitment, retention, acceptability of randomization (if appropriate), duration of follow-up, outcome measures, and assessment of potential risks. Avoiding calculating effect estimates as the study is likely underpowered due to inadequate sample size (Cook et al., 2018). Using both qualitative and quantitative methods to capture meaningful and useful data. Engaging with relevant stakeholders before and throughout the feasibility stage to ensure relevant data are collected at the feasibility and/or full evaluation stage for future policy and practice decisions. Taking the opportunity of the feasibility stage to modify the intervention and/or evaluation design. Focusing on producing a refined program theory at the end of the feasibility stage. Using progression criteria, which are predefined before the feasibility study begins, to help assess whether the intervention should proceed to a full evaluation (Hallingberg et al., 2018). This may not be relevant to all evaluation designs. Considering issues related to economic evaluation (discussed further in Chapter 26, this volume).

Evaluation of Behavior Change Interventions

22.3.2 Full Evaluation The approach taken to the full evaluation will depend on the research questions and the purpose of the evaluation. For example, is the aim to assess the efficacy of the intervention or to explore how context interacts with the intervention to produce outcomes? In the case of experimental designs, where the researcher or interventionist may have undertaken a feasibility evaluation prior to full evaluation, key aspects to consider are: •

• •

•

Calculating sample size following DELTA2 guidance, which states that this should be based on a review of the current literature alongside identification of an important and realistic difference as identified by one or more key stakeholders (Cook et al., 2018). Refining the intervention and program theory prior to starting the full evaluation. Addressing fidelity: How much variation in delivery of, and engagement with, the intervention is acceptable without jeopardizing the intervention integrity? (See Walton et al., 2017.) Including an internal pilot with preset progression criteria related to, for example, recruitment, retention, and adherence.

• • •

Inclusion of a process evaluation. Engagement with relevant stakeholders. Inclusion of a health economic evaluation.

A range of evaluation designs are available and some of the key designs are outlined in the next section.

22.4 Types of Evaluation Design Evaluation designs can be divided into three categories, each described in the next sections: 1. Experimental 2. Quasi-experimental 3. Nonexperimental

Experimental and quasi-experimental designs can be used to evaluate behavior change interventions. Nonexperimental designs typically generate post hoc data about an intervention but, as they do not involve manipulation of a variable, they do not identify causal effects. Given the large number of available evaluation designs, only a few of the commonly used designs will be described as well as some relatively new designs that offer promise. Examples of some of these designs are provided in Table 22.1. References for further reading are provided for all other designs mentioned.

Table 22.1 Examples of experimental and quasi-experimental evaluation designs • • • • • • • • • • •

Individually randomized controlled trialsa (Torgerson & Torgerson, 2008) Cluster randomized trialsa (Hemming et al., 2017) Stepped wedge designsa (Hemming et al., 2015) Preference trial designsa (Torgerson & Sibbald, 1998) N-of-1 designsa (McDonald et al., 2017) Adaptive designs (Kairalla et al., 2012) Factorial designs (Collins, Dziak, & Li, 2009) Crossover/repeated measures designs (Kraska, 2010) Interrupted time series designsa (Bernal, Cummins, & Gasparrini, 2017); Natural experimentsa (Craig et al., 2017) Regression discontinuity design (O’Keeffe et al., 2014)

Note. aDesigns elaborated on in the current chapter.

321

322

LYNSAY MATTHEWS AND SHARON A. SIMPSON

22.4.1 Experimental Designs Experimental designs are characterized by the random allocation of participants to different conditions; control of all variables with the exception of the one being tested (the “experimental manipulation”); and the ability of the design to be replicated by other researchers. Typically, this involves one or more groups being assigned to an intervention condition, while one group is assigned as a control. The intervention group(s) receives the experimental manipulation or intervention, while the control group receives comparator content, such as a placebo, standard treatment, or remains on an intervention waiting list. Their strengths include offering the ability to draw inferences about causality; reduced bias; higher levels of internal validity (i.e., the observed effect is a result of the intervention and not another factor); and replicable research designs. Their limitations typically include lower external validity (i.e., the ability to generalize the results to other populations or settings); lack of contextual insight and limited theorizing as to how mechanisms interact with context; inability to account for the complexity needed to evaluate systems or policy-level interventions (Sanson-Fisher et al., 2014); and that they are resource-intensive (Deaton & Cartwright, 2018). Although designs like RCTs can lack “real-world” applicability, it should be noted that “realist randomized controlled trials,” as proposed by Bonell et al. (2012), address many of these issues. Examples of experimental designs are outlined in the next section. SansonFisher et al. (2014) provide a useful summary of alternative designs to a standard RCT, based on trade-offs related to practicality, validity, and ethical considerations.

22.4.1.1 Individually Randomized Controlled Trials Individuals are randomly allocated to two or more groups: intervention and control. Outcomes are then compared across the groups. Example:

Parenting intervention in Sure Start services (Hutchings et al., 2007).

22.4.1.2 Cluster Randomized Trials Participants are randomized as groups rather than individuals – for example all participants in school A are randomized to the intervention group and all in school B to a control group. This reduces the risk of intervention contamination caused by, for instance, participants across each condition sharing intervention material and minimizes bias in estimates of effect size (Hemming et al., 2017). Experts advise using individual allocation where possible due to potential biases associated with poorly designed cluster RCTs, for instance selection bias (Hahn et al., 2005). Example: Effects of the Learning Together INCLUSIVE intervention (Bonell et al., 2018).

22.4.1.3 Stepped Wedge Designs Although this design can be used for an individual or cluster RCT, it is most often used in cluster RCTs. In a cluster stepped wedge design, at the beginning of the evaluation no clusters are exposed to the intervention and data are collected during this baseline period. At regular intervals, clusters are randomized to receive the intervention. Data are collected at these points, until all clusters have received the intervention, followed by a final data collection. In this design, the baseline period data of individuals or clusters represent the control arm. This may be used to overcome practical or ethical considerations related to withholding the intervention in the control group for policy and/or health service–led interventions (Hemming et al., 2015). However, this design increases the complexity of analyses and participant burden and is more costly. Example: Study protocol for a substance abuse lifestyle intervention (Kelly et al., 2015).

22.4.1.4 Preference Trial Designs This approach may be appropriate for interventions where strong participant preferences for an intervention may bias the study via refusal to be

Evaluation of Behavior Change Interventions

randomized or resentful demoralization and dropout after being randomized to their nonpreferred group. Preference trials result in four groups: an intervention and control randomized as normal, in addition to an intervention and control allocated by participant preference (Torgerson & Sibbald, 1998). Although preference trials have the advantage of minimizing treatment bias of participant preferences, they increase the size and cost of the evaluation. Example: An internet-based/delivered stop-smoking preference trial (Schueller et al., 2013).

22.4.1.5

N-of-1 Designs

An N-of-1 trial is typically where a single participant is the unit of observation. It can also, however, be one cluster (e.g., an organization or a family). It is a randomized study where the efficacy or side effects of different interventions are tested. This can be advantageous for behavior change interventions where outcomes of the intervention are likely to vary widely between

323

participants. In an N-of-1 study, the participant receives different interventions over time, they act as their own control, the outcomes can be measured repeatedly, and the effects of the intervention can be reversed (Davidson et al., 2014). They can also provide information about the mechanisms by which an intervention operates (Kwasnicka et al., 2019). Findings from multiple N-of-1 studies can be combined to advance the evidence base. They should not be a replacement for traditional RCTs due to issues of generalizability. However, they can be beneficial for participants who have not responded as expected to previous evidence-based interventions. Example: Using behavioral analytics to increase exercise (Yoon et al., 2018). See Sidebar 22.1 for an additional example.

22.4.2 Quasi-experimental Designs A quasi-experimental design is one that resembles an experimental design but lacks the key

Sidebar 22.1 Case study: Personalizing behavioral interventions through single-patient (N-of-1) trials (Davidson et al., 2014)

The case study. A seven-year-old, underweight boy, diagnosed with attention deficit hyperactivity disorder, has been treated with medication as guided by the evidence base. He demonstrated no clear improvement and continued to exhibit harmful behavioral symptoms. After researching the evidence base, and finding no clear guidance on next treatment steps, a range of alternative interventions were given to the individual based on his unique circumstances. These ranged from increased dosage of medication to behavioral and parenting interventions. The individual’s progress on one intervention (e.g., change in medication dose) was monitored and, in the instance of nonimprovement, the intervention was stopped and replaced with an alternative intervention (e.g., behavioral intervention). Key point. This example illustrates the advantage of the N-of-1 approach for an individual who has not responded as expected to interventions based on evidence from traditional RCTs. In this scenario, it allowed evaluation of a series of interventions targeted to one individual, based on their specific history, characteristics, and treatment effects to date, until one intervention demonstrated improved outcomes.

324

LYNSAY MATTHEWS AND SHARON A. SIMPSON

element of randomization. Although the main variable is manipulated in an experimental fashion, the lack of randomization makes it difficult to reliably eliminate the impact of confounding variables. Quasi-experimental designs lie on the continuum of internal validity between experimental and true nonexperimental designs. They are typically used in circumstances where randomization is not possible, for example limited time and resources, where randomization is unethical, or where practitioners are not in equipoise regarding the treatments being tested and are therefore unwilling to randomize patients (Fairhurst & Dowrick, 1996).

22.4.2.1 Interrupted Time Series Design This design uses longitudinal data to evaluate an intervention effect (Kontopantelis et al., 2015). It is characterized by multiple observations gathered over time from participants, both before and after they have received an intervention. The time series is interrupted by an intervention at one or more time points. This allows a population to act as their own control, allowing for patterns of change to be detected pre- and postintervention. The multiple “time series” measurements help to increase statistical power by improving internal validity as opposed to a pretest-posttest design, which has a single pre- and postmeasurement. This approach can be useful for populations where the potential sample size is small, such as rural communities or minority ethnic groups. Similar to the stepped wedge design, it has the advantage that every individual receives the intervention rather than being allocated to a waiting list or control group. The staggered start times also help alleviate logistical difficulties of small research teams implementing interventions in rural or multiple locations (Fok, Henry, & Allen, 2015). Example: Three behavioral economics “nudges” on grocery and convenience store sales of promoted nutritious foods (Chapman et al., 2019).

22.4.2.2 Natural Experiments Natural experiments generally evaluate interventions that are not developed by a researcher, where a population is split into exposed and unexposed groups (Craig et al., 2012). They have been growing in popularity for research where experimental manipulation is not suitable. For example, they can be particularly useful for evaluating policy-led interventions (Barnighausen et al., 2017; Basu, Meghani, & Siddiqi, 2017; MacMillan et al., 2018). Ideally, natural experiments should produce a process of exposure that is similar to randomization. This is a challenge in policy and practice but one that is addressed in the MRC guidance on natural experiments (Craig et al., 2012). An example is Prins et al.’s (2016) natural experiment examining behavior change in the context of new transport infrastructure and cycling to work. See Sidebar 22.2 for an additional example of a natural experiment.

22.4.3 Nonexperimental Approaches to Evaluation Nonexperimental designs should not be used alone to evaluate behavior change interventions. They can, however, gather helpful data to support understanding of the main evaluation findings (e.g., qualitative methods; Salkind, 2010) or predicting behavior change using modeling approaches (e.g., agent-based modeling).

22.4.3.1 Qualitative Methods This approach to evaluation is often used alongside experimental and quasi-experimental designs to gather additional data to help understand the process, mechanism, and outcomes of the intervention as well as explore acceptability of evaluation methods. Qualitative methods, such as interviews, focus groups, and observations, can be useful for behavior change research by exploring people’s experience of barriers and facilitators to change, as well as mechanisms of impact and the interaction of intervention elements with contextual factors. They can also help identify

Evaluation of Behavior Change Interventions

325

Sidebar 22.2 How do individuals’ health behaviors respond to an increase in the supply of health care? Evidence from a natural experiment (Fichera, Gray, & Sutton, 2016)

The case study. This study assessed how individuals’ health behavior responded to the introduction of the quality and outcomes framework in England (Forbes et al., 2017). It found that an improved supply of health care, whereby health providers received incentives, had positive effects of the lifestyle behaviors of individuals with targeted health conditions. Key point. A strength of this natural experiment was the ability to explore associations between human behavior and a policy change affecting health care. The use of the natural experiment approach in this case study therefore allowed the intervention to be evaluated within the context to which it was intended.

modifications needed to evaluation methods as well as intervention content and/or delivery. A good example of this is Desveaux and Shaw’s (2018) qualitative realist evaluation of their mobile app to improve self-management of Type 2 diabetes (see also Chapter 30, this volume).

22.4.3.2 Modeling Designs This approach models the impact on behavior of different influences, interventions, or policies. An example of this approach can be seen in Yang and Diez-Roux (2013), where an agent-based model simulated decisions about whether children would walk to school based on perceptions of safety and distance.

22.5 Process Evaluation and Mixed Methods Process evaluation is an important part of many types of evaluation. It can provide information on fidelity, reach, retention, mechanisms, and important contextual factors, as well as insights into why an intervention has or has not been successful. G. F. Moore et al. (2015) provide a framework and guidance on conducting process evaluation, from the planning stage through design, analysis, and reporting. Within an evaluation, or a process evaluation, quantitative and qualitative methods can be

combined in a mixed methods approach. This can provide important insights beyond those obtained using one method alone and can facilitate corroboration of quantitative or qualitative findings, providing clarification or explanation to aid in interpreting results. Quantitative data collected from experiments, trials, or surveys and qualitative data from focus groups, interviews, or observations are combined and analyzed to help address the research questions (O’Cathain, Murphy, & Nicholl, 2010). RCTs can be strengthened by the use of mixed methods at the development stage (Chapter 21, this volume), early on in the trial, and also when interpreting the findings. An example is provided by the HelpMeDoIt! study (Matthews et al., 2017; Simpson et al., 2019).

22.6 Choosing an Evaluation Design The choice of evaluation design should be based on the most appropriate method to answer the research questions posed by the researcher or interventionist. As set out at the beginning of this chapter, evaluators should spend time consulting with stakeholders to identify the purpose of the research and the research questions (see Chapter 24, this volume). Careful consideration of different designs and their pros and cons is required in the development stage of the research.

326

LYNSAY MATTHEWS AND SHARON A. SIMPSON

Table 22.2 Issues to consider when evaluating a behavior change intervention Issues

Description

Stakeholders

Gather input from relevant stakeholders on what questions or outcomes are of interest to them in order to collect useful data to inform future policy and practice decisions. Also consider whether the intervention will fit within the context or system in which it is to be delivered. Further details on involving stakeholders are provided in Chapter 24, this volume. Develop a program theory capturing intervention prerequisites, inputs, mechanisms, contextual dependencies, and outcomes as well as unintended consequences. Further details on program theory/logic model/theory of change are provided in Chapters 18, 19, and 21, this volume. Identify aspects of the context that (1) influence the intervention; (2) are influenced by the intervention; and/or (3) influence the evaluation. Identifying important contextual aspects in the evaluation will be useful when implementing the intervention in the “real world.” Consider involving an economist at an early stage. Further details on economic evaluation are provided in Chapter 26, this volume. Improve likelihood of successful implementation by addressing the issues already mentioned above, as well as (1) considering issues of implementation at the earliest stages of research; (2) making use of models, frameworks, and theories of implementation; (3) learning from the success/failure of similar interventions; and (4) proactively engaging decision makers. Further details on implementation are provided in Chapter 23, this volume.

Program theory

Context

Economic considerations Implementation

22.6.1 Key Issues to Consider When Undertaking an Evaluation There are a number of key issues to consider when designing and undertaking an evaluation. Thinking about these issues before and throughout the evaluation process will improve the quality of the evaluation design and potential usefulness of the findings. Key issues include (1) involving relevant stakeholders; (2) developing and updating program theory; (3) considering the wider context (and system); (4) including appropriate economics input; and (5) addressing issues of implementation. Table 22.2 outlines these issues further.

22.7 Choice of Evaluation Outcomes A development in the field of intervention evaluation has been a shift in thinking toward assessing the “usefulness” of findings generated from

the evaluation, as opposed to the standard use of outcomes that focus on effect size (Leviton & Melichar, 2016; Public Health England, 2018). Understanding the change process resulting from an intervention and producing data that are both useful and meaningful for stakeholders are important. Evaluating “What happened?” (process evaluation) versus “Did it work?” (outcome evaluation) may sometimes be more appropriate, particularly if the “Did it work?” question has been prioritized too early. For example, data may include real-world outcomes, elements of refined theory, or information on the mechanisms, context, and processes of the intervention. This type of data is useful for understanding and refining interventions for further evaluation or implementation. Overall, the outcome of evaluation should be the collation of useful data that inform future iteration and implementation. A summary

Evaluation of Behavior Change Interventions

327

Sidebar 22.3 Summary of key points for choosing evaluation outcomes • • • • • • •

Engage with policy, practice, and patients/the public in order to produce something that is useful to these stakeholders. Include elements of economic evaluation; consider modeling approaches. Choose appropriate measures, giving consideration to subjective versus objective measures, validity, and reliability. Measure long-term behavior change where feasible. Think about proximal as well as distal outcomes based on the program theory of the intervention. Use process evaluation to understand what has happened and explore contextual factors. Refine the program theory as a study outcome.

of key points related to evaluation outcomes is presented in Sidebar 22.3.

22.7.1 General Considerations Regarding Outcomes and Measures in Full Trial Evaluations Consideration should be given to which outcomes are primary and secondary and how these will be analyzed. Although a single primary outcome, and a small number of secondary outcomes, may be the most straightforward from the point of view of statistical analysis, it may not represent the best use of the data and it could miss important outcomes. A good theoretical understanding of the intervention, articulated in the program theory, is key to choosing suitable outcomes. As mentioned in Table 22.2, the program theory should also consider the potential mechanisms, processes, outcomes, and influences across a range of domains. Stakeholders should be involved in this process (see also Chapter 24, this volume). Measures should be appropriate to the design of the evaluation – for example, subjective or self-report outcomes may be unreliable due to response bias. The length of follow-up should be chosen to align with the proposed timeline of

behavior change – for example, if the program theory suggests a change in behavior at three months, it will be unhelpful to collect followup data at two months. Additional sources of variation in outcomes should be considered for subgroup analyses, such as analyses by demographic variables (e.g., age, gender, socioeconomic status), if statistical power allows for this, and such subgroup analyses and hypotheses should be prespecified in a protocol paper or statistical analyses plan developed prior to analyses commencing. A potential outcome of evaluation is refined program theory. Improved theory or theory development or refinement could be the main outcome of an evaluation. Improved theory will help inform the transferability of an intervention across settings and produce data that are relevant for informing policy decisions.

22.7.2 Challenges of Measuring Behavior Change Human behavior is complex and measuring behavior change presents a challenge to those evaluating behavior change interventions. A number of key issues need to be considered, including the practicalities of choosing between subjective and objective measures,

328

LYNSAY MATTHEWS AND SHARON A. SIMPSON

difficulties in measuring the long-term maintenance of behavior change, measuring population-level benefits, and measuring system change. These will be outlined in the following sections.

22.7.2.1 Practicalities of Choosing Between Subjective and O bjective Measures Objective measures (e.g., observation of behavior) are the least biased means to evaluate behavior change. However, in many cases, objective measures are not possible to administer due to practical reasons, cost, or because an objective measure does not exist. In addition, the complexity of behavior change may not be fully captured if only objective measures are used. By contrast, subjective measures like questionnaires often have limitations with validity and reliability and are also open to biases like social desirability bias (Paulhus, 1984).

22.7.2.2 Difficulty Measuring Long-Term Maintenance of Behavior Change It is typically impractical to have studies of many years in duration due to the cost and dropout of participants, which will introduce bias in the analyses. The longest feasible follow-up timescale should be considered alongside practical constraints such as funding, staffing, and other resources. Modeling approaches can be used to estimate longer-term effectiveness and costeffectiveness.

Modeling can also be a useful approach to evaluate intervention effects if introduced at the population level.

22.7.2.4 Measuring System Change Interventions for behavior change may influence and be influenced by the wider system. For example, an intervention designed to impact behaviors in school pupils may create change in other areas of the school, such as staff morale and rates of pupil disciplinary referrals and absenteeism. These changes may, in turn, feedback and influence further behavior change. These changes may be missed if they are not included in the program theory and measured. Egan and colleagues have published guidance on how to measure change in a system (Egan, McGill, Anderson de Cuevas, et al., 2019; Egan, McGill, Penney et al., 2019).

22.8 Implications for Research, Practice, and Translation Based on the current available evidence, the following recommendations are proposed for the evaluation of interventions in research, practice, and for translation: •

22.7.2.3 Measuring Population-Level Benefits Population approaches to change behavior can often have quite small changes at an individual level or can have small changes across numerous outcomes that can still lead to population-level benefits. Evaluations powered on relatively large changes on a primary outcome, or which only assess one or two primary outcomes, may miss these effects. Program theory should, therefore, be used to determine the most appropriate outcomes to measure across a range of domains.

•

•

Considering the following key issues before and throughout an evaluation will improve its potential impact: (1) involving relevant stakeholders; (2) developing and updating program theory; (3) considering the wider context (and system); (4) including appropriate economics input; and (5) addressing issues of implementation. Clarifying the purpose of the evaluation should guide the choice of evaluation design. Considering alternative evaluation designs, based on the best method for addressing the research questions, can positively impact on the generation of “useful” data and reduce research waste. Choosing evaluation methods appropriate to the stage of evaluation (e.g., feasibility stage vs. full evaluation) can improve the usefulness of the data.

Evaluation of Behavior Change Interventions

•

Choosing outcome and process measures covering multiple aspects of the program theory ensures that benefits (or harms) are not missed.

22.9 Summary and Conclusion The most appropriate evaluation design depends on the research questions. This chapter provided an overview of four evaluation approaches, three categories of evaluation designs, and five key issues to consider throughout the evaluation process, as well as specific issues related to outcome measurement. The chapter focused on highlighting essential components to be considered in designing and implementing an evaluation and serves as a starting point for further reading.

References Barnighausen, T., Tugwell, P., Rottingen, J. A. et al. (2017). Quasi-experimental study designs seriespaper 4: Uses and value. Journal of Clinical Epidemiology, 89, 21–29. https://doi.org/10.1016/ j.jclinepi.2017.03.012 Basu, S., Meghani, A., & Siddiqi, A. (2017). Evaluating the health impact of large-scale public policy changes: Classical and novel approaches. Annual Review of Public Health, 38, 351–370. https://doi.org/10.1146/annurev-publhealth031816-044208 Bernal, J. L., Cummins, S., & Gasparrini, A. (2017). Interrupted time series regression for the evaluation of public health interventions: A tutorial. International Journal of Epidemiology, 46, 348–355. https://doi.org/10.1093/ije/dyw098 Bonell, C., Allen, E., Warren, E. et al. (2018). Effects of the Learning Together intervention on bullying and aggression in English secondary schools (INCLUSIVE): A cluster randomised controlled trial. The Lancet, 392(10163), 2452–2464. https:// doi.org/10.1016/S0140-6736(18)31782-3 Bonell, C., Fletcher, A., Morton, M., Lorenc, T., & Moore, L. (2012). Realist randomised controlled trials: A new approach to evaluating complex public health interventions. Social Science and

329

Medicine, 75, 2299–2306. https://doi.org/ 10.1016/j.socscimed.2012.08.032 Bonell, C., Jamal, F., Melendez-Torres, G. J., & Cummins, S. (2015). “Dark logic”: Theorising the harmful consequences of public health interventions. Journal of Epidemiology and Community Health, 69, 95–98. https://doi.org/ 10.1136/jech-2014-204671 Chapman, L., Sadeghzadeh, C., Koutlas, M., Zimmer, C., & Marco, M. D. (2019). Evaluation of three behavioral economics “nudges” on grocery and convenience store sales of promoted nutritious foods (OR16-05–19). Public Health Nutrition, 22, 3250–3260. https://doi.org/10.1017/S13689800 1900179 Collins, L. M., Dziak, J. J., & Li, R. (2009). Design of experiments with multiple independent variables: A resource management perspective on complete and reduced factorial designs. Psychological Methods, 14, 202–224. https://doi.org/10.1037/ a0015826 Cook, J. A., Julious, S. A., Sones, W. et al. (2018). DELTA2 guidance on choosing the target difference and undertaking and reporting the sample size calculation for a randomised controlled trial. British Medical Journal, 363. https://doi.org/10.1136/bmj.k3750 Craig, P., Cooper, C., Gunnell, D. et al. (2012). Using natural experiments to evaluate population health interventions: New Medical Research Council guidance. Journal of Epidemiology and Community Health, 66, 1182–1186. https://doi .org/10.1136/jech-2011-200375 Craig, P., Di Ruggiero, E., Frolich, K. L., Mykhalovskiy, E., White, M., & the Canadian Institutes of Health Research (CIHR)–National Institute for Health Research (NIHR) Context Guidance Authors Group. (2018). Taking Account of Context in Population Health Intervention Research: Guidance for Producers, Users and Funders of Research. Southampton: NIHR Journals Library. Craig, P., Katikireddi, S. V., Leyland, A., & Popham, F. (2017). Natural experiments: An overview of methods, approaches, and contributions to public health intervention research. Annual Review of Public Health, 38, 39–56. https://doi.org/10.1146/ annurev-publhealth-031816-044327

330

LYNSAY MATTHEWS AND SHARON A. SIMPSON

Davidson, K. W., Peacock, J., Kronish, I. M., & Edmondson, D. (2014). Personalizing behavioral interventions through single-patient (n-of-1) trials. Social and Personality Psychology Compass, 8, 408–421. https://doi.org/10.1111/spc3.12121 Deaton, A., & Cartwright, N. (2018). Understanding and misunderstanding randomized controlled trials. Social Science and Medicine, 210, 2–21. https://doi.org/10.1016/j.socscimed.2017.12.005 Desveaux, L., & Shaw, J. (2018). A mobile app to improve self-management of individuals with type 2 diabetes: Qualitative realist evaluation. Journal of Medical Internet Research, 20, e81. https://doi.org/10.2196/jmir.8712 Egan, M., McGill, E., Anderson de Cuevas, R. et al. (2019). NIHR SPHR Guidance on Systems Approaches to Local Public Health Evaluation. Part 1: Introducing Systems Thinking. London: National Institute for Health Research. Egan, M., McGill, E., Penney, T. et al. (2019). NIHR SPHR Guidance on Systems Approaches to Local Public Health Evaluation. Part 2: What to Consider When Planning a Systems Evaluation. London: National Institute for Health Research. Fairhurst, K., & Dowrick, C. (1996). Problems with recruitment in a randomized controlled trial of counselling in general practice: Causes and implications. Journal of Health Services Research and Policy, 1, 77–80. https://doi.org/10.1177/ 135581969600100205 Fichera, E., Gray, E., & Sutton, M. (2016). How do individuals’ health behaviours respond to an increase in the supply of health care? Evidence from a natural experiment. Social Science and Medicine, 159, 170–179. https://doi.org/10.1016/ j.socscimed.2016.05.005 Fok, C. C., Henry, D., & Allen, J. (2015). Research designs for intervention research with small samples II: Stepped wedge and interrupted timeseries designs. Prevention Science, 16, 967–977. https://doi.org/10.1007/s11121-015-0569-4 Forbes, L. J., Marchand, C., Doran, T., & Peckham, S. (2017). The role of the Quality and Outcomes Framework in the care of long-term conditions: A systematic review. British Journal of General Practice, 67, e775–e784. https://doi.org/10.3399/ bjgp17X693077

Glasgow, R. E., Harden, S. M., Gaglio, B. et al. (2019). RE-AIM planning and evaluation framework: Adapting to new science and practice with a 20year review. Frontiers in Public Health, 7, 64. https://doi.org/10.3389/fpubh.2019.00064 Hahn, S., Puffer, S., Torgerson, D. J., & Watson, J. (2005). Methodological bias in cluster randomised trials. BMC Medical Research Methodology, 5, 10. https://doi.org/10.1186/1471-2288-5-10 Hallingberg, B., Turley, R., Segrott, J. et al. (2018). Exploratory studies to decide whether and how to proceed with full-scale evaluations of public health interventions: A systematic review of guidance. Pilot Feasibility Studies, 4, 104. https:// doi.org/10.1186/s40814-018-0290-8 Hemming, K., Eldridge, S., Forbes, G., Weijer, C., & Taljaard, M. (2017). How to design efficient cluster randomised trials. British Medical Journal, 358, j3064. https://doi.org/10.1136/bmj.j3064 Hemming, K., Haines, T. P., Chilton, P. J., Girling, A. J., & Lilford, R. J. (2015). The stepped wedge cluster randomised trial: Rationale, design, analysis, and reporting. British Medical Journal, 350, h391. https://doi.org/10.1136/bmj.h391 Hutchings, J., Bywater, T., Daley, D. et al. (2007). Parenting intervention in Sure Start services for children at risk of developing conduct disorder: Pragmatic randomised controlled trial. British Medical Journal, 334(7595), 678. https://doi.org/ 10.1136/bmj.39126.620799.55 Kairalla, J. A., Coffey, C. S., Thomann, M. A., & Muller, K. E. (2012). Adaptive trial designs: A review of barriers and opportunities. Trials, 13, 145. https://doi.org/10.1186/1745-6215-13-145 Kelly, P. J., Baker, A. L., Deane, F. P. et al. (2015). Study protocol: A stepped wedge cluster randomised controlled trial of a healthy lifestyle intervention for people attending residential substance abuse treatment. BMC Public Health, 15, 465. https://doi.org/10.1186/s12889-0151729-y Kontopantelis, E., Doran, T., Springate, D. A., Buchan, I., & Reeves, D. (2015). Regression based quasiexperimental approach when randomisation is not an option: Interrupted time series analysis. British Medical Journal, 350, h2750. https://doi.org/ 10.1136/bmj.h2750

Evaluation of Behavior Change Interventions

Kraska, M. (2010). Repeated measures design. In N. J. Salkind (Ed.), Encyclopedia of Research Design. Thousand Oaks, CA: SAGE. Kwasnicka, D., Inauen, J., Nieuwenboom, W. et al. (2019). Challenges and solutions for N-of-1 design studies in health psychology. Health Psychology Review, 13, 163–178. https://doi.org/ 10.1080/17437199.2018.1564627 Leviton, L. C., & Melichar, L. (2016). Balancing stakeholder needs in the evaluation of healthcare quality improvement. BMJ Quality and Safety, 25, 803–807. https://doi.org/10.1136/bmjqs-2015004814 MacMillan, F., George, E. S., Feng, X. et al. (2018). Do natural experiments of changes in neighborhood built environment impact physical activity and diet? A systematic review. International Journal of Environmental Research and Public Health, 15, 217. https://doi.org/10.3390/ijerph15020217 Matthews, L., Pugmire, J., Moore, L. et al. (2017). Study protocol for the “HelpMeDoIt!” randomised controlled feasibility trial: An app, web and social support-based weight loss intervention for adults with obesity. BMJ Open, 7, e017159. https://doi.org/10.1136/bmjopen-2017017159 McDonald, S., Quinn, F., Vieira, R. et al. (2017). The state of the art and future opportunities for using longitudinal n-of-1 methods in health behaviour research: A systematic literature overview. Health Psychology Review, 11, 307–323. https://doi.org/ 10.1080/17437199.2017.1316672 Moore, G. F., Audrey, S., Barker, M. et al. (2015). Process evaluation of complex interventions: Medical Research Council guidance. British Medical Journal, 350(h1258), h1258. https://doi .org/10.1136/bmj.h1258 Moore, L., Hallingberg, B., Wight, D. et al. (2018). Exploratory studies to inform full-scale evaluations of complex public health interventions: The need for guidance. Journal of Epidemiology and Community Health, 72, 865– 866. https://doi.org/10.1136/jech-2017-210414 Murad, M. H., Asi, N., Alsawas, M., & Alahdab, F. (2016). New evidence pyramid. BMJ EvidenceBased Medicine, 21, 125–127. https://doi.org/ 10.1136/ebmed-2016-110401

331

O’Cathain, A., Croot, L., Duncan, E. et al. (2019). Guidance on how to develop complex interventions to improve health and healthcare. BMJ Open, 9, e029954. https://doi.org/10.1136/ bmjopen-2019-029954 O’Cathain, A., Murphy, E., & Nicholl, J. (2010). Three techniques for integrating data in mixed methods studies. British Medical Journal, 341, c4587. https://doi.org/10.1136/bmj.c4587 O’Keeffe, A. G., Geneletti, S., Baio, G., Sharples, L. D., Nazareth, I., & Petersen, I. (2014). Regression discontinuity designs: An approach to the evaluation of treatment efficacy in primary care using observational data. British Medical Journal, 349, g5293. https://doi.org/10.1136/bmj.g5293 Paulhus, D. L. (1984). Two-component models of socially desirable responding. Journal of Personality and Social Psychology, 46, 598. https://doi.org/10.1037/0022-3514.46.3.598 Prins, R. G., Panter, J., Heinen, E., Griffin, S. J., & Ogilvie, D. B. (2016). Causal pathways linking environmental change with health behaviour change: Natural experimental study of new transport infrastructure and cycling to work. Preventive Medicine, 87, 175–182. https://doi.org/ 10.1016/j.ypmed.2016.02.042 Public Health England. (2018). Evaluation in Health and Wellbeing. London: Public Health England. Salkind, N. J. (2010). Encyclopedia of Research Design. Thousand Oaks, CA: SAGE. Sanson-Fisher, R. W., D’Este, C. A., Carey, M. L., Noble, N., & Paul, C. L. (2014). Evaluation of systems-oriented public health interventions: Alternative research designs. Annual Review of Public Health, 35, 9–27. https://doi.org/10.1146/ annurev-publhealth-032013-182445 Schueller, S. M., Leykin, Y., Pérez-Stable, E. J., & Muñoz, R. F. (2013). Selection of intervention components in an internet stop smoking participant preference trial: Beyond randomized controlled trials. Psychiatry Research, 205, 159–164. https:// doi.org/10.1016/j.psychres.2012.08.030 Simpson, S. A., Matthews, L., Pugmire, J. et al. (2019). An app, web and social support based weight loss intervention for adults with obesity: The “HelpMeDoIt!” feasibility RCT. Public Health Research.

332

LYNSAY MATTHEWS AND SHARON A. SIMPSON

Skivington, K., Matthews, L., Craig, P., Simpson, S., & Moore, L. (2018). Developing and evaluating complex interventions: Updating Medical Research Council guidance to take account of new methodological and theoretical approaches. The Lancet, 392, S2. Meeting abstract: Public Health Science 2018, Belfast, Northern Ireland, November 23, 2018. http://dx.doi.org/10.1016/ s0140-6736(18)32865-4 Torgerson, C., & Torgerson, D. J. (2008). Designing Randomised Trials in Health, Education and the Social Sciences: An Introduction. London: Palgrave Macmillan. Torgerson, D. J., & Sibbald, B. (1998). Understanding controlled trials: What is a patient preference trial? British Medical Journal, 316, 360. https://doi.org/ 10.1136/bmj.316.7128.360

Walton, H., Spector, A., Tombor, I., & Michie, S. (2017). Measures of fidelity of delivery of, and engagement with, complex, face-to-face health behaviour change interventions: A systematic review of measure quality. British Journal of Health Psychology, 22, 872–903. https://doi.org/ 10.1111/bjhp.12260 Yang, Y., & Diez-Roux, A. V. (2013). Using an agentbased model to simulate children’s active travel to school. International Journal of Behavioral Nutrition and Physical Activity, 10, 67. https://doi .org/10.1186/1479-5868-10-67 Yoon, S., Schwartz, J. E., Burg, M. M. et al. (2018). Using behavioral analytics to increase exercise: A randomized n-of-1 study. American Journal of Preventive Medicine, 54, 559–567.

23 Implementation Science and Translation in Behavior Change Aleksandra Luszczynska, Karolina Lobczowska, and Karolina Horodyska

Practical Summary When planning, delivering, and reporting behavior change interventions, the main focus is usually on the efficacy of the intervention in affecting behavior change and on elucidating mechanisms responsible for the effects of the intervention on behavior, often informed by behavior change theories. However, to improve the transferability of the intervention from a laboratory setting to “real-world” contexts, planning, delivering, and reporting of interventions should also account for advances in implementation science. In general, implementation science is a discipline aimed at studying and informing the development and investigation of methods to promote the systematic uptake of research findings and other evidence-based practices into realworld settings and routine practice. This chapter outlines the theories, models, and frameworks used in the implementation of behavior change interventions. Implementation processes, implementation determinants (barriers and facilitators), the evaluation of implementation processes, and implementation outcomes are discussed. Finally, the chapter introduces an example of a translation model aimed at explaining the translation of behavioral interventions on a continuum of research approaches, spanning studies on basic processes and mechanisms (e.g., laboratory experiments) to practicebased research, aiming at improved applications in the clinical practice and greater adoption of an intervention across target populations, contexts, and settings.

23.1 Introduction Behavior change interventions will only be likely to result in practically significant behavior change in “real-world” contexts if they are implemented with careful attention to implementation theories, models, or frameworks and evidence-based bestpractice guidelines (Ory, Jordan, & Bazzarre, 2002). However, research on behavior change interventions has hitherto had a heavy emphasis on efficacy (see also Chapter 22, this volume), with insufficient attention paid to implementation (see also Hagger & Weed, 2019). This resulted in an

“implementation gap,” which refers to the hiatus between what is known as efficacious in controlled studies, often conducted in academic institutions or research institutes, and what is delivered to the target population for evidence-based practice in a specific setting (Proctor et al., 2009). New interventions are estimated to “languish” for between fifteen and twenty years before they are incorporated into the standard practice or usual care (see Proctor et al., 2009). Implementation science aims to close this gap through the development of https://doi.org/10.1017/9781108677318.023

334

ALEKSANDRA LUSZCZYNSKA, KAROLINA LOBCZOWSKA, AND KAROLINA HORODYSKA

implementation theories as well as the accumulation and synthesis of research evidence. Implementation science is an innovative discipline aimed at studying and informing the development and investigation of methods to promote the systematic uptake of research findings, and other evidence-based practices, into routine

practice. Evidence derived from implementation science can be used to inform intervention development, implementation, and evaluation to improve the quality and effectiveness of behavior change interventions and policies targeting individual, organizational, and societal change (Eccles, & Mittman, 2006). Implementation is

Sidebar 23.1 Key constructs used in implementation science

Implementation science is a discipline aimed at studying and informing the development and investigation of methods to promote the systematic uptake of research findings and other evidence-based practices into real-world contexts and routine practice (Greenhalgh et al., 2005; Proctor et al., 2009; Rabin & Brownson, 2012). Implementation is a social process through which interventions are operationalized within an organization or community (see the context and implementation of complex interventions framework; Pfadenhauer et al., 2017). The process involves implementation actors (e.g., teachers delivering the intervention, managers or principals who organize resources for the intervention) and a specific implementation setting (e.g., characteristics of a school where the implementation takes place, such as the availability of equipment or built facilities necessary to conduct an intervention) (Pfadenhauer et al., 2017). The implementation process involves the diagnostic of needs; the readiness of the target population and implementation agents for the intervention; building resources and capacity in the target setting; planning and initiation of an implementation; and adoption of the intervention by the target population, its setting, and implementation actors. The implementation process involves implementation strategies, that is, the methods and means to ensure adoption and sustained delivery of an intervention (Pfadenhauer et al., 2017). The strategies are tailored to the implementation actors and setting. For example, to build and maintain the awareness of the intervention, school principals may add information about the intervention to the school website or send regular messages about the intervention to parents and students. The implementation process is accompanied by the careful monitoring of fidelity and adoptions made. Implementation monitoring and evaluation accounts for implementation outcomes (e.g., levels of acceptability of the intervention by students, teachers, and parents; costs of implementation covered by a school). The implementation process takes place in the implementation context, which includes several key domains: geographical, epidemiological, sociocultural, socioeconomic, ethical, legal, and political (Pfadenhauer et al., 2017). For example, the geographical context may refer to the accessibility of an intervention to students who live in geographically isolated areas, a long distance from the school where the intervention takes place. The characteristics of the context, setting, and implementation actors may constitute

Implementation Science and Translation in Behavior Change

335

barriers or facilitators of the implementation process and therefore influence implementation outcomes (Nilsen, 2015). For example, teachers’ knowledge and beliefs about the content of the intervention may hinder or prompt successful implementation. Implementation may be planned as a linear process but, in real-world contexts and settings, it may be also explained with complex system approaches, assuming multiple feedback loops, interactions, and constant adaptations (Rutter et al., 2017). The characteristics of the planned implementation processes constantly interact and form feedback loops with the characteristics of actors, setting, implementation strategies, evaluation and monitoring efforts, and the domains of the context (Pfadenhauer et al., 2017; Rutter et al., 2017). Those implementing interventions to change behavior must, therefore, be aware of the potential factors and the processes involved and account for them when designing their implementation.

part of a diffusion-dissemination-implementation continuum (see Sidebar 23.1). Diffusion is defined as the passive, untargeted, and unplanned spread of new practices, whereas dissemination is defined as the active spread of new practices to the target audience using planned strategies. Implementation is defined as the process of putting to use or integrating new practices within a setting (Greenhalgh et al., 2005; Rabin & Brownson, 2012). When designing and delivering a behavior change intervention, researchers and practitioners may focus mainly on the extent to which the intervention leads to change in the key behavioral outcome of interest. This is often evaluated through efficacy trials (Bauer et al., 2015), which focus on the systematic evaluation of intervention effects on behavior and associated outcomes of interest (see Chapter 22, this volume). Such a focus does not provide an extensive account of the conditions and components important for the implementation of the intervention in a real-world setting. Without sufficient attention to implementation plans and procedures, it may be difficult to attribute whether the effects of an intervention on a change in behavior are due to the intervention content or due to components relating to its implementation. In cases where interventions have been shown to be ineffective in changing behavior, a lack of a theoretical

basis and inadequacies in implementation design are the most common reasons (see Bartholomew Eldredge et al., 2016; Chapter 21, this volume). An alternative approach to an efficacy trial is to test the effects of an intervention on behavior and associated outcomes while simultaneously collecting data on the implementation process to facilitate subsequent implementation, known as hybrid trials (Bauer et al., 2015). Finally, the primary objective of research may be to test the effects of an implementation strategy on implementation outcomes while, at the same time, identifying their influence on behavior in real-world or ecologically valid settings or contexts (implementation trials; Bauer et al., 2015). A process of developing and evaluating interventions encompassing efficacy, hybrid, and implementation trials forms the core stages of the research translation continuum (Drolet & Lorenzi, 2011).

23.2 Implementation Theories, Frameworks, and Taxonomies The implementation of behavior change interventions may be guided by implementation theories, frameworks, and taxonomies, each focusing on specific aspects of implementation (Nilsen, 2015). There are more than sixty theories, models, and

336

ALEKSANDRA LUSZCZYNSKA, KAROLINA LOBCZOWSKA, AND KAROLINA HORODYSKA

frameworks that explain or enhance the implementation of behavior change interventions (Tabak et al., 2012). The models and frameworks vary in terms of their focus (e.g., implementation vs. dissemination), specificity of the constructs included (broad vs. specific or operational), and the levels accounted for (e.g., considering individual, organizational, and community constructs relevant for implementation of an intervention) (Tabak et al., 2012). Theories and models of implementation link the implementation constructs (e.g., implementation barriers, implementation strategies) in a specific manner, elucidating the complex interplay between these constructs (see Bauer et al., 2015). For example, the Ottawa implementation process model proposes that identifying implementation barriers and determinants usually precedes a selection of implementation strategies (Logan & Graham, 2010). Frameworks usually organize included constructs into broader categories and do not explicitly specify the relationships between these constructs (Bauer et al., 2015). For example, according to the consolidated framework for implementation research (Damschroder et al., 2009), a broader category, organizational readiness for implementation (a decision and commitment to implement a behavior change intervention), encompasses three implementation constructs: leadership engagement for implementation, available organizational resources for implementation, and access to information about an intervention for its implementers, participants, and leaders. Finally, taxonomies aim to specify comprehensive lists of relevant constructs or processes and organize those constructs and processes into specific, hierarchically organized categories. For example, the taxonomy proposed by Proctor et al. (2011) lists eight constructs assumed to form key outcomes of implementation, including feasibility, acceptability, and penetration. The purpose of this chapter is to provide an overview of theories, models, frameworks, and taxonomies that attempt to explain or classify (1) implementation processes; (2) implementation

determinants, including barriers and facilitators; and (3) evaluation of implementation processes and implementation outcomes (for a comparable classification of theories and frameworks of implementation, see Nilsen, 2015). Figure 23.1 displays the decision process that may take place when planning for the development, implementation, and evaluation of a behavior change intervention. Translational research on the processes, determinants, and evaluation of behavior change interventions is essential to the implementation of behavior change interventions. Such research links laboratory-based investigation with evidence-based practice targeting behavior change in real-world clinical, educational, organizational, or work settings.

23.3 Theories, Models, and Frameworks of Implementation Processes Multiple theories, models, and frameworks have been developed to detail the key determinants and processes that should lead to the effective implementation of behavior change interventions. The Ottawa model for research use (Graham & Logan, 2004; Logan & Graham, 2010) is an example of an implementation process model. The model proposes a number of consecutive steps, essential to the effective implementation of behavior change interventions. It is operational in focus and specifies the sequence of actions required for optimal implementation. The first step focuses on identification and assessment of barriers and facilitators for implementation. The process starts with identifying those in charge within organizations and determining available resources and those who will be responsible for implementation. Barriers referring to an intervention or policy practice and environment (e.g., characteristics of the existing educational system, centralization of power in the organization) are identified and assessed. Next, barriers and facilitating factors referring to the implementation actors (e.g., a lack of skills to deliver the

Implementation Science and Translation in Behavior Change

337

Consider approaches informing implementation of behavior change interventions in three areas (e.g., Nilsen, 2015)

Models and frameworks used to guide implementation processes

e.g., the Ottawa model for research use: Setting the stage, identification of barriers and facilitations → Selection/monitoring of implementation strategies → Monitoring of the adoption → Evaluation of the effects of implementation process

Models and frameworks used to explain implementation determinants

Models and frameworks used to guide implementation process evaluation and understand implementation outcomes

e.g., the consolidated framework for implementation research:

e.g., the approach to implementation outcomes by Proctor et al. (2011):

Barriers and facilitators referring to: (1) characteristics of interventions, (2) outer setting, (3) inner setting, (4) individual-level moderators (target groups; implementers), (5) determinants of implementation processes

(1) acceptability, (2) adoption (3) appropriateness, (4) costs, (5) feasibility, (6) fidelity, (7) penetration, (8) sustainability or the RE-AIM framework: (1) reach, (2) efficacy, (3) adoption, (4) implementation, (5) maintenance

Figure 23.1 Three overarching aims of using implementation science theories, models,

and frameworks

intervention, absence of incentives for teachers to engage in extracurricular activities in the organization) are measured. Finally, the perceptions of the innovative intervention are evaluated in the target population (e.g., students) and implementation actors (e.g., teachers). For example, research informing the intervention may be perceived as of low credibility or validity, or an intervention targeting a specific topic (e.g., sexual violence) may be considered of limited relevance for the target population (e.g., secondary school students). The second step outlines the selection and monitoring of implementation strategies, including awareness raising and training of those implementing the intervention. The third step

involves monitoring of the actual adoption of a policy or intervention, including changes and adjustments made during adoption. The fourth step includes evaluation of a policy/intervention target group, implementers, and settings (or systems) to determine the effects of the implementation process. The Ottawa model may be used when planning and delivering interventions targeting various populations and outcomes. For instance, the model was used to enhance learning outcomes and the uptake of a community-based interprofessional education course for medical and nursing students. Although interprofessional learning courses are recommended for undergraduate and graduate programs, their uptake is usually limited

338

ALEKSANDRA LUSZCZYNSKA, KAROLINA LOBCZOWSKA, AND KAROLINA HORODYSKA

(see Luebbers et al., 2017). Using the steps of the Ottawa model, Luebbers et al. (2017) developed an intervention to enhance the uptake of an interprofessional education course aimed at knowledge and skills in obesity screening. In the first step, a diagnosis of barriers was conducted and key barriers, such as a perception of the intervention as an extracurricular activity not embedded in the content of existing courses, were identified. Next, these barriers were addressed, for example, by integrating the course with the content taught in the university’s curricula. The intervention resulted in 100 percent of the first-year medical students participating in the interprofessional course for four semesters. In addition, 100 percent of students enrolled in the interprofessional course passed the clinical and knowledge skills in the respective screening areas.

23.4 Theories, Models, and Frameworks Focusing on Implementation Determinants 23.4.1 The Consolidated Framework for Implementation Research The consolidated framework for implementation research (Damschroder et al., 2009) is a hybrid approach developed via an analysis and synthesis of other implementation models and frameworks (e.g., the conceptual model for considering the determinants, dissemination, and implementation of innovations by Greenhalgh et al. [2005] and the Ottawa model for research use by Graham and Logan [2004]). The consolidated framework provides a detailed description of key determinants of implementation grouped into five broad categories. The first category refers to characteristics of interventions, such as the source (e.g., do stakeholders perceive an intervention as internally or externally developed?), design quality and packaging (e.g., is an intervention excellent in terms of how it is bundled, presented, and

assembled?), complexity (e.g., how difficult is it to implement the intervention when its duration, number of steps, and disruptiveness of other processes or tasks are accounted for?), and costs (e.g., what are the costs associated with the implementation of an intervention?). The second category refers to the outer setting in which the intervention is to be implemented, including target groups’ needs and resources, pressure from other competing organizations that have already implemented or are bidding for similar interventions, and external policies or guidelines that may either facilitate or interfere with the implementation of an intervention. The third category of determinants deals with inner setting characteristics, such as the organizational climate; networks and communication channels in the setting; the relative priority and compatibility of implemented interventions, policies, leadership engagement, and norms and culture within an organization that may facilitate or interfere with the implementation of an intervention. The fourth group of determinants captures individual-level moderators. In particular, this group identifies the characteristics of the target group and the individuals, organizations, or networks involved in implementing the intervention and refers to the knowledge and beliefs of these groups with respect to the intervention content, their beliefs in their capability to enact or use the resources of an intervention (e.g., implementation self-efficacy beliefs), and their identification with the organization or setting where an intervention/policy takes place. The final group is the determinants of implementation processes. In particular, it refers to preparing implementation plans, the use of strategies for sustainable engagement of participants, engagement of implementation leaders, securing qualitative feedback, regular debriefing among implementers, and the presence of implementation champions who dedicate their efforts to supporting, marketing, and “driving through” implementation.

Implementation Science and Translation in Behavior Change

The consolidated framework is the most frequently used implementation framework. Research suggests that it has been applied by 20.6 percent of practitioners and researchers specializing in the implementation of behavior change interventions (Birken et al., 2017). However, studies have indicated that only a relatively small number of intervention trials and trial protocols explicitly referred to the framework and its components (Birken et al., 2017). This may be due to the complexity of the framework, in particular the high number of determinants and their multilevel character.

23.4.2 The Taxonomy of Key Implementation Strategies Another recent development is a taxonomy of key implementation strategies (Leeman et al., 2017). The taxonomy differentiates between two sets of actors involved in the implementation of the intervention. The first set includes delivery system actors, that is, the personnel responsible for the implementation of intervention, namely the individuals and teams who implement an intervention in their respective settings (e.g., a schoolteacher who delivers the intervention in a school

339

setting). The second set consists of support system actors, that is, the decision makers responsible for the adoption of an intervention and promoting and supporting its implementation. Support system actors build up the capacity of organizations or systems to implement a specific intervention (e.g., a school principal, a parentteacher association, or a local authority) (Leeman et al., 2017). The implementation strategies identified in the taxonomy may target delivery and support system actors or they may be used by delivery and support system actors to facilitate implementation. Five sets of implementation strategies (see Table 23.1) are identified in the framework (Leeman et al., 2017). The first set includes dissemination strategies that target the support system actors and the delivery system actors. Examples of dissemination strategies include developing social media messaging, packaging customized to the audience, developing information materials for newsletters, or continuing education opportunities. Second, implementation process strategies include the activities necessary for the delivery system actors to adapt and integrate an intervention to the target context and population (Leeman et al.,

Table 23.1 Implementation strategies, involved actors, and implementation characteristics targeted by the strategies (Leeman et al., 2017) Implementation Strategy

Who Enacts the Strategya

Target of the Strategyb

Dissemination

Delivery system actors, support system actors Delivery system actors Delivery system actors Support system actors Support system actors

Intervention and individual

Implementation process Integration strategies Capacity-building strategies Scaling-up strategies

Implementation process Individual and inner setting Individual and implementation process Individual, inner setting, and outer setting

Note. aThe delivery system actors (e.g., people who implement the intervention) or/and support system actors (e.g., decision makers, administrators); bUsing the consolidated framework for implementation research (Damschroder et al., 2009), the target may refer to the characteristics of: the intervention (e.g., complexity, adaptability); individuals (e.g., knowledge, beliefs about the intervention); the inner setting (e.g., organizational culture, existing communication, and training networks); the outer setting (e.g., pressure from other organizations); and the implementation process (e.g., planning, presence of implementation champions).

340

ALEKSANDRA LUSZCZYNSKA, KAROLINA LOBCZOWSKA, AND KAROLINA HORODYSKA

2017). The process strategies may include engaging key stakeholders; adapting interventions and policies to the context; prioritizing implementation goals; and monitoring the process of implementation. Third, integration strategies involve the identification of, and dealing with, factors that facilitate or impede optimal integration of an intervention in a setting (Leeman et al., 2017). Integration strategies are applied by the delivery system actors. For example, supervision of the members of the implementation personnel by a team leader responsible for implementation, organizational reminder systems, and adaptation of professional roles may facilitate the integration and implementation of a newly adopted policy or intervention. Fourth, capacity-building strategies are proposed. These strategies are developed and delivered by support system actors. Capacity-building strategies aim at fostering motivation and capacity among the implementers (delivery system actors). Examples of capacitybuilding strategies include training to build general implementation capacity and technical assistance with implementation provided by specialists employed by support system actors. Fifth, Leeman et al.’s (2017) taxonomy proposes scaling-up strategies that aim to develop motivation and the capacity to integrate an intervention/policy into practice. These strategies are developed and used by the support system actors. Scaling-up strategies may include quality improvement collaboration, and the use of implementation toolkits, to build up recognition systems or benchmarking.

23.4.3 The Theoretical Domains Framework The theoretical domains framework is another example of a framework developed to facilitate the implementation of behavioral interventions (Cane, O’Connor, & Michie, 2012; Michie et al., 2005). The framework was developed through an integration of theoretical constructs relevant for behavior change and combining

them into a single framework (Michie et al., 2005). The primary aim of the original version of the framework was to identify cognitive, affective, social, and environmental determinants that influence the behavior of professionals who deliver the intervention and support system actors, that is, people who build capacity for implementing an intervention or policy. Practitioners and researchers who develop and implement behavior change interventions in health, organizational, and occupational settings have listed the theoretical domains framework among the most frequently indicated implementation frameworks; however, only 5.4 percent of them apply the framework (Birken et al., 2017). The refined theoretical domains framework (Cane et al., 2012) includes fourteen domains (and eighty-four constructs): knowledge (e.g., knowledge of task environment); skills (e.g., interpersonal skills); social/professional role and identity (e.g., leadership); beliefs about capabilities (e.g., professional confidence); optimism (e.g., unrealistic optimism); beliefs about consequences (e.g., outcome expectancies); reinforcement (e.g., incentives); intentions (e.g., stability of intentions); goals (e.g., goal/target setting); memory, attention, and decision processes (e.g., cognitive overload/tiredness); environmental context and resources (e.g., organizational culture/climate); social influences (e.g., intergroup conflicts); emotions (e.g., burnout); and behavioral regulation (e.g., self-monitoring). The use of the theoretical domains framework, or the other implementation models and frameworks, may help to (1) identify key barriers and facilitators that are enabling or hindering the implementation of an intervention and (2) develop an intervention targeting delivery and support system actors (see Leeman et al., 2017). When applying the theoretical domains framework, researchers may first need to identify which of the constructs specified in the theoretical domains framework are the key determinants of the implementation of a specific intervention or policy, delivered in a specific setting. For example, a

Implementation Science and Translation in Behavior Change

study of implementation of a healthy diet promotion program in primary school canteens (Reilly et al., 2019) tested which of fourteen constructs aligned with the theoretical domains framework were related to a lack of full implementation of the program. Results indicated that only 20 percent of the 184 canteens reached full implementation. The most common barriers to the full implementation of the program, identified among school canteen managers, included issues in the following areas: behavioral regulation (identified in 65 percent of canteen managers), skills (57 percent of managers), beliefs about capabilities (55 percent of managers), reinforcement (52 percent of managers), and goals (52 percent of managers) (Reilly et al., 2019). After eliciting the key implementation barriers and facilitators, the next step is to develop an intervention targeting these barriers and facilitators. For instance, Nathan et al. (2016) applied implementation strategies in an intervention developed to enhance healthy diet promotion programs in Australian primary school canteens. The intervention addressed theoretical domains framework–based barriers and used dissemination, process, and implementation strategies, for example sending awareness-raising materials and text messages on key barriers, training for performance monitoring, and feedback (Nathan et al., 2016). The target population included delivery system actors and support system actors, that is, school principals, canteen managers, and parents representing parent-teacher associations. The outcomes included food sale indicators, for example, at least reaching the target of 50 percent of foods sold at schools being classified as fruit, vegetables, reduced-fat dairy products, lean meat/ fish/poultry, and bottled water. A randomized controlled trial enrolling fifty-three schools showed full implementation of a healthy diet program occurred more frequently in schools participating in the implementation intervention condition compared to the control condition (Nathan et al., 2016).

341

23.5 Theories, Models, Frameworks, and Taxonomies Focusing on Implementation Process Evaluation and Implementation Outcomes The frameworks and models of implementation may focus on the evaluation or monitoring of implementation processes or implementation outcomes. The focus here is not on the evaluation of the effects of the intervention on behavior change outcomes but instead on, for example, how successful the intervention was in reaching the target group (e.g., how many people from the potential target population took part in all stages of the intervention?; how socially and culturally inclusive has the intervention been?) or the extent to which the intervention was consistently delivered throughout its stages. The implementation process evaluation frameworks specify the implementation domains or specific constructs that may be assessed, monitored, and evaluated to obtain data indicating the level of implementation of an intervention.

23.5.1 The RE-AIM Framework The RE-AIM framework (Glasgow et al., 2004; Glasgow, Vogt, & Boles, 1999) is an example of an implementation process evaluation framework. “RE-AIM” is an acronym that stands for reach, evaluation, adoption, implementation, and maintenance (Glasgow et al., 2004). The REAIM framework is one of the “top-three” listed approaches to the implementation of behavioral interventions in terms of meeting the criteria of accounting for multilevel contexts and the characteristics of real-world settings (Tabak et al., 2012). In contrast to process-based models (e.g., the Ottawa model for research use; Graham & Logan, 2004; Logan & Graham, 2010) or determinants-based frameworks (e.g., the theoretical domains framework; Cane et al., 2012), the REAIM framework (Glasgow et al., 2004; Glasgow,

342

ALEKSANDRA LUSZCZYNSKA, KAROLINA LOBCZOWSKA, AND KAROLINA HORODYSKA

Vogt, & Boles, 1999) may be considered an implementation process evaluation framework (Nilsen, 2015). The RE-AIM framework is the second most frequently used implementation framework behind the theoretical domains framework, indicated by 13.9 percent of practitioners and researchers specializing in implementation (Birken et al., 2017). According to the RE-AIM framework (Glasgow et al., 2004), there are five aspects of implementation that should be monitored and evaluated by researchers, practitioners, and policy makers: (1) reach of the intervention in the target population and securing its representativeness; (2) efficacy of the intervention in evoking meaningful changes in relevant outcomes; (3) adoption of the intervention by the target staff, group, or institution; (4) consistency, lower costs, and better adaptations in the implementation process; and (5) maintenance or institutionalization of the program and its incorporation into routine practice. The RE-AIM framework does not clarify which strategies may be used to optimize reach, adoption, and maintenance of the intervention and instead raises researchers’ and practitioners’ awareness of these components and the necessity for accounting for all of them in order to successfully implement an intervention.

23.5.2 The Implementation Outcomes Taxonomy Several existing frameworks identify implementation outcomes as an important consideration when evaluating an intervention. Implementation outcomes should be distinguished from service outcomes (e.g., safety of intervention or securing the equity in delivery across populations) or client outcomes (e.g., satisfaction of the target group; Proctor et al., 2011). According to Proctor et al. (2011), there are eight main implementation outcomes that should be monitored and evaluated in behavioral interventions. Acceptability is the perception that an intervention or policy is agreeable,

palatable, or satisfactory for the target population, that is, those whose behavior needs to be changed (Proctor et al., 2011). While acceptability is mainly assessed in the target group of participants, it can also be evaluated among those who deliver/implement an intervention or any stakeholders involved in the process. The outcome of adoption (or uptake) refers to the initial decision or attempt to initiate or employ an intervention in a respective setting (Proctor et al., 2011). Adoption could be measured from the perspective of the provider or organization, with such indices as the number of settings/ organizations agreeing to participate in an intervention. Appropriateness reflects the perceived fit, relevance, or compatibility of an intervention for the aims and/or structures of a target setting of a provider or a consumer. Appropriateness is assessed among providers, managers, and intervention participants to decide if an intervention fits well with the key problems of the target population (e.g., childhood obesity, low academic achievements, counterproductive behaviors in the workplace) (Proctor et al., 2011). An intervention may be considered a “good fit” for the goals of an organization (e.g., eliminating counterproductive behaviors) but some features of an intervention or policy may render it unacceptable to the implementers or the decision makers in the organization. For example, the intervention may be too costly or have a protocol that is viewed as too rigid or lengthy for the organization. The costs of implementation may be divided into costs of the intervention itself (e.g., its complexity, length), costs of the implementation of the intervention (e.g., costs of strategies needed to implement, such as training intensity and resources needed to launch the intervention), and setting-specific costs (e.g., costs that the organization may charge for using its built facilities, etc.) (Proctor et al., 2011). The fifth outcome, feasibility, refers to the extent to which an intervention or policy can be successfully adopted or executed in a particular setting (Proctor et al., 2011). Poor feasibility is usually reflected in poor recruitment and high dropout rates among participants and those

Implementation Science and Translation in Behavior Change

implementing the intervention. An intervention may be appropriate for a setting (i.e., fitting well with the goals of the setting/organization) but may not be feasible due to the extensive resources required or training requirements necessary to develop staff to implement the intervention. Fidelity deals with the degree to which an intervention was implemented as it was designed and described in the original protocol (Proctor et al., 2011). Penetration (or reach) refers to the degree to which an intervention is incorporated across the facets of the setting or a proportion of a target population that has been reached (Proctor et al., 2011). For example, an intervention with poor penetration reaches only a subsample of the intended/targeted population, such as well-educated, young, and motivated individuals. Finally, sustainability refers to the extent to which an intervention was incorporated and institutionalized into the organizational setting and is currently an essential part of ongoing operations (Proctor et al., 2011). A sustainable intervention becomes a key element of services delivered constantly in a respective setting. The eight implementation outcomes identified by Proctor et al. (2011) are measurable. A systematic review (Lewis et al., 2015) indicated that at least 104 different measures have been used to evaluate the eight implementation outcomes. The most frequently used measures were those assessing intervention acceptability (n = 50) and adoption (n = 19), with the lowest number of measures assessing penetration (n = 4).

23.5.3 The Taxonomy of Implementation Monitoring and Evaluation of Implementation Recent research has used systematic review methods to identify the key aspects of monitoring and evaluation of implementation (Horodyska, Luszczynska, van den Berg et al., 2015; Horodyska, Luszczynska, Hayes et al., 2015). For example, the evidence-based taxonomy of

343

best-practice development, implementation, and evaluation of interventions (Horodyska, Luszczynska, van den Berg et al., 2015) suggests six broad categories of evaluation and monitoring constructs (see Sidebar 23.2). Besides constructs referring to the evaluation of efficacy (the influence on behavior change) and underlying mechanisms (the evaluation of active components), four evaluation and monitoring categories include constructs that refer to the evaluation and monitoring of implementation (Horodyska, Luszczynska, van den Berg et al., 2015).

23.6 Translational Models and Research Translational research is usually defined as the process by which research findings are moved from basic research (e.g., laboratory) to a realworld target community or population (e.g., organizational systems, patients’ beds, schools) (Rubio et al., 2010). Translational scientists have proposed a “meta-model,” the research translation continuum (Drolet & Lorenzi, 2011), which provides a set of overarching descriptions of the processes linking basic science findings with the evidence-based applications. The continuum suggests that any research can be situated at some point along a continuum of translational activities, linking basic science discoveries and evidence-based practice (Drolet & Lorenzi, 2011). After conducting laboratory-based research, further research efforts may focus on translation from the laboratory to the target population, which may then lead to a proposed application of the intervention. Next, translational activities focus mainly on further evaluation of safety and efficacy and contribute to evidence supporting the established application of the intervention in a specific population and in a given context (see also Chapters 21 and 22, this volume). In the next stage, further translation activities may include in-depth investigation of implementation of a procedure/intervention in the

344

ALEKSANDRA LUSZCZYNSKA, KAROLINA LOBCZOWSKA, AND KAROLINA HORODYSKA

Sidebar 23.2 The monitoring and evaluation of implementation of behavior change interventions according to taxonomies of best practices

The taxonomy of best practices in behavior change interventions and policies (Horodyska, Luszczynska, van den Berg et al., 2015) suggests the evaluation and monitoring of the following aspects of implementation: (1) total financial costs of the intervention (e.g., total budget per participant, including costs of implementation); (2) the use of reach strategies (e.g., increasing the likelihood of involving a large percentage of the target population); (3) the degree of inclusiveness (e.g., low mobility or high comorbidity, age and gender groups); (4) cultural competence and social inclusion strategies used during implementation (e.g., accounting for cultural minority issues in recruitment processes, content development, setting selection); (5) participation rates (across stages of implementation); and (6) monitoring and measurement of delivery and monitoring and measurement of materials (e.g., adaptations made). Another taxonomy focusing solely on implementation characteristics (Horodyska, Luszczynska, Hayes et al., 2015) suggests the evaluation and monitoring of the following aspects of implementation: (1) participants’ general satisfaction with the intervention implementation; (2) the feasibility of intervention implementation among providers, participants, and other stakeholders; (3) the acceptability of the program among participants (e.g., acceptability of the group size, the type of participants, interventionists’ skills, materials); (4) the dissemination of results to participating communities and stakeholders; and (5) the difficulties in assessing the impact of one intervention separately from ancillary policies/interventions. To validate the implementation characteristics taxonomy, Muellmann et al. (2017) conducted a study eliciting key implementation factors for multilevel interventions and policies promoting healthy eating and physical activity participation across five European countries: Belgium, Germany, Ireland, Norway, and Poland. Findings indicated that common implementation themes in the intervention implementation included the evaluation of continuous involvement of relevant stakeholders; good communication between coordinating organizations; and the tailoring of materials to match the needs of socially and culturally diverse target groups to secure feasibility and acceptability of the intervention in the target population.

target settings and population (Drolet & Lorenzi, 2011). This stage may account for research on implementation processes, implementation determinants, and evaluation of implementation. Finally, practice-based research is conducted to further improve applications in the clinical practice, foster greater adoption of the intervention across populations and settings, and provide feedback to previous stages for further examination.

23.7 Implications When designing, enacting, conducting, and reporting interventions aimed at behavior change, researchers usually emphasize mechanismrelated constructs and models that focus on intervention efficacy. This relatively narrow focus may result in limited insight into intervention effects and may be relatively silent on whether

Implementation Science and Translation in Behavior Change

345

Sidebar 23.3 An example of the questionnaire to measure behavior change intervention adoption

The adoption questionnaire by Steckler et al. (1992) is an example of a measure of adoption, one of the key implementation outcomes according to the taxonomy proposed by Proctor et al. (2011). This nine-item questionnaire was originally developed in the context of change of smoking behavior at schools. The questionnaire accounts for three factors: relative advantage of an innovative intervention compared to existing practices (e.g., “The new intervention is better than our current curriculum practices for tobacco prevention”); complexity of the innovative intervention (e.g., “The new intervention would be hard for teachers to understand”), and observability of effects (e. g. “Methods for assessing the new intervention’s impact on teachers are readily available”). The responses are given on five-point scales, ranging from “strongly agree” to “strongly disagree.” The questionnaire is administered to support system actors (e.g., school administration) and delivery system actors (e.g., teachers implementing the intervention). Lewis et al. (2015) rated Steckler et al.’s (1992) adoption questionnaire among the best in terms of structural validity, norms, and usability.

the intervention can be effectively implemented (Luszczynska, Horodyska et al., 2016). Use of theories, models, frameworks, and taxonomies to inform evaluation and implementation efforts may strengthen transferability and replicability of behavior change interventions and enrich understanding of the components of the interventions (Nilsen, 2015). Education for researchers and practitioners developing behavior change interventions should account for implementation approaches, which may help to design, enact, and report interventions promoting successful and meaningful behavior change in a target population and taking into account implementation considerations already in the early phases of development (see also Chapters 21 and 22, this volume). Although the use of implementation theories, models, frameworks, and taxonomies adds complexity to research and practice, the benefits outweigh the costs. They increase the ability to enhance the effectiveness and deepen understanding of interventions and their implementation. Their use would often require a pilot study using appropriate methods, as well as collection

of data from those involved in delivery of intervention/policy and appropriate stakeholders in the target setting (see Chapters 21 and 22, this volume). The number of tools to assess implementation constructs and participant-relevant outcomes is growing (Lewis et al., 2015; SIRC, 2018). Many of these tools are well established and psychometrically sound (see Sidebar 23.3).

23.8 Summary and Conclusion When planning and developing a behavior change intervention, three broad issues should be accounted for: the efficacy of the intervention in evoking change in the target behavior (a response to the question, “Will the intervention work?”; Abraham et al., 2014; Peters, de Bruin, & Crutzen, 2015; see Chapter 19, this volume); the mechanisms of the intervention responsible for behavior change (a response to the question: “how will the intervention work?”; Abraham et al., 2014; Peters et al., 2015; see Chapters 20 and 21, this volume); and the conditions that determine the effective implementation of the intervention in the real-world settings and contexts (a response to

346

ALEKSANDRA LUSZCZYNSKA, KAROLINA LOBCZOWSKA, AND KAROLINA HORODYSKA

the question, “Under what conditions will the intervention work?”; Horodyska, Luszczynska, Hayes et al., 2015). The final component should include specification of the conditions necessary for the intervention to be implemented in realworld settings and contexts (e.g., resources needed to deliver the intervention); characteristics of the target population (e.g., a consideration of the needs of the intervention participants); characteristics of providers (e.g., the skills needed to implement an intervention/policy), physical and social characteristics of the setting in which the intervention will take place (e.g., environmental barriers such as a lack of a built infrastructure); and the political, economic, and legal systems characteristics involved in implementing the intervention (e.g., conflicting policies that operate during processes of adoption of an intervention in a respective setting or region) (Horodyska, Luszczynska, Hayes et al., 2015). The theories, frameworks, and taxonomies of implementation discussed in this chapter outline the important considerations that should inform the implementation process for behavior interventions. The application of these approaches allows intervention designers to oversee and tackle possible barriers to intervention implementation, identify resources enabling implementation, and integrate implementation outcomes into the process of intervention evaluation (see also Chapter 21, this volume). The implementation theories, models, frameworks, and taxonomies may guide researchers and practitioners to strengthen the effectiveness of behavior change interventions and help identify the conditions responsible for their effectiveness in changing the target behavior in the target population and in the target context.

References Abraham, C., Johnson, B. T., de Bruin, M., & Luszczynska, A. (2014). Enhancing reporting of behavior change intervention evaluations. Journal of Acquired Immune Deficiency Syndromes, 66,

S293–S299. https://doi.org/10.1097/QAI.000000 0000000231 Bartholomew Eldredge, L. K., Markham, C. M., Ruiter, R. A. C., Fernández, M. E., Kok, G., & Parcel, G. S. (2016). Planning Health Promotion Programs: An Intervention Mapping Approach (4th ed.). San Francisco: Jossey-Bass. Bauer, M. S., Damschroder, L., Hagedorn, H., Smith, J., & Kilbourne, A. M. (2015). An introduction to implementation science for the non-specialist. BMC Psychology, 3, 32. https://doi.org/10.1186/ s40359-015-0089-9 Birken, S. A., Powell, B. J., Shea, C. M. et al. (2017). Criteria for selecting implementation science theories and frameworks: Results from an international survey. Implementation Science, 12, 124. https://doi.org/10.1186/s13012-017-0656-y Cane, J., O’Connor, D., & Michie, S. (2012). Validation of the theoretical domains framework for use in behaviour change and implementation. Implementation Science, 7, 37. https://doi.org/ 10.1186/1748-5908-7-37 Damschroder, L. J., Aron, D. C., Keith, R. E. et al. (2009). Fostering implementation of health services research findings into practice: A consolidated framework for advancing implementation science. Implementation Science, 4, 50. https://doi.org/10.1186/1748-5908-4-50 Drolet, B. C., & Lorenzi, N. M. (2011). Translational research: Understanding the continuum from bench to bedside. Translational Research, 157, 1– 5. https://doi.org/10.1016/j.trsl.2010.10.002 Eccles, M. P., & Mittman, B. S. (2006). Welcome to implementation science. Implementation Science, 1, 1. https://doi.org/10.1186/1748-5908-1-1 Glasgow, R. E., Klesges, L. M., Dzewaltowski, D. A., Bull, S. S., & Estabrooks, P. (2004). The future of health behavior change research: What is needed to improve translation of research into health promotion practice? Annals of Behavioral Medicine, 27, 3–12. https://doi.org/10.1207/ s15324796abm2701_2 Glasgow, R. E., Vogt, T. M., & Boles, S. M. (1999). Evaluating the public health impact of health promotion interventions: The RE-AIM framework. American Journal of Public Health, 89, 1322–1237.

Implementation Science and Translation in Behavior Change

Graham, I. D., & Logan, J. (2004). Innovations in knowledge transfer and continuity of care. Canadian Journal of Nursing Research, 36, 89–103. Greenhalgh, T., Robert, G., Bate, P., Macfarlane, F., Bate, P., & Kyriakidou, O. (2005). Diffusion of innovations in service organisations: Systematic review and recommendations. Milbank Quarterly, 82, 581–629. https://doi.org/10.1111/j.0887-378X .2004.00325.x Hagger, M. S., & Weed, M. E. (2019). DEBATE: Do behavioral interventions work in the real world? International Journal of Behavioral Nutrition and Physical Activity, 16, 36. http://dx.doi.org/ 10.1186/s12966-019-0795-4 Horodyska, K., Luszczynska, A., Hayes, C. B. et al. (2015). Implementation conditions for diet and physical activity interventions and policies: An umbrella review. BMC Public Health, 15, 1250. https://doi.org/10.1186/s12889-015-2585-5 Horodyska, K., Luszczynska, A., van den Berg, M. et al. (2015). Good practice characteristics of diet and physical activity interventions and policies: An umbrella review. BMC Public Health, 15, 19. https://doi.org/10.1186/s12889-015-1354-9 Leeman, J., Birken, S. A., Powell, B. J., Rohweder, C., & Shea, C. M. (2017). Beyond “implementation strategies”: Classifying the full range of strategies used in implementation science and practice. Implementation Science, 12, 125. https://doi.org/ 10.1186/s13012-017-0657-x Lewis, C. C., Fischer, S., Weiner, B. J., Stanick, C., Kim, M., & Martinez, R. G. (2015). Outcomes for implementation science: An enhanced systematic review of instruments using evidence-based rating criteria. Implementation Science, 10, 155. https:// doi.org/10.1186/s13012-015-0342-x Logan, J., & Graham, I. D. (2010). The Ottawa model of research use. In J. R. M. Bucknall (Ed.), Models and Frameworks for Implementing EvidenceBased Practice: Evidence to Action (pp. 83–108). Oxford: Wiley-Blackwell. Luebbers, E. L., Dolansky, M. A., Vehovec, A., & Petty, G. (2017). Implementation and evaluation of community-based interprofessional learning activity. Journal of Interprofessional Care, 31, 91–97, https:// doi.org/10.1080/13561820.2016.1237936

347

Luszczynska, A., Horodyska, K., Zarychta, K., Liszewska, N., Knoll, N., & Scholz, U. (2016). Planning and self-efficacy interventions encouraging replacing energy-dense foods intake with fruit and vegetable: A longitudinal experimental study. Psychology and Health, 31, 40–64. https://doi.org/ 10.1080/08870446.2015.1070156 Michie, S., Johnston, M., Abraham, C., Lawton, R., Parker, D., & Walker, A. (2005). Making psychological theory useful for implementing evidence based practice: A consensus approach. Quality and Safety in Health Care, 14, 26–33. https://doi.org/10.1136/qshc.2004.011155 Muellmann, S., Steenbock., B., De Cocker, K. et al. (2017). Views of policy makers and health promotion professionals on factors facilitating implementation and maintenance of interventions and policies promoting physical activity and healthy eating: results of the DEDIPAC project. BMC Public Health, 17, 932. https://doi.org/ 10.1186/s12889-017-4929-9 Nathan, N., Yoong S. L., Sutherland, R. et al. (2016). Effectiveness of a multicomponent intervention to enhance implementation of a healthy canteen policy in Australian primary schools: A randomised controlled trial. International Journal of Behavioral Nutrition and Physical Activity, 13, 106. https://doi.org/10.1186/s12966-016-0431-5 Nilsen, P. (2015). Making sense of implementation theories, models and frameworks. Implementation Science, 10, 53. https://doi.org/10.1186/s13012015-0242-0 Ory, M. G., Jordan, P. J., & Bazzarre, T. (2002). The behavior change consortium: Setting the stage for a new century of health behavior-change research. Health Education Research, 17, 500–511. https:// doi.org/10.1093/her/17.5.500 Peters, G. -J. Y., de Bruin, M., & Crutzen, R. (2015). Everything should be as simple as possible, but no simpler: Towards a protocol for accumulating evidence regarding the active content of health behaviour change interventions. Health Psychology Review, 9, 1–14. https://doi.org/ 10.1080/17437199.2013.848409 Pfadenhauer, L. M., Gerhardus, A., Mozygemba, K. et al. (2017). Making sense of complexity in context and implementation: The Context and

348

ALEKSANDRA LUSZCZYNSKA, KAROLINA LOBCZOWSKA, AND KAROLINA HORODYSKA

Implementation of Complex Interventions (CICI) framework. Implementation Science, 12, 21. https://doi.org/10.1186/s13012-017-0552-5 Proctor, E. K., Landsverk, J., Aarons, G., Chambers, D., Gilsson, C., & Mittman, B. (2009). Implementation research in mental health services: An emerging science with conceptual, methodological, and training challenges. Administration and Policy in Mental Health, 36, 22–34. https://doi.org/10.1007/ s10488-008-0197-4 Proctor, E., Silmere, H., Raghavan, R. et al. (2011). Outcomes for implementation research: Conceptual distinctions, measurement challenges, and research agenda. Administration and Policy in Mental Health, 38, 65–76. https://doi.org/ 10.1007/s10488-010-0319-7 Rabin B. A., & Brownson, R. C. (2012). Developing the terminology for dissemination and implementation research. In Brownson, R., C, Colditz, G. A., & Proctor, E. K, (Eds.), Dissemination and Implementation Research in Health (pp. 23–51). New York: Oxford University Press. Reilly, K., Nathan, N., Grady, A. et al. (2019). Barriers to implementation of a healthy canteen policy: A survey using the theoretical domains framework. Health Promotion Journal of Australia. https:// doi.org/10.1002/hpja.218

Rubio, D. M., Schoenbaum, E. E., Lee, L. S. et al. (2010). Defining translational research: Implications for training. Academic Medicine, 85, 470–475. https://doi.org/10.1097/ ACM.0b013e3181ccd618. Rutter, H., Savona, N., Glonti, K. et al. (2017). The need for a complex systems model of evidence for public health. Lancet, 390, 2602–2604. https://doi .org/10.1016/S0140-6736(17)31267-9 SIRC (Society for Implementation Research Collaboration). (2018). The SIRC Instrument Review Project (IRP): A Systematic Review and Synthesis of Implementation Science Instruments. www.societyforimplementationre searchcollaboration.org/sirc-projects/sirc-instru ment-project Steckler, A., Goodman, R. M., McLeroy, K. R., Davis, S., & Koch, G. (1992). Measuring the diffusion of innovative health promotion programs. American Journal of Health Promotion, 6, 214–224. https://doi.org/10.4278/ 0890-1171-6.3.214 Tabak, R. G., Khoong, E. C., Chambers, D. A., & Brownson, R. C. (2012). Bridging research and practice: Models for dissemination and implementation research. American Journal of Preventive Medicine, 43, 337–350.

24 Engagement of Stakeholders in the Design, Evaluation, and Implementation of Complex Interventions Joanna L. Hudson, Zoe Moon, Lyndsay D. Hughes, and Rona Moss-Morris

Practical Summary Randomized controlled trials (RCTs) are considered the gold standard approach to evaluating behavior change interventions. While many RCTs show that behavior change interventions are effective, few get put into routine practice. One reason for this is likely because stakeholders are not involved in their design, evaluation, and subsequent implementation into routine practice. For example, in health care contexts, potential stakeholders include patients, health care professionals, and clinical commissioners. This chapter outlines the importance of engaging stakeholders from conceiving the idea for the intervention right through to putting it into practice once RCTs have shown them to be effective. This is illustrated with three health care examples: (1) involving patient stakeholders during the design of an “app” to support women prescribed hormone therapy for breast cancer; (2) interviewing patients and health care professionals during a feasibility study to help identify and address barriers to implementing online cognitive behavioral therapy (CBT) for distress in dialysis; and (3) engaging with commercial partners to support the implementation and scalability of an online CBT intervention for treating irritable bowel syndrome (IBS) symptoms.

24.1 Introduction Developing evidence-based behavior change interventions that work is a costly and timeconsuming process. Even when this is achieved, the uptake of these interventions into routine practice, particularly when they include complex multicomponents, very seldom happens. For example, in the UK alone, more than two million people are living post-cancer and a third of these report reductions in quality of life. Participating in physical exercise, mindfulness-based stress reduction, and cognitive behavioral therapy (CBT) have all been shown to improve the quality of life of cancer

survivors (Duncan et al., 2017). Yet very few of these people are prescribed or indeed have routine access to these evidence-based treatments. How does the field move from evidence-based treatments to routine implementation? In the UK, the Medical Research Council (MRC) has developed a framework to support the development and evaluation of complex interventions that focus on changing health-related behaviors and/or health outcomes (Craig et al., 2008; Moore et al., 2015; Skivington et al., 2018). The framework has relevance to health, https://doi.org/10.1017/9781108677318.024

350

JOANNA L. HUDSON, ZOE MOON, LYNDSAY D. HUGHES, AND RONA MOSS-MORRIS

education, transport, housing, and environmental challenges, all of which can have an impact on a person’s health related outcomes. It outlines the factors that need to be considered during the following four phases: (1) intervention development (e.g., using a theory to hypothesize what factors might influence a person’s behavior); (2) feasibility and piloting (e.g., estimating recruitment and retention rates for a future full-scale trial); (3) evaluation (e.g., designs and statistical analyses needed to perform robust effectiveness and costeffectiveness studies); and (4) implementation (e. g., dissemination, surveillance, and long-term follow-up of the intervention). For a more detailed explanation of how the MRC framework is applied to the development and evaluation of behavior change interventions, see Chapters 21 and 22, this volume. This chapter discusses how embedding stakeholder engagement into the four phases of the MRC framework may substantially enhance the potential for downstream implementation. Currently the MRC framework only provides guidance on implementation once a robust evaluation is completed. It is argued that, to improve the likelihood that effective interventions are adopted and embedded into routine practice, it is vital that barriers and facilitators to implementation are considered and managed throughout each of the four phases of the MRC framework. The focus of this chapter is to provide an overview of methods of including stakeholders. First, the chapter introduces the reader to the field of implementation science. The meaning of the term “stakeholder” is initially defined and a brief introduction provided to a common theory used to study the processes of implementation known as normalization process theory (May et al., 2007; May et al., 2009). Methods of including stakeholders in the research process are then discussed, and three examples are used to illustrate points presented.

24.2 What Is Implementation and Stakeholder Engagement? Implementation science is an emerging field of research that aims to understand the factors that promote or inhibit the implementation of interventions in routine practice (Eccles & Mittman, 2006; May et al., 2007; see also Chapter 23, this volume). Implementing a new intervention requires users of the intended intervention to change their behavior. Implementation is therefore a complex intervention in itself. A stakeholder is defined as any proposed user of the intervention whose behavior may need to change (Chapter 19, this volume). In the health context, this may include patients, carers, health care professionals, clinical commissioners or health insurance companies, policy makers, service managers, and contractual managers. In addition, it may include businesses who supply access to resources needed to deliver the intervention (e.g., IT software). Given that implementation requires behavior change from multiple stakeholders, it requires a theoretical understanding of the factors that facilitate the uptake, maintenance, and spread of interventions (May et al., 2007; May et al., 2009). This will help to inform the development of specific strategies to support intervention implementation by successfully changing stakeholder behavior. A common theory used to understand the processes by which an intervention does, or does not, become part of routine practice is normalization process theory (May et al., 2007; May et al., 2009). This theory focuses specifically on groups and organizational processes that promote or inhibit the adoption and maintenance of interventions (May et al., 2007). The theory considers complex interventions to consist of three core components: 1. Actors of the intervention – all individuals who will encounter the intervention. In this chapter, actors are considered to be synonymous with how stakeholders are defined previously. The terms actors, stakeholders, and users are used synonymously throughout the chapter.

Engagement of Stakeholders

2. Objects of the intervention – which include the means through which an intervention is executed, for example through the use of standardized protocols, drug-prescription charts, evidence-based treatment manuals, digital delivery, and specific hardware. 3. Context of the intervention – which includes physical, organizational, institutional, and legislative structures surrounding the delivery of the intervention. Therefore, to support the translation of interventions into practice it is important to understand the processes through which actors (stakeholders), objects, and contexts interact with one and other. Normalization process theory has empirically identified four processes shaping these interactions (May, 2013): 1. Coherence – how users make sense of the intervention (Finch, 2008; May, 2013). For example, do users see the intervention as dramatically different from usual practice? Do all users of the intervention have the same understanding of its purpose? Do users of the intervention understand their roles when engaging with the intervention? Do users think that the intervention is needed? 2. Cognitive participation – whether users of the intervention “sign up” to the idea of it (May et al., 2007; May et al., 2009). For example, are key users of the intervention driving it forward (e.g., senior management teams, patient advocate groups, research team)? Do intended intervention users view it as part of their role? Are intended users open to exploring new ways of working with each other to deliver the intervention? Are people using and engaging with the intervention? 3. Collective action – the change needed in knowledge and behaviors to deliver the intervention (Mair et al., 2012; May et al., 2007; May et al., 2009). For example, do interactions between providers (e.g., doctors) and receivers (e.g., patients) of the intervention need to

351

change? Do working relationships within the organization providing the intervention need to change? (e.g., do doctors need to work more closely with psychologists to support the mental and physical health needs of patients?) Do intervention providers have the skills they need to deliver the intervention? Is it possible to deliver the intervention with current resources? 4. Reflexive monitoring (May et al., 2009) – how users of the intervention check if it is working. For example, are systems in place to formally appraise this? Do all users of the intervention agree on how to evaluate if the intervention is working? Do all users of the intervention feel they make a valuable contribution to its implementation? Are there systems in place to feed back views on how the intervention is working? Having an understanding of each of these four core processes from normalization process theory can help to forecast and plan for implementation (Murray et al., 2010). When an intervention is launched into practice there are three potential outcomes (May et al., 2007; May et al., 2009). Outcome one is that an intervention is immediately rejected by the end users/stakeholders. Outcome two is that an intervention may be initially adopted but it subsequently fails to become part of routine practice. Outcome three is the successful embedding of the intervention so that it becomes part of normal practice. Once embedded, the implementation focus needs to be on integration so that the intervention is scaled up and sustained within and across organizations over time. It is important to bear in mind that there always remains the potential for an intervention to become de-normalized. This may occur because it becomes outdated as, for example, new technologies emerge. Alternatively, the factors that promote the normalization of an intervention may become disrupted (e.g., stakeholders responsible for driving the intervention forward move on to a different job role and no longer remain within the organizational context to drive it forward). Given this brief overview of normalization process theory,

352

JOANNA L. HUDSON, ZOE MOON, LYNDSAY D. HUGHES, AND RONA MOSS-MORRIS

the chapter next explains how aspects of this approach are integrated with a wider range of methods to ensure stakeholders are at the heart of the example interventions developed across the MRC stages.

24.3 Methods of Involving Stakeholders There are two key ways of involving stakeholders in the development and testing of behavioral interventions: (1) as research partners and (2) as research participants. Research partnerships with stakeholders is often referred to as patient and public involvement (PPI) in the UK (NIHR, 2019). PPI partnerships comprise patients and/or members of the public or relevant charity organizations who act as an advisory group to the researchers and who may also be coresearchers. PPI is now considered an integral part of most UK funding agreements, whether it be intervention development, feasibility trial, or RCT, and PPI partners are paid for their time commitment. The commitment usually starts at the ideas or design phase of the study and continues through the setup of trials, the running of the trials, and the analysis phase. Helping to disseminate the results is also a core part of PPI and can also be crucial for helping with implementation. This partnership arrangement is distinct from gathering stakeholder views as part of the research process/participation where both quantitative and qualitative methods can be used. Surveys can be used to gather quantitative information from many different stakeholders. There is a useful, openaccess, normalization process theory website that describes in detail how to set up surveys within the framework of the theory. The types of questions relevant to gathering information about coherence, cognitive participation, collective action, and reflexive monitoring are included (for details, see May et al., 2015). To get a richer, more detailed view from a smaller number of stakeholders, qualitative methods including individual interviews and focus groups can be

used and analyzed deductively in accordance with the theory constructs. This is particularly helpful in terms of understanding what people do and how the context in which they work is likely to facilitate or impede future implementation. An illustration of this procedure is provided in Sidebar 24.1. If the research aims to get a better understanding of the users’ attitudes and reactions to a new behavior change intervention, then drawing from the person-based approach to intervention development may be more useful (Yardley et al., 2015). This approach provides a framework to elicit the views and reactions of the intended users of the intervention using qualitative methods such as focus groups, interviews, and “think-aloud” studies. The data here are often inductively analyzed so that new themes and information can arise from the data. Further examples of this approach are provided in the next section (see also Chapter 25, this volume).

24.4 Stakeholder Engagement at the Intervention Development Phase Interventions are needed to support breast cancer survivors and improve hormone therapy adherence rates, which are commonly reported to be as low as 50 percent (Brito, Portela, & Vasconcellos, 2014). Adherence to hormonal therapy reduces the risk of breast cancer recurrence in patients who have estrogen receptor-positive breast cancer (EBCTCG, 1998, 2011). The following example of how the guidance in Stage I of the MRC framework for intervention development was adhered to by applying two common social cognition models of health behavior to explore predictors of adherence to hormone therapy over twelve months. The two social cognition models used included an extended version of the common-sense model of illness self-regulation (Leventhal et al., 2012; Chapter 5, this volume) that includes the necessity-concerns framework (Horne, Weinman, & Hankins, 1999) and the theory of planned behavior (Ajzen, 1991; see also Chapter 2, this volume). Variables drawn from both models

Engagement of Stakeholders

were found to explain intentional nonadherence and forgetting to take hormonal therapy (Moon, MossMorris, Hunter, Norton, & Hughes, 2019). Relevant psychosocial targets for intervention were identified from this empirical work, including the perceived necessity of treatment and self-efficacy in managing side effects and medication taking. Additional treatment targets were also drawn from a thematic analysis of qualitative data on the experience of taking hormonal therapy (Moon, Moss-Morris, Hunter, & Hughes, 2017). A theory-based self-management booklet was drafted by the research team and further developed with extensive feedback from patient representatives and clinicians (Moon, MossMorris, Hunter, & Hughes., 2019). This approach was tested in a small RCT, which showed the intervention to be acceptable and feasible in this population and that the intervention was promising in terms of changing unhelpful illness and treatment beliefs, improving adherence, and reducing the experience of side effects (Moon, Moss-Morris, Hunter, Goodliffe, & Hughes, 2019). The next phase of the work was to develop an intervention, which was more likely to be scalable and implementable. Using the normalization process theory framework, the actors were defined as breast cancer survivors, commissioners of services, and the health care professionals who prescribed hormonal therapies (oncologists) and provided ongoing support (breast cancer nurses). The context in which the intervention was to be deployed was also taken into consideration. In the UK, breast cancer follow-up services are currently moving from prescheduled follow-ups to an open-access model where women are encouraged to contact their breast cancer nurse if they have any concerns. Breast cancer nurses are already overstretched so any implementable intervention needed to both engage the patient in self-management and ensure minimal input was needed from a health care professional. While the means of delivery of the initial intervention was paper-based, the changing context suggested that a change to a digital intervention largely managed by the patients was likely to be

353

more scalable (see Chapter 29, this volume). Creating an interactive digital platform was also in line with the ten-year forward view of a national digital transformation to the health service in England (NHS, 2019). To ensure that this patient-led intervention met the normalization process theory criteria of coherence and cognitive participation, the initial goal was to ensure that the intervention was useful, usable, accessible, and acceptable to the main stakeholders: patients. A normalization process theory–based survey, PPI, and the person-based approach were therefore used to achieve these aims. These findings informed essential changes that were made to the app-based intervention as described in Sidebar 24.1.

24.5 Stakeholder Engagement at the Feasibility and Piloting Phase Further stakeholder work during the feasibility and piloting phase of the MRC framework can provide critical feedback on the factors that promote or prevent an intervention from being embedded into practice. Here, illustrative examples are provided from work focused on manualized and web-based CBT-informed self-management for managing distress and symptoms in people with end-stage kidney disease (Hudson, Moss-Morris, Game, Carroll, & Chilcot, 2016; Hudson, Moss-Morris, Game, Carroll, McCrone et al., 2016; Hudson et al., 2017). CBT includes a wide range of cognitive and behavioral techniques to reduce unhelpful coping behaviors such as avoidance and increase a wide range of health promoting behaviors such as maintaining activity in the face of symptoms, engaging in pleasurable activities, developing regular daily routines, and improving sleep. A web-based CBT self-management program for treating distress in hemodialysis patients, iDiD (Improving Distress in Dialysis), was developed over a period of a year using theory and rigorous stakeholder codesign, incorporating the person-centered principles described in Section 24.4 (Hudson,

354

JOANNA L. HUDSON, ZOE MOON, LYNDSAY D. HUGHES, AND RONA MOSS-MORRIS

Sidebar 24.1 Using stakeholders to guide the development of the myHT app to support adherence to hormone therapy in breast cancer survivors

Table 24.1 shows the steps taken to develop and digitize the intervention in line with stakeholder feedback. It highlights the key learning points from each of these steps. This extensive patient involvement was an essential step in ensuring that the digital intervention will be acceptable to the users and will hopefully overcome some of the low uptake and engagement seen with other eHealth interventions. Had this work not been carried out, it is likely that the resulting product developed would not be as user-friendly and could include several elements that are “off-putting” to the users. This work also helped develop an intervention that fits with the guiding principles identified as part of the person-based approach: user autonomy and competence, positive emotional experience, and a sense of relatedness (Yardley et al., 2015; see also Chapters 8 and 35, this volume). User autonomy was promoted by giving patients choices with regards to which sections of the “app” they focused on and by giving options regarding the number and timing of notifications. User competence was promoted by breaking the app into smaller sections and by creating a linear progress sequence that was easy to follow. Positive emotional experience was promoted by using supportive, nonjudgmental language and by providing personally relevant and tailored information. Table 24.1 Stakeholder engagement and key learning points while developing myHT, an app-based intervention to promote adherence to hormonal therapy in breast cancer survivors Stakeholder Involvement

Key Learning Points

Qualitative interviews with women who • took part in the feasibility study for the paper-based intervention (n = 17) •

• •

Interviews were conducted over the • telephone and were analyzed using thematic analysis The interviews lasted around 20 minutes • For more information, see Moon et al. (2019) •

Qualitative interviews with women to • determine preferences for app or website (n = 4) •

Women want a digital version so they can carry it around and update on the go as well as access links to online resources Having it as an app feels more private. Do not have to worry about confidential information Some women would still prefer paper booklet to reduce screen time. They like the action of physically writing things down Majority of information in paper booklet was very acceptable and helpful Preference for both an app and a website but, if could only have one, then an app was preferred Having the option to download the information within the app would be useful Continued

Engagement of Stakeholders

355

Table 24.1 (cont.)

Stakeholder Involvement

Key Learning Points

Online survey to establish app/website usage and needs (n = 137)

• • •

Feedback from patient representatives (PPI) on written materials (n = 5) • •

•

Patients identified through social media, charities, and support groups • Feedback obtained through email and telephone calls

Focus groups to get feedback on app (n = 13) • •

•

• •

Patients identified through social media, charities, and support groups Two focus groups were held, questions • were posed to the group regarding specific aspects of the “app,” and • responses were noted •

•

• •

Comprehensive “app” testing using think-aloud methods (n = 4) •

• •

Women asked to think aloud as they complete the app and verbalize their • thought processes. This helps to identify • any areas where the clarity, functioning, • or content could be improved

97 percent owned a smartphone and 74 percent owned a tablet. 83 percent used apps more than once a day Preference for both an app and a website but, if could only have one, then an app was preferred Addition of other key side effects (urinary frequency, cognitive effects) Changes made to language to make it more personal Making sure all the information appears logical and relevant Name myHT (my Hormonal Therapy) and logo of app determined Women want the option of having narration for the animations that they can turn on and off Discussion of number of quotes within the app Notification timing/frequency should all be determined by the user The option to download content within the app would be very well received Should be able to add multiple symptom monitoring scores throughout the day Emotional nature of risk information needs to be considered Name of “vaginal dryness” and similar symptoms should be renamed to protect privacy if using in public place Clarification of instructions for some activities Break some pages up into smaller sections Add some more color where possible Reduce font size Addition of new resources

356

JOANNA L. HUDSON, ZOE MOON, LYNDSAY D. HUGHES, AND RONA MOSS-MORRIS

Moss-Morris, Game, Carroll, & Chilcot, 2016). Despite the intensive user input into the intervention development, during the subsequent feasibility study, a range of recruitment issues were encountered, which meant that recruitment fell well short of targets (Hudson, Moss-Morris, Game, Carroll, McCrone et al., 2016; Hudson et al., 2017). In the feasibility study, potential participants were recruited to the study by completing self-report mood questionnaires on a computer tablet (iPad) while attending for dialysis. Nurses were responsible for administering the tablet to complete the screening as part of a hospital-wide initiative to screen all patients with physical health conditions (Rayner et al., 2014). However, when the study was launched, several nurses refused to offer the screening to patients. In addition, when patients were offered the opportunity to complete screening for their mood, 26 percent reported finding the screening unacceptable and a further 30 percent found it unfeasible to complete while on dialysis (Hudson et al., 2017). At the end of the trial, in-depth qualitative patient and health care practitioner interviews were conducted to better understand these perceived barriers and facilitators to screening from both the health professional and the patient points of view. These findings were mapped onto the four normalization process theory constructs to inform how screening could be better implemented by focusing on service-level features. The findings and planned actions are summarized in Table 24.2. A key learning point from this work is that implementation needs to focus on the whole treatment pathway, from screening to referral, and not just the intervention itself.

24.6 Stakeholder Engagement at the RCT Evaluation and Implementation Phases More recently, another body of work focused on IBS has addressed how to move from a large effectiveness RCT to implementation. Over a number of

years, a CBT treatment specific for IBS was developed iteratively using the MRC framework, including substantial user input into the theory-based intervention, feasibility and efficacy RCTs, process analysis of mechanisms of change to test the original theory, and qualitative feedback from RCT participants (Bogalo & Moss-Morris, 2006; Chilcot & Moss-Morris, 2013; Everitt et al., 2010; Kennedy et al., 2005; Moss-Morris, McAlpine, Didsbury, & Spence, 2009; Tonkin-Crine et al., 2013). A key focus of this development was the web-based selfmanagement version of the CBT called Regul8. Findings from a large effectiveness RCT of CBT for IBS (N = 558) developed by Everitt et al. (2019). Individuals who had access to Regul8, alongside 2.5 hours of therapist support on the telephone, showed a significantly greater improvement in both primary outcomes (IBS symptom severity and the impact of the symptoms on their lives) up to twelve months’ follow-up when compared to patients receiving treatment as usual. Regul8 patients also showed significantly greater improvements in mood and ability to manage their IBS. CBT is rarely available to patients with IBS, who make up as many as 10–20 percent of the population (Lovell & Ford, 2012), therefore the scalability of CBT in this group is important. Despite now having a potentially scalable, evidence-based product developed over a twenty-year period, funders were not able to support the rolling-out of the intervention to patients. The cost of maintaining a digital treatment is considerable as there are issues of maintaining and storing data safely and the ability to innovate quickly as software browsers and hardware such as smartphones are ever-evolving (see Chapter 29, this volume). At this point, the research team involved in the development and evaluation of Regul8 realized that the stakeholder needed to be a commercial company with the financial backing and ability to bring Regul8 not only to the UK but, potentially, worldwide (Everitt et al., 2019). As this is an ongoing process, full details cannot be reported but the team learned the importance of engagement with a whole

Table 24.2 Patient and health care professional stakeholder views of screening for depression in acute care settings and suggested actions to increase its implementation Normalization Process Theory Processes Staff Stakeholder Perspectives Coherence

Cognitive participation

Collective action

Reflexive monitoring

Patient Stakeholder Perspectives

Planned Actions to Improve Screening Implementation

Improve patient coherence by providNurses viewed screening as a completely new Did not understand why they were being ing more information about screenscreened for depression intervention and an additional task ing in advance of their appointment Thought they were being asked to answer Understood potential value of identifying and train staff to introduce the purquestionnaires for a research study not to symptoms of depression and/or anxiety for pose of screening more clearly assess their own mental well-being improving patient care Mixed views. Some firmly believed that mind Clearly define job roles. Managers to Mixed views. Some firmly believed that include screening in job description and body should be managed together. screening for mental health was part of their for all new members of the team Others viewed each health care professional role, while others felt that it was not in their Improve patient coherence further by as having their own expertise and the remit outlining how the role of screening boundaries should not be crossed is to identify depression and anxiety so that a referral to an expert can be made Lack of clarity about who staff could refer to Mixed views. Some patients viewed screening Generate clear referral pathways between mental and physical health as helpful for opening up the conversation should they identify mild to moderate care pathways. Resource permitting about depression. Others felt as though symptoms of depression and anxiety. This integrate psychologists within the more time was needed deterred some from screening. However, physical health care setting staff were clear on what to do if patients were acutely suicidal Majority of patients reported that staff rarely Improve methods of communication Staff had no way of knowing if a patient between physical and mental health had the time to talk about their screening engaged with their recommendations if care settings results with them referred outside of the immediate care team Improve staff training to ensure that all patients have their results discussed with them even if they do not screen positive for depression

358

JOANNA L. HUDSON, ZOE MOON, LYNDSAY D. HUGHES, AND RONA MOSS-MORRIS

new group of stakeholders. This included university and legal commercialization expertise and finding a commercial company to work with in partnership who understands the importance of the evidence base of the product. With digital products, the object of the intervention, as defined by normalization process theory, also needs to include data safety and regulatory procedures, which are different across countries, and licensing agreements with the health service regulatory approvals allowing a medical device to be covered by health insurance companies. Finally, once the product is launched and made readily available, finding ways to evaluate the ongoing effectiveness in the real world should be a priority. Here the principles of reflexivity delineated by normalization process theory, as described in Section 24.2, are important. Monitoring the product in real-world practice can include the routine collection of very large data, which can enable further improvements and innovations through the use of methods such as artificial intelligence.

24.7 Summary and Conclusion This chapter has outlined several ways of including stakeholders in the development of behavioral interventions and in building the evidence base of these products. The focus of the chapter has been on stakeholders in the health care context as this is where much of the work has been conducted, but these methods could be similarly applied to behavior change interventions in other domains such as public health, social policy, education, transport, and the environment. All the frameworks discussed, including the MRC framework, normalization process theory, and the person-centered approach, can be applied. The methods for stakeholder involvement have improved both the products developed and their potential to be incorporated in interventions within health services. In the next few years, it is expected that some of these products can be integrated into routine care. It is worth noting that high-quality and effective stakeholder involvement requires time

and financial commitment. Funders need to be prepared to support both. It is also important not only to engage with stakeholders who support behavior change interventions under development but to engage with those who see the difficulties and challenges in what is proposed so that the potential reach of the intervention if these issues can be resolved is greater.

References Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50, 179–211. https://doi.org/10.1016/ 0749-5978(91)90020-T Bogalo, L., & Moss-Morris, R. (2006). The effectiveness of homework tasks in an irritable bowel syndrome self-management programme. New Zealand Journal of Psychology, 35, 120–125. https://psycnet.apa.org/record/2007-01035-003 Brito, C., Portela, M. C., & Vasconcellos, M. T. L. d. (2014). Factors associated to persistence with hormonal therapy in women with breast cancer. Revista de saude publica, 48, 284–295. https://doi .org/10.1590/S0034-8910.2014048004799 Chilcot, J., & Moss-Morris, R. (2013). Changes in illnessrelated cognitions rather than distress mediate improvements in irritable bowel syndrome (IBS) symptoms and disability following a brief cognitive behavioural therapy intervention. Behaviour Research and Therapy, 51, 690–695. https://doi.org/ 10.1016/j.brat.2013.07.007 Craig, P., Dieppe, P., Macintyre, S., Michie, S., Nazareth, I., & Petticrew, M. (2008). Developing and evaluating complex interventions: The new Medical Research Council guidance. British Medical Journal, 337, a1655. https://doi.org/ 10.1136/bmj.a1655 Duncan, M., Moschopoulou, E., Herrington, E. et al. (2017). Review of systematic reviews of nonpharmacological interventions to improve quality of life in cancer survivors. BMJ open, 7, e015860– e015860. https://doi.org/10.1136/bmjopen-2017015860 EBCTCG (Early Breast Cancer Trialists’ Collaborative Group). (1998). Tamoxifen for early breast cancer: An overview of the randomised trials. The

Engagement of Stakeholders

Lancet, 351, 1451–1467. https://doi.org/10.1016/ S0140-6736(97)11423-4 EBCTCG (Early Breast Cancer Trialists’ Collaborative Group). (2011). Relevance of breast cancer hormone receptors and other factors to the efficacy of adjuvant tamoxifen: Patient-level meta-analysis of randomised trials. The Lancet, 378, 771–784. https://doi.org/10.1016/S0140-6736(11)60993-8 Eccles, M. P., & Mittman, B. S. (2006). Welcome to Implementation Science. Implementation Science, 1, 1. https://doi.org/10.1186/1748-5908-1-1 Everitt, H. A., Landau, S., O’Reilly, G. et al. (2019). Assessing telephone-delivered cognitive– behavioural therapy (CBT) and web-delivered CBT versus treatment as usual in irritable bowel syndrome (ACTIB): A multicentre randomised trial. Gut, 68, 1613–1623. https://doi.org/10.1136/ gutjnl-2018-317805 Everitt, H. A., Moss-Morris, R. E., Sibelli, A. et al. (2010). Management of irritable bowel syndrome in primary care: Feasibility randomised controlled trial of mebeverine, methylcellulose, placebo and a patient self-management cognitive behavioural therapy website. (MIBS trial). BMC Gastroenterology, 10, 136–145. https://doi.org/ 10.1186/1471-230X-10-136 Finch, T. (2008). Teledermatology for chronic disease management: Coherence and normalization. Chronic Illness, 4, 127–134. https://doi.org/ 10.1177/1742395308092483 Horne, R., Weinman, J., & Hankins, M. (1999). The beliefs about medicines questionnaire: The development and evaluation of a new method for assessing the cognitive representation of medication. Psychology and Health, 14, 1–24. https://doi.org/10.1080/08870449908407311 Hudson, J. L., Moss-Morris, R., Game, D., Carroll, A., & Chilcot, J. (2016). Improving Distress in Dialysis (iDiD): A tailored CBT self-management treatment for patients undergoing dialysis. Journal of Renal Care, 42, 223–238. https://doi .org/10.1111/jorc.12168 Hudson, J. L., Moss-Morris, R., Game, D., Carroll, A., McCrone, P. et al. (2016). Improving distress in dialysis (iDiD): A feasibility two-arm parallel randomised controlled trial of an online cognitive behavioural therapy intervention with and without

359

therapist-led telephone support for psychological distress in patients undergoing haemodialysis. BMJ Open, 6. https://doi.org/10.1136/bmjopen2016-011286 Hudson, J. L., Moss-Morris, R., Norton, S. et al. (2017). Tailored online cognitive behavioural therapy with or without therapist support calls to target psychological distress in adults receiving haemodialysis: A feasibility randomised controlled trial. Journal of Psychosomatic Research, 102, 61–70. https://doi.org/10.1016/j .jpsychores.2017.09.009 Kennedy, T., Jones, R., Darnley, S., Seed, P., Wessely, S., & Chalder, T. (2005). Cognitive behaviour therapy in addition to antispasmodic treatment for irritable bowel syndrome in primary care: Randomised controlled trial. British Medical Journal, 331, 435. https://doi.org/10.1136/ bmj.38545.505764.06 Leventhal, H., Bodnar-Deren, S., Breland, J. Y. et al. (2012). Modeling Health and Illness Behaviour: The approach of the Commonsense Model. In A. B. Baum, T. A. Revenson, & J. Singer (Eds.), Handbook of Health Psychology (pp. 3–36). New York: Psychology Press. Lovell, R. M., & Ford, A. C. (2012). Global prevalence of and risk factors for irritable bowel syndrome: A meta-analysis. Clinical Gastroenterology and Hepatology, 10, 712–721, e714. https://doi.org/ 10.1016/j.cgh.2012.02.029 Mair, F. S., May, C., O’Donnell, C., Finch, T., Sullivan, F., & Murray, E. (2012). Factors that promote or inhibit the implementation of e-health systems: An explanatory systematic review. Bulletin of the World Health Organization, 90, 357–364. https:// doi.org/10.2471/BLT.11.099424 May, C. (2013). Towards a general theory of implementation. Implementation Science, 8, 18. https://doi.org/10.1186/1748-5908-8-18 May, C., Finch, T., Mair, F. et al. (2007). Understanding the implementation of complex interventions in health care: The normalization process model. BMC Health Services Research, 7, 148. https:// doi.org/10.1186/1472-6963-7-148 May, C., Rapley, T., Mair, F. et al. (2015). Normalization process theory on-line users’ manual, toolkit and NoMAD instrument.

360

JOANNA L. HUDSON, ZOE MOON, LYNDSAY D. HUGHES, AND RONA MOSS-MORRIS

Normalization Process Theory. Website. www .normalizationprocess.org May, C. R., Mair, F., Finch, T. et al. (2009). Development of a theory of implementation and integration: Normalization process theory. Implementation Science, 4, 29. https://doi.org/ 10.1186/1748-5908-4-29 Moon, Z., Moss-Morris, R., Hunter, M. S., & Hughes, L. D. (2017). Understanding tamoxifen adherence in women with breast cancer: A qualitative study. British Journal of Health Psychology, 22, 978– 997. https://doi.org/10.1111/bjhp.12266 Moon, Z., Moss-Morris, R., Hunter, M. S., & Hughes, L. D. (2019). Development of a self-management intervention to improve tamoxifen adherence in breast cancer survivors using an intervention mapping approach. Unpublished manuscript, King’s College London. Moon, Z. E., Moss-Morris, R., Hunter, M. S., Goodliffe, S., & Hughes, L. D. (2019). Acceptability and feasibility of a selfmanagement intervention for women prescribed tamoxifen. Health Education Journal. https://doi .org/10.1177/0017896919853856 Moon, Z. E., Moss-Morris, R., Hunter, M. S., Norton, S., & Hughes, L. D. (2019). Non-adherence to tamoxifen in breast cancer survivors: A 12 month longitudinal analysis. Health Psychology, 38, 888–899. https://doi.org/10.1037/hea0000785 Moore, G. F., Audrey, S., Barker, M. et al. (2015). Process evaluation of complex interventions: Medical Research Council guidance. BMJ: British Medical Journal, 350, h1258. https://doi.org/ 10.1136/bmj.h1258 Moss-Morris, R., McAlpine, L., Didsbury, L. P., & Spence, M. J. (2009). A randomized controlled trial of a cognitive behavioural therapy-based selfmanagement intervention for irritable bowel syndrome in primary care. Psychological Medicine, 40, 85–94. https://doi.org/10.1017/ S0033291709990195 Murray, E., Treweek, S., Pope, C. et al. (2010). Normalisation process theory: A framework for

developing, evaluating and implementing complex interventions. BMC Medicine, 8, 63. https://doi.org/10.1186/17417015-8-63 NIHR (National Institute for Health Research). (2019). Patient and Public Involvement in Health and Social Care Research: A handbook for researchers. www.nihr.ac.uk/about-us/CCF/fund ing/how-we-can-help-you/RDS-PPI-Handbook2014-v8-FINAL.pdf NHS (National Health Service). (2019). The NHS Long-Term Plan. www.longtermplan.nhs.uk/pub lication/nhs-long-term-plan/ Rayner, L., Matcham, F., Hutton, J. et al. (2014). Embedding integrated mental health assessment and management in general hospital settings: Feasibility, acceptability and the prevalence of common mental disorder. General Hospital Psychiatry, 36, 318–324. https://doi.org/10.1016/ j.genhosppsych.2013.12.004 Skivington, K., Matthews, L., Craig, P., Simpson, S., & Moore, L. (2018). Developing and evaluating complex interventions: Updating Medical Research Council guidance to take account of new methodological and theoretical approaches. The Lancet, 392, S2. Meeting abstract: Public Health Science 2018, Belfast, Northern Ireland, November 23, 2018. http://dx.doi.org/10.1016/s0140-6736(18) 32865-4 Tonkin-Crine, S., Bishop, F. L., Ellis, M., Moss-Morris, R., & Everitt, H. (2013). Exploring patients’ views of a cognitive behavioral therapy-based website for the self-management of irritable bowel syndrome symptoms. Journal of Medical Internet Research, 15, e190–e190. https://doi.org/10.2196/ jmir.2672 Yardley, L., Morrison, L., Bradbury, K., & Muller, I. (2015). The person-based approach to intervention development: Application to digital health-related behavior change interventions. Journal of Medical Internet Research, 17, e30–e30.

25 Maximizing User Engagement with Behavior Change Interventions Lucy Yardley, Leanne Morrison, Ingrid Muller, and Katherine Bradbury

Practical Summary Interventions can only produce positive behavior change if people actually engage with them. Often people find interventions unappealing, unpersuasive, or difficult to put into practice. This chapter describes the steps intervention developers should take to make sure that all target users will find their intervention easy and enjoyable to engage with, motivating, and practical – whether they are children or older people, students or employees, patients or health care professionals. By gathering in-depth feedback about users’ views of all elements of interventions, the developer can understand what users need and want. Gaining this understanding is vital to avoid wasting resources developing and administering interventions that people do not want or cannot engage with. Careful development with a wide range of users is especially important to ensure that people from underserved groups can engage with the intervention, including people with lower levels of education, with limited resources, or from different cultures.

25.1 Introduction Many psychological theories of behavior change (for more details, see Chapter 1; see also the chapters in Part I) were originally developed and tested in controlled conditions and samples of people who were more educated and committed than the wider population – for example, laboratory studies of students or surveys and trials in motivated or paid volunteers (e.g., van Beurden et al., 2016).1 This research has provided valuable evidence about what behavior change techniques (for more details, see Chapter 20, this volume) may have efficacy – that is, can change behavior if applied correctly (Singal, Higgins, & Waljee, 2014). Yet these behavior change techniques will only be effective in the real world if users engage with them as intended. Often the effectiveness of

theory- and evidence-based behavior change interventions is undermined by poor user engagement. Poor engagement may take the form of low initial uptake, high early dropout, or failure to sustain long-term adherence to the techniques intended to support behavior change (Yardley et al., 2016). The aim of this chapter is to describe how to increase the effectiveness of behavior change interventions by designing and optimizing 1

It is important to note that not all theories have been developed in experimental contexts and in educated, motivated samples. For example Prochaska and DiClemente’s transtheoretical model (see Chapter 10, this volume) and ecological theories of behavior change (see Chapter 17, this volume), and practices like motivational interviewing (see Chapter 45, this volume), were developed in nonexperimental, clinical contexts. https://doi.org/10.1017/9781108677318.025

362

LUCY YARDLEY, LEANNE MORRISON, INGRID MULLER, AND KATHERINE BRADBURY

interventions to make them accessible and engaging to the target users – whether these are members of the public or the people who themselves deliver interventions, such as teachers, doctors, and providers of social care. First, key methods for designing engaging interventions are introduced; these involve gaining an understanding of the needs and characteristics of the target users and then tailoring and adapting the intervention to meet the needs of different users. The chapter then outlines methods for optimizing engagement using qualitative and quantitative data about how users engage with it. Lastly, the chapter discusses harnessing social and environmental support to sustain behavior.

25.2 Creating Interventions that Will Maximize Engagement 25.2.1 Needs Assessment and Characterizing Users When beginning to plan a behavior change intervention it is essential to conduct a needs assessment to identify what an intervention must include in order to overcome barriers to engagement (Yardley et al., 2015). For example, an assessment might seek to understand the needs of schoolchildren who demonstrate challenging behavior in class, to identify potential barriers to their engagement with an intervention designed to help them to manage their emotions and engage better with learning activities. Conducting a needs assessment can help intervention developers to: •

•

•

Select theory- and evidence-based intervention techniques that will be most acceptable, salient, and feasible for a target population. Avoid the inclusion of intervention techniques that might be off-putting, unpersuasive, or unnecessary. Suggest new intervention features (which are not yet evidence-based as they have not previously been studied).

The person-based approach provides a guide to conducting needs analyses to inform the planning of behavior change interventions (Yardley et al., 2015). The first step is to create a detailed understanding of who the target intervention users are and what challenges they might face in trying to change the target behavior (Yardley et al., 2015). This process shares similarities with the development of “personas” and “scenarios” sometimes used by computer scientists developing digital products to increase engagement with them (Van Velsen et al., 2013). However, the person-based approach, which is rooted in the discipline of health psychology, focuses more specifically on factors important to engagement with behavior change. Qualitative evidence (e.g., collected through interviews, focus groups, or observations of behavior) is particularly useful for creating an in-depth understanding of target users as it provides detailed insight into how users live, think, and feel. If sufficient qualitative research exists about a target user group, then this can be synthesized. If not, then it is necessary to conduct a primary qualitative study to understand the needs of a target population (for more details on qualitative approaches, see Chapter 30, this volume). Thinking about the example of examining the needs of schoolchildren who demonstrate challenging behavior in class, qualitative interviews with the children, their teachers, and parents would be beneficial, as would observations of classroom behavior. Important things to understand about a user group include their: • • • • •

Demographic characteristics and current behavior. Beliefs and feelings about behaviors targeted by the intervention. Needs, capabilities, and preferences. Social identities and context. Environmental barriers and facilitators.

Creating a detailed understanding of target users means that key challenges that the user

Maximizing User Engagement with Behavior Change Interventions

might face in engaging with an intervention can be identified. Based on this, “guiding principles” can be formulated that provide a short summary of what an intervention needs to contain in order to maximize engagement (Yardley et al., 2015). Guiding principles are made up of two parts: 1. Intervention design objectives – what an intervention needs to do in order to overcome each challenge (e.g., if a key challenge is that a child feels anxious, which is causing them to be disruptive in the classroom, then a design objective might be to provide support to reduce anxiety). 2. Key intervention features that will be employed to achieve each intervention design objective.

363

Key intervention features might include particular arguments an intervention needs to draw on to be persuasive or to avoid being off-putting, how the intervention needs to be delivered (e.g., through a particular technology or in a particular setting), or behavior change techniques needed to maximize engagement. The particular features needed will vary depending on the needs of the target group, behavior(s) being targeted, and context. Sidebar 25.1 illustrates the development of guiding principles for an intervention for cancer survivors. Examples of guiding principles are available in Appendixes 25.1 and 25.2 (supplemental materials).

Sidebar 25.1 Developing guiding principles for an intervention for cancer survivors

An intervention was developed that aimed to improve quality of life and prevent cancer recurrence in people who have experienced breast, colorectal, or prostate cancer. A literature review and interviews with cancer survivors suggested that cancer survivors can be sensitive to information that suggests that their behavior might have been in any way responsible for their illness (Corbett et al., 2018). Although unhealthy behaviors (e.g. poor diet, sedentary behavior) are involved in causing breast, colorectal, and prostate cancers, survivors wanted to view cancer as simply “bad luck” (Corbett et al., 2018). This was important to understand as it meant that arguments about reducing the chances of cancer recurrence by making behavior changes could be off-putting, as they would present survivors with evidence that their previous or current behavior might have contributed to their cancer. At the same time, if people do not view a problem as under their control, then they have little motivation to attempt change. Therefore, a guiding principle to address this barrier to engagement was formulated: The intervention design objective was to ensure that promoting behavior change did not stigmatize cancer survivors’ current behavior. Key intervention features designed to achieve this objective were to: 1. Avoid arguments that could be viewed as blaming people for their cancer (this included avoiding mentioning that behavior changes could reduce the chances of cancer recurrence). 2. At the same time, show users the benefits of behavior change – benefits that were more acceptable to cancer survivors were drawn on, such as improvement in symptoms (such as fatigue/distress) and protection against other illnesses.

364

LUCY YARDLEY, LEANNE MORRISON, INGRID MULLER, AND KATHERINE BRADBURY

25.2.2 Tailoring and Self-Tailoring Individuals who use, deliver, or support the use of the same intervention can be quite different. They may: •

•

• •

Have different demographic characteristics (e.g., gender, age, ethnicity, socioeconomic background, literacy or education levels). Think differently about behavior (e.g., in terms of motivation or confidence to do and support it). Be starting from different points (e.g., baseline patterns of behavior, prior experience). Be trying to change or support behavior in different environmental contexts (e.g., availability of support, nature of daily routines).

Interventions that take these differences into account by tailoring the information, support, and advice delivered can maximize user engagement and improve the likelihood of positive behavior change. Tailoring refers to the practice of delivering different information relevant to different users who vary on particular characteristics (Kreuter et al., 1999). For example, research has shown that better energy savings are seen when individuals are provided with behavioral suggestions that are tailored to their current energy-related behaviors, such as encouraging individuals to use energy-saving lightbulbs only if they were not already using them (Abrahamse et al., 2007). The need to tailor an intervention may become apparent through needs assessment and characterization of users (see Section 25.2.1). Tailoring intervention content can maximize engagement with behavior change interventions in two important ways: 1. It can improve personal relevance, helping users quickly and easily find the advice and support that best fit their specific circumstances. 2. It can stop users seeing information or advice that is likely to be off-putting or even harmful (e.g., advice to undertake activities that are beyond a user’s current level of physical capability).

However, tailoring is not always straightforward. Often, it is not possible to tailor an intervention according to all the possible ways in which individual users of an intervention differ. Data collected during needs assessment (and optimization – see Section 25.3) can tell an intervention designer if particular characteristics reflect important differences in how individuals or groups of users are likely to react to and engage with the intervention or behavior change process. For example, younger individuals with Type 2 diabetes using an online intervention to encourage physical activity described finding it difficult to relate to images of older people. The decision was then taken to tailor the visual content of the intervention by age (Rowsell et al., 2015). Tailoring does not always have the desired effect on engagement. Sometimes users feel that they have not provided enough detailed or accurate information about themselves for the intervention to deliver the best advice (Morrison et al., 2014). A potential solution is to allow self-tailoring. This technique invites users to choose what information to look at and try out different tools and techniques and lets them decide for themselves what is most relevant and useful (Lorig & Holman, 2003).

25.2.3 Targeting and Adapting to Different Populations Tailoring interventions is a method used to modify, or personalize, aspects within an intervention to the individual user. Targeting, on the other hand, refers to the specific group or population that the whole intervention is aimed at. Interventions may target different populations, in which case they may need adaptation for different user groups. This can be especially important to ensure that interventions are engaging for underserved populations, such as socioeconomically deprived communities or ethnic minorities. Adaptation may involve taking into account

Maximizing User Engagement with Behavior Change Interventions

365

Sidebar 25.2 Illustration of using nonjudgmental language

An intervention was developed to increase physical activity for people with low health literacy and Type 2 diabetes (Rowsell et al., 2015). A core part of the intervention was a knowledge quiz, designed to highlight some lesser known benefits of physical activity. Initial qualitative research with users revealed that people with lower literacy levels were upset when informed that their answer to the quiz was incorrect. To address this, if an incorrect response was given the correct answer was introduced by “Surprise!” rather than “Wrong” and with reassurance that their incorrect answer was reasonable – for example, “it is not surprising that you think that controlling your sugar levels is the most important thing to do in diabetes; in the past this is what doctors thought too.” Participants also preferred nondirective language when giving advice. For example, instead of saying “you should” or “you must,” advice was phrased as “you could” or “many people find it helpful to …”

differences in language, culture, or context (Harris, Sillence, & Briggs, 2011). When developing or adapting an intervention, steps can be taken to make the intervention accessible and usable for a wide range of people, including those with lower levels of literacy or cognitive impairments. Trust and credibility of interventions are particularly important for facilitating engagement in underserved populations (Sbaffi & Rowley, 2017). These can be promoted by providing details about the team who developed or are delivering the intervention (which should include members of the target population) and offering reassurance about what engaging with the intervention will involve (e.g., details of how personal data will be used). It is also useful to minimize the reading burden by only presenting people with relevant information and using audiovisual formats and tailoring where appropriate (Mayer, 2002; Rowsell et al., 2015). Following best-practice guidance (e.g., Krug, 2006; Mahr, 2014) will maximize the accessibility of any written material (e.g., using simple language, short sentences, large fonts, clear layouts, bullet points, and signposting).

It is important to use nonjudgmental language at all times, to ensure people feel they are respected and to avoid provoking negative feelings (see Sidebar 25.2). It is often helpful to give clear explanations and evidence to support the recommended behavior change and suggest trying out the behavior for a limited period. This invites people to decide for themselves if the suggestion will benefit them, which builds trust and provokes less resistance (Yardley et al., 2015).

25.3 Optimizing Interventions to Maximize Engagement 25.3.1 Optimization of Interventions Using Qualitative Methods Once a behavior change intervention has been designed (for more details on intervention design, see Chapter 21, this volume) and prototype materials developed (e.g., an app, poster, leaflet, or manual), it is vital to check how target users perceive the planned intervention components and prototype materials. Feedback from target users enables the identification of barriers to engagement, which can ensure that the intervention is

366

LUCY YARDLEY, LEANNE MORRISON, INGRID MULLER, AND KATHERINE BRADBURY

acceptable, feasible, persuasive, enjoyable, motivating, engaging, and, crucially, as likely to change behavior as possible. This is an iterative process and several iterations are usually needed to develop an optimal intervention. Various approaches exist for optimizing interventions by gaining feedback from target users. For instance, usability testing, which is based in the discipline of human-computer interaction, is often used to explore the attractiveness and ease of use of digital interventions. The person-based approach focuses on whether an intervention engages users with the elements of the intervention that are likely to change behavior and so is particularly relevant to developing behavior change interventions (Yardley et al., 2015). The person-based approach uses qualitative feedback to optimize prototype interventions as this provides very detailed feedback about potential barriers to behavior change, including unexpected barriers, which an intervention developer might not have thought to ask about in a questionnaire. Two different types of qualitative study design (see also Chapter 30, this volume) are valuable for optimizing behavior change interventions: (1) think-aloud studies and (2) longitudinal qualitative studies.

25.3.1.1

disengagement when there is no supporter present to help address concerns and maintain engagement. It is important to sample people from a variety of backgrounds, especially those with lower literacy levels, to ensure the intervention will be accessible and acceptable to a wide range of target users. It is useful to create a table of the feedback and how it can be addressed to allow prioritization of modifications that are most important for behavior change (Bradbury et al., 2018). See Sidebar 25.3 for an example of how feedback from a think-aloud study was used to optimize an intervention to promote physical activity. An example interview guide from a think-aloud study is available in Appendix 25.3 (supplemental materials). An illustration of the table to support prioritization of changes is available in Appendix 25.4 (supplemental materials).

25.3.1.2

Longitudinal qualitative studies, in which target users engage with an intervention over several weeks and then provide feedback in an interview, are another valuable approach that can inform optimization. This study design has several advantages: •

Think-Aloud Studies

In think-aloud studies participants engage with prototype intervention materials while sitting with a researcher and provide immediate feedback on everything they engage with by saying their thoughts out loud as they try it out (Bradbury et al., 2018; Van den Haak et al., 2007). This ensures that problems with intervention content or format that could trigger early disengagement are identified. This is particularly useful for optimizing interventions that have written content, for example in booklets, posters, websites, or apps, as it is vital to identify any content that is confusing, distressing, or unconvincing, as this is likely to lead to early

Longitudinal Qualitative Studies

•

•

•

Participants engage with the intervention as they would normally, without the “artificial” situation of a researcher being present, which may affect or bias their responses by, for example, motivating them or inadvertently prompting them to engage more than they would on their own (Cotton & Gresty, 2006). Participants can try out behavior changes, which can identify barriers to actually engaging with the behavior change (e.g., using their car less) – which is not possible within think-aloud studies. This study design can highlight potential problems with sustaining behavioral engagement. Problems with implementation can be identified, such as health care practitioners not having enough time to sufficiently support patients.

Maximizing User Engagement with Behavior Change Interventions

367

Sidebar 25.3 Using think-aloud methods to optimize an activity planner

A key feature of an online intervention developed to promote increased physical activity was an interactive physical activity planner designed to help people form achievable plans to increase their current activity level by building on the activities they already do. People were given a list of common activities and asked to select the ones they currently do. They were then asked how many times a week they do each activity and for how many minutes each time. The website then calculated total activity levels and provided tailored activity recommendations based on their current activity level. Qualitative think-aloud interviews during the development phase enabled observation of people filling in the activity planner. Soon, it was discovered that people were struggling to complete the planner correctly. This was a big problem because it meant that people with sedentary lifestyles were receiving tailored feedback congratulating them on already being active enough. Think-aloud interviews highlighted how and why people were not filling in the planner correctly and the following steps were taken: 1. The original planner was presented as a single table where people were asked to fill in how long was spent per activity and on how many days. The first change made was to split this into two tables to make it easier to complete. 2. The original activity planner also included “climbing stairs” as one of the possible activities but users were not able to confidently or accurately estimate how much time was spent on this and the decision was made to remove “climbing stairs” from the activity planner. 3. Subsequent think-aloud interviews showed that the usability of the planner had been improved but that more changes were needed as people were still overestimating their activity levels. Further changes included adding guidance notes, such as “only count fast walking for at least 10 minutes non-stop” and “very gentle activities such as walking slowly or washing the dishes do not count!” 4. Finally, the algorithm was modified to raise the minimum total amount of physical activity required to receive feedback congratulating users that they are already active enough, to compensate for users overreporting activity levels. Further think-aloud interviews were completed after each change until the researchers were confident that the physical activity planner was as usable and accurate as possible. Qualitative think-aloud interviews were crucial for improving the feasibility of the activity planner, highlighting the importance of thoroughly testing intervention materials before dissemination or evaluation in a trial.

368

LUCY YARDLEY, LEANNE MORRISON, INGRID MULLER, AND KATHERINE BRADBURY

25.3.2 Optimization of Interventions Using Quantitative Methods Quantitative data can also provide valuable information about how or why an intervention did or did not change behavior (see also Chapter 22, this volume). This might include data on: • • •

•

User characteristics, attitudes, or beliefs that are expected to influence behavior. Views of or reactions to the intervention content. Usage of the intervention elements (e.g., printed materials, digital platforms, contact with intervention providers). Usage of behavior change techniques in daily life (such as pursuing goals, implementing plans).

The popularity of digital intervention platforms provides more opportunity to collect detailed data on usage patterns (see also Chapter 29, this volume). Yet interpreting usage data is not simple, as more or longer periods of intervention use may not always be optimal or necessary. Users may stop engaging with an intervention once they have found the advice or support they needed and have successfully changed their behavior (Yardley et al., 2016). Combining quantitative usage data with qualitative data on users’ experiences (see Section 25.3.1) can suggest ways in which the intervention can be adjusted to maximize future engagement. Study and trial design approaches are available to support the optimization of interventions by testing the effect on engagement (or other outcomes) of different intervention components or adapting the exposure to intervention components over time. Examples include the multiphase optimization strategy (MOST) and the sequential multiple assignment randomized trial (SMART; Collins, Murphy, & Strecher, 2007). These designs may be particularly useful when there is expected to be a strong or important effect of using different intervention versions (e.g., cost-effectiveness). Large samples are needed if only small differences between intervention versions are expected.

25.4 Providing Support for Sustained Engagement 25.4.1 Social Support One thing to consider when developing a behavior change intervention is whether target users might benefit from social support. This could include support from friends, family, colleagues, or professionals. Social support can help users to sustain behavior changes (see also Chapter 44, this volume). For example, reviews indicate that social support is associated with older adults engaging in more physical activity (Smith et al., 2017). Harnessing the support of target users’ existing social networks may be useful, since social networks have important influences on behavior – for example, voting behavior (Christakis & Fowler, 2009). Social influence approaches, such as an influential person demonstrating a behavior, stating the importance of a behavior, or making a public commitment to perform a behavior, can be effective in sustaining engagement with behavior change. For example, a meta-analysis showed that these three social influence approaches increase conservation of resources to protect the environment (Abrahamse & Steg, 2013). Adding human support may be especially important in digital interventions, where it can increase engagement and effectiveness (Baumeister et al., 2014). However, if delivered by professionals, then the cost of this additional support may need to be considered. Remote support (by email or phone) may be a cost-effective way of providing professional support, since it is cheaper than inperson support and can be equally effective (Little et al., 2016).

25.4.2 Harnessing Habit and Environment Behavioral change can involve one-off or infrequently occurring behaviors that are motivated by a deliberate, conscious choice (e.g., consenting to

Maximizing User Engagement with Behavior Change Interventions

organ donation; switching to solar energy). Alternatively, behavioral change can require a set of behaviors to be repeated regularly over time (e.g., weight management; lifestyle changes). These repeated behaviors need to be considered in the context of an individual’s dayto-day routine and physical and social environment (see Chapter 41, this volume). For instance, how do their friends and family behave? What organizational systems and structures influence their behavior? Habitual and environmental factors can either support or pose a barrier to engaging with behavior change. For individuals to successfully change behavior, there is a need to move from a deliberate, supported intention to change (typically facilitated by early engagement with an intervention) to automatically performed behavior that is sustained over the longer term. Habitual behavior is automatically prompted by the environment (e.g., a particular place or time of day) and so long-term engagement will likely be maximized if interventions are designed to fit with usual routines or the existing environment (see Chapter 41, this volume). Interventions are also more likely to be used and more easily scaled-up to reach larger numbers of people if they can be integrated within the existing social and organizational structures that individuals already engage with, such as their family, workplace, college, or health care system. Advancing technology now allows us to sense and measure habitual behavior or aspects of a person’s environment, such as activity levels, location, social connections, or time of day. This means that interventions can be tailored to deliver information and advice at just the right time, sometimes known as just-in-time adaptive interventions (Nahum-shani, Hekler, & Spruijt-Metz, 2015). For example, in a recent study, a mobile app was used to unobtrusively monitor the locations where users reported smoking, prior to a quit attempt. During their quit attempt, support messages were automatically triggered when the

369

app detected that the user had entered a location where they frequently smoked (Naughton et al., 2016). Location sensing has also been explored in the context of consumer behavior; “locationbased adverts” could be sent to or displayed on an individual’s mobile device when they enter specific locations (e.g., proximity to specific retailers or product offers) (Bauer & Strauss, 2016).

25.5 Implications for Research The methods used to ensure interventions are engaging are continuously evolving. For example, stakeholders are increasingly involved in research relating to intervention development and implementation. Consequently, current developments of the person-based approach are focusing on how best to combine stakeholder codesign with in-depth qualitative research. Stakeholder engagement with providers of interventions from the early stages of intervention development can help to ensure that existing systems and structures are harnessed appropriately, allowing the intervention to be sustained in future real-world implementation. Stakeholder codesign can also be ideal for quickly gaining suggestions and reactions relevant to design decisions from stakeholders with different perspectives and types of expertise – including views of the intervention modifications suggested by the person-based approach. The in-depth qualitative research employed by the person-based approach can complement codesign by allowing more in-depth investigation of a wider range of user experiences and perceptions over time. Another exciting field for future methodological development is the potential to optimize interventions efficiently by collecting data on uptake, usage, and outcomes in real-world implementation and using the findings for continuous quality improvement (Mohr et al., 2017). These methods are already used by many digital intervention providers to continuously improve engagement with apps and social media, are starting to be introduced

370

LUCY YARDLEY, LEANNE MORRISON, INGRID MULLER, AND KATHERINE BRADBURY

into evaluation of health care interventions (Friedman et al., 2017), and have great potential for optimizing behavioral interventions in the future.

25.6 Conclusion The key to maximizing engagement with interventions is deceptively simple: Listen to and understand the target users of the intervention and address their needs, concerns, and preferences. Intervention developers often underestimate the time and effort needed to optimize interventions and take shortcuts such as quickly consulting a small, atypical sample of users. Unfortunately, poor design and inadequate optimization simply result in interventions that fail, underperform, or are only engaging for a small proportion of potential users. The methods described in this chapter can be used flexibly and adapted to whatever time and resources are available in order to anticipate and overcome barriers to engagement and maximize the uptake and effectiveness of the intervention.

References Abrahamse, W., & Steg, L. (2013). Social influence approaches to encourage resource conservation: A meta-analysis. Global Environmental Change, 23, 1773–1785. https://doi.org/https://doi.org/ 10.1016/j.gloenvcha.2013.07.029 Abrahamse, W., Steg, L., Vlek, C., & Rothengatter, T. (2007). The effect of tailored information, goal setting, and tailored feedback on household energy use, energy-related behaviors, and behavioral antecedents. Journal of Environmental Psychology, 27, 265–276. https://doi.org/10.1016/ j.jenvp.2007.08.002 Bauer, C., & Strauss. (2016). Location-based advertising on mobile devices. Management Review Quarterly, 66, 159–194. https://doi.org/ 10.1007/s11301-015-0118-z Baumeister, H., Reichler, L., Munzinger, M., & Lin, J. (2014). The impact of guidance on Internet-based

mental health interventions: A systematic review. Internet Interventions, 1, 205–215. https://doi.org/ 10.1016/j.invent.2014.08.003 Bradbury, K., Morton, K., Band, R. et al. (2018). Using the Person-Based Approach to optimise a digital intervention for the management of hypertension. PloS ONE, 13, e0196868. https://doi.org/10.1371/ journal.pone.0196868 Christakis, N. A., & Fowler, J. H. (2009). Connected: The Surprising Power of Our Social Networks and How They Shape Our Lives. New York: Little, Brown. Collins, L. M., Murphy, S. A., & Strecher, V. (2007). The multiphase optimization strategy (MOST) and the sequential multiple assignment randomized trial (SMART): New methods for more potent eHealth interventions. American Journal of Preventive Medicine, 32, S112–S118. https://doi.org/10.1016/j.amepre.2007.01.022. Corbett, T., Cheetham, T., Müller, A. M. et al. (2018). Exploring cancer survivors’ views of health behaviour change: “Where do you start, where do you stop with everything?” PsychoOncology, 27, 1816–1824. https://doi.org/ 10.1002/pon.4732 Cotton, D., & Gresty, K. (2006). Reflecting on the think-aloud method for evaluating e-learning. British Journal of Educational Technology, 37, 45–54. https://doi.org/10.1111/j.14678535.2005.00521 Friedman, C. P., Allee, N. J., Delaney, B. C. et al. (2017). The science of learning health systems: Foundations for a new journal. Learning Health Systems, 1, e10020. https://doi.org/10.1002/ lrh2.10020 Harris, P. R., Sillence, E., & Briggs, P. (2011). Perceived threat and corroboration: Key factors that improve a predictive model of trust in internet-based health information and advice. Journal of Medical Internet Research, 13, e51. https://doi.org/10.2196/jmir.1821 Kreuter, M. W., Bull, F. C., Clark, E. M., & Oswald, D. L. (1999). Understanding how people process health information: A comparison of tailored and nontailored weight-loss materials. Health Psychology, 18, 487. https://doi.org/10.1037/ 0278-6133.18.5.487

Maximizing User Engagement with Behavior Change Interventions

Krug, S. (2006). Don’t Make Me Think! A Common Sense Approach to Web Usability. (2nd ed.). New York: New Riders. Little, P., Stuart, B., Hobbs, F. R. et al. (2016). An internet-based intervention with brief nurse support to manage obesity in primary care (POWeR+): A pragmatic, parallel-group, randomised controlled trial. The Lancet Diabetes and Endocrinology, 4, 821–828. https://doi.org/ 10.1016/S2213-8587(16)30099-7 Lorig, K. R., & Holman, H. R. (2003). Selfmanagement education: History, definition, outcomes, and mechanisms. Annals of Behavioral Medicine, 26, 1–7. https://doi.org/10.1207/ S15324796ABM2601_01 Maher, T. (2014). Plain English Campaign. Website. www.plainenglish.co.uk. Mayer, R. E. (2002). Multimedia learning. Psychology of Learning and Motivation, 41, 85–139. https:// doi.org/10.1016/S0079-7421(02)80005-6 Mohr, D. C., Lyon, A. R., Lattie, E. G., Reddy, M., & Schueller, S. M. (2017). Accelerating digital mental health research from early design and creation to successful implementation and sustainment. Journal of Medical Internet Research, 19, e153. https://doi.org/10.2196/ jmir.7725 Morrison, L., Moss-Morris, R., Michie, S., & Yardley, L. (2014). Optimizing engagement with Internetbased health behaviour change interventions: Comparison of self-assessment with and without tailored feedback using a mixed methods approach. British Journal of Health Psychology, 19, 839–855. https://doi.org/10.1111/bjhp.12083 Nahum-Shani, I., Hekler, E. B., & Spruijt-Metz, D. (2015). Building health behavior models to guide the development of just-in-time adaptive interventions: A pragmatic framework. Health Psychology, 34, 1209. https://doi.org/10.1037/ hea0000306 Naughton, F., Hopewell, S., Lathia, N. et al. (2016). A context-sensing mobile phone app (Q sense) for smoking cessation: A mixed-methods study. JMIR mHealth and uHealth, 4, e106. https://doi.org/ 10.2196/mhealth.5787 Rowsell, A., Muller, I., Murray, E. et al. (2015). Views of people with high and low levels of health

371

literacy about a digital intervention to promote physical activity for diabetes: A qualitative study in five countries. Journal of Medical Internet Research, 17, e230. https://doi.org/10.2196/ jmir.4999 Sbaffi, L., & Rowley, J. (2017). Trust and credibility in web-based health information: A review and agenda for future research. Journal of Medical Internet Research, 19, e218. https://doi.org/ 10.2196/jmir.7579 Singal, A. G., Higgins, P. D., & Waljee, A. K. (2014). A primer on effectiveness and efficacy trials. Clinical and Translational Gastroenterology, 5, e45. https://doi.org/10.1038/ctg.2013.13 Smith, G. L., Banting, L., Eime, R., O’Sullivan, G., & van Uffelen, J. G. (2017). The association between social support and physical activity in older adults: A systematic review. International Journal of Behavioral Nutrition and Physical Activity, 14, 56. https://doi.org/10.1186/s12966017-0509-8. van Beurden, S. B., Greaves, C. J., Smith, J. R., & Abraham, C. (2016). Techniques for modifying impulsive processes associated with unhealthy eating: A systematic review. Health Psychology, 35, 793. https://doi.org/10.1037/ hea0000337. Van den Haak, M. J., De Jong, M. D., & Schellens, P. J. (2007). Evaluation of an informational web site: Three variants of the think-aloud method compared. Technical Communication, 54, 58–71. Van Velsen, L., Wentzel, J., & Van Gemert-Pijnen, J. E. (2013). Designing eHealth that matters via a multidisciplinary requirements development approach. JMIR Research Protocols, 2, e21. https://doi.org/10.2196/resprot.2547 Yardley, L., Morrison, L., Bradbury, K., & Muller, I. (2015). The Person-Based Approach to intervention development: Application to digital health-related behavior change interventions. Journal of Medical Internet Research, 17, e30. https://doi.org/10.2196/jmir.4055 Yardley, L., Spring, B. J., Riper, H. et al. (2016). Understanding and promoting effective engagement with digital behavior change interventions. American Journal of Preventive Medicine, 51, 833–842.

26 Cost-Effectiveness Evaluations of Behavior Change Interventions Tianjiao Wang, Martin Downes, Joshua Byrnes, and Paul Scuffham

Practical Summary There is an expanse of interventions or programs that have been developed to modify people’s behavior, all aimed at benefiting the individual and/or society more broadly. Precise assessment of efficacy and effectiveness of an intervention is needed to guide “clinical” decisions on whether the intervention should potentially be implemented or not. Moreover, owing to limited resources and budgets, governments, organizations, and key stakeholders need evidence that clearly demonstrates the benefits of the intervention are greater than the cost of the intervention itself. Such evidence is key when making decisions on whether to allocate future funds to an intervention. Economic evaluation in the form of cost-effectiveness analysis (CEA) helps provide that evidence and is considered essential to inform funding decisions. This chapter introduces theory and approaches to CEA and outlines methods and tools to help assess the quality of CEA. In addition, the chapter provides a discussion on challenges to conducting CEA for behavior change interventions and recommendations on how these should be conducted.

26.1 Introduction Behavior change interventions have made an indelible mark on addressing problems that require behavioral solutions, such as voting turnout (Neri, Leifer, & Barrows, 2016), saving money and prudent financial decisions (Fertig, Lefkowitz, & Fishbane, 2015), health-promoting behaviors (Hallsworth et al., 2016), adaptive education, criminal behaviors, and environmental behaviors (Battista et al., 2018). Such interventions can bring about population-level change that leads to net benefits to society, particularly because many of the aforementioned issues are associated with human cost in terms of suffering and ill-being. However, these interventions also come at a financial cost through, for example, the provision

of service and the delivery of intervention messages and other intervention components (see Chapters 19, 21, and 22, this volume). In the case of population-level interventions, such as campaigns and programs run by governmental organizations, funds are provided through public finances from tax revenues. While the general public may be largely supportive of some interventions (e.g., interventions to quit smoking or stop overconsumption of alcohol), they may not be supportive of others (e.g., education programs to avoid illegal fishing or sort refuse into recyclables and nonrecyclables) (Battista et al., 2018). Therefore, governments and policy makers may be unwilling to finance behavioral interventions https://doi.org/10.1017/9781108677318.026

Cost-Effectiveness Evaluations of Behavior Change Interventions

with public funds that are unlikely to gain public support and, thus, economic evidence may help to provide evidence for funding. Similarly, finance advisors in organizations such as schools and hospitals are often presented with behavioral programs that could provide benefit; however, budgetary constraints may place limits on decisions to implement interventions and/or the choices made over which interventions to support (WHO, 2016). These considerations for government personnel and advisors make economic evaluation of behavioral interventions essential to inform decisions on whether to implement an intervention and to ensure that the maximum benefit is achieved for available funds. Economic evaluation forms part of the systematic evaluation of behavior change interventions, which should complement other forms of evaluation, such as efficacy or effectiveness evaluation and process evaluation (see Chapter 22, this volume). Economic evaluation is often considered an essential part in conducting high-quality evaluation of a behavioral intervention and provides important evidence for decision-making when it comes to implementing behavior change interventions (Fragoulakis et al., 2015). Methods and theory of cost-effectiveness evaluation have been widely applied in academic research and policy decisionmaking and in multiple fields such as health care, transportation, education, and environment (AlayliGoebbels et al., 2013). A variety of forms of economic evaluation exist. One of the most prominent and frequently applied is cost-effectiveness analysis (CEA), which aims to evaluate the costs and benefits of interventions and alternative options. CEA measures the relative costs and effects of the intervention in question and compares it to the costs and effects of alternatives, including, for example, current practice or “usual practice,” with results presented as incremental cost-effectiveness ratios (ICERs; Drummond et al., 2015). For example, an environmental behavior change intervention may look at the cost per reduction

373

in kilowatts used. Cost utility analysis is another form of economic evaluation, where the outcome is quality adjusted life years (QALY). This analysis is commonly regarded as a special type of CEA often used for health care interventions. QALYs can be generated by calculating the products of time spent and utility values in the health state (Mpofu, 2014). Yet another form of economic evaluation is cost-benefit analysis, which measures both costs and benefits by monetary units that can be compared across different fields (Drummond et al., 2015). For example, if the benefits of reducing the mortality rate by intervention A and that of improving air quality by intervention B can be measured by monetary units, the process of making decisions on budget allocation across various aspects of human life is more straightforward. Although this approach at face value seems very useful, it is often criticized for measuring every aspect of life in monetary units (Gray et al., 2011). The aim of this chapter is to outline the methods and approaches to CEA of behavior change interventions and discuss the role of economic evaluation in this setting. The chapter starts by presenting a framework for conducting economic evaluations of behavior change interventions. The framework sets out how to identify participants, interventions, comparators, and outcomes for economic evaluation studies. The chapter then provides an appraisal of CEA by applying the consolidated health economic evaluation reporting standards (CHEERS) checklist (Husereau et al., 2013). Finally, it discusses the implications of, and recommendations for, the application of CEA, including a discussion of the appropriate measurement of benefits, feasible model approaches, and issues underlying political considerations when funding behavior change interventions.

26.2 A Framework for Conducting Economic Evaluations A fundamental skill required before an economic evaluation can be conducted, and which facilitates

374

TIANJIAO WANG, MARTIN DOWNES, JOSHUA BYRNES, AND PAUL SCUFFHAM

the literature searching for evidence synthesis that is used in CEA, is the formulation of a well-built research question (Schardt et al., 2007). The population, intervention, comparator, and outcome (PICO) framework is commonly applied and assists in organizing the research question into a searchable query (e.g., Huang, Lin, & DemnerFushman, 2006). Each element of the PICO should be clearly defined with respect to the targeted behavior change intervention. In regards to the first element, population, demographic characteristics of the participants are considered. This includes, for example, age, gender, ethnicity, socioeconomic status, and other key demographic factors relevant to the target behavior to be changed. Regarding the second element, intervention, characteristics of the intervention are identified. For example, five strategies to change behavior have been identified based on economic and behavior economic theory: providing information, financial incentives, taxes and subsidies, regulations, and nudge interventions (for further discussion, see Chapter 42, this volume). The third element, comparator, reflects what, if anything, the intervention is being compared with – for example, another treatment, a standard treatment such as “usual practice,” or “doing nothing” (for further discussion on types of evaluation designs, see Chapter 22, this volume). The last element, outcome, reflects the target problem-focused outcome of the intervention, which could be changes in the target behavior itself (e.g., participating in more physical activity, smoking fewer cigarettes, turning off lights when leaving a room) or some index or meaningful outcome that indicates behavior change (e.g., lowered cholesterol levels, less money spent on cigarettes, reduced energy consumption). Outcomes can also be more complex, depending on the target behavior, such as decreases in the crime rate, improvements in education quality, reductions in the smoking uptake rate, or decreases in alcohol misuse. What is important to consider is that, in an economic framework, the outcomes need to be appropriate for developing the economic

analysis as well as relevant for decision-making when using a CEA (e.g., cost per light turned off, cost per kilowatt reduction, cost per reduction in CO2 emissions). In addition to the PICO elements, another consideration for a cost-effectiveness framework is perspective. This refers to the standpoint at which costs and outcomes are realized. For example, societal perspective (considering all possible costs and benefits in the analysis), payer perspective (government or insurance), or person perspective (participant or patient). Generally, in the case of interventions rolled out at a national level, the perspective is the payer (Mpofu, 2014). For example, if a government is interested in reducing carbon emissions and is considering implementing an intervention aimed at targeting this, then a cost per reduction in CO2 emissions may be important. This is not to say such interventions cannot also take a person or societal perspective. For example, the person perspective might be examining the cost savings associated with reduced energy consumption and demonstrating the cost savings associated with the time it takes to turn off the light (i.e., cost per minute). From a societal perspective, it might be examining air quality and how society in general would benefit from improved environmental quality (e.g., reduced illness, reduction in absence from work). Thus, depending on the perspective taken, different CEA findings might be observed. Another consideration for a costeffectiveness framework is time. In an economic evaluation, this is considered to be the time horizon (Drummond et al., 2015). For example, if an intervention is likely to have ongoing benefits (or harms), these need to be considered in the CEA analysis. When conducting the CEA, the time horizon is declared and often based on the evidence that is available for the intervention. This can be a short (e.g., weeks or months, based on the duration of a trial) or long (e.g., ten years or a lifetime, depending on whether the benefit is expected to last that long) time horizon.

Cost-Effectiveness Evaluations of Behavior Change Interventions

This section presented a framework for conducting economic evaluations of behavior change interventions. Specifically, the framework outlined the PICO model, which sets out how to identify participants, interventions, comparators, and outcomes for economic evaluation studies. In Section 26.3, an overview of economic theory and methodology used when developing CEA is discussed.

26.3 Economic Theory and Methodology of CEA 26.3.1 CEA as a Form of Economic Evaluation Neoclassical economics, often considered the theoretical foundation of economic evaluation, was developed on the basis of several assumptions (Perloff, 2015; see also Chapter 42, this volume). Neoclassical economics suggests that individuals, when considering various options, are usually clear on their preferred tastes and preferences. In addition, individuals are believed to be able to rank these options according to their preferences, and then make decisions on which can give them greatest pleasure (i.e., rational consumers). Individuals are also believed to hold a “more is better” belief; that is, more consumption of goods means more welfare (Binger & Hoffman, 1988). On the other hand, owing to the nature of limited resources and budget constraint, people could not have infinite welfare from consumption. Therefore, they maximize their welfare by comparing costs and benefits among different alternatives (Perloff, 2015). Economists apply ordinal numerical values (i.e., a utility function) to predict individuals’ preferences and tastes (Perloff, 2015). Consequently, the theoretical assumption underlying economic evaluation is that individuals are rational and aim to maximize their utility. In other words, they will select a particular option only when the benefits of doing so outweigh the costs (Battista et al., 2018).

375

While the above has been heavily criticized for being unrealistic at an individual level, the principles of such a structured, rational approach to decision-making is often heralded as what is expected of government, policy, and organizational decision makers (Drummond et al., 2015; Folland, Goodman, & Stano, 2016; see also Chapter 42, this volume). Economic evaluation provides a systematic structure to compare between various options by ranking them in terms of costs and benefits. Researchers in many aspects of policy, such as health care (see Sidebar 26.1), education, and environment evaluating the introduction of programs aimed at social benefit, such as behavior change interventions, therefore, seek a completed and evidence-based structure to guide decision makers (Battista et al., 2018; Hallsworth et al., 2016).

26.3.2 Measurement of Costs and Benefits Measuring costs and benefits of interventions is another important consideration in the economic evaluation process. This is not just simply adding prices and numerical values given the chosen perspective as discussed in Section 26.2, it requires comprehensive consideration in the data collection process (Gray et al., 2011). For example, researchers should identify which resource categories are to be included (e.g., types of outcome measure), what instruments are to be used to collect data (e.g., data records or surveys), and what definition of cost units will be used (Gray et al., 2011). When estimating costs, the time horizon and how the current monetary value can be applied to future value need to be considered. According to time preference theory, the monetary value in the future needs to be discounted compared to the value at present (Bhattacharya, Hyde, & Tu, 2013). The calculation formula is displayed as below (Gray et al., 2011): Cp ¼

Cf1 ð1þrÞ

þ

Cf2 ð1þrÞ2

þ…þ

Cfn ð1þrÞn

376

TIANJIAO WANG, MARTIN DOWNES, JOSHUA BYRNES, AND PAUL SCUFFHAM

Sidebar 26.1 The increase of health care expenditure and the application of economic evaluation

Economic evaluation has been widely applied in health care, as many countries face economic and political pressure regarding the allocation of their limited health budgets (WHO, 2016). In 2017, the total health expenditure for the top one-third of Organisation for Economic Co-operation and Development (OECD) countries, accounted for 10 percent or more of their gross domestic product (GDP) (OECD, 2017). The percentage in US is the highest in these countries, accounting for 17.1 percent of its GDP (OECD, 2017). In addition, this growth is increasing year on year. For example, the resources allocated in health care in Australia have been rising annually since 2001 (AIHW, 2013). Furthermore, with medical advancements, health expenditure is undeniably set to increase (Baker et al., 2003; Bodenheimer, 2005). In January 1993, Australia became the first country to establish a mandatory requirement for the economic evaluation of pharmaceuticals to be considered before a medicine could become eligible for government subsidization (Drummond et al., 2015). Since then, many jurisdictions with government-subsidized health care have enacted similar policies to require economic analysis for government subsidization programs, including the UK, the European Union, Canada, and New Zealand (Drummond et al., 2015).

where Cp = the present cost Cfn = the future cost at the nth year r = the rate of discount The benefits, or improvements, in outcome used in CEA are usually associated with the outcomes measured in studies of the intervention (Gray et al., 2011). For example, if an intervention is shown to decrease mortality and/or increase survival rates, the CEA could be measured as cost per life year gained. If an intervention can improve the indicator in the population, the benefits can be measured by the change in this indicator. The CEA can only be based on one outcome (though a composite outcome can be used), so careful consideration of the choice of outcome is important. The outcome used is also dependent on the perspective of the economic evaluation and this is important for behavior change interventions (Mpofu, 2014). The effects of these interventions may be beneficial from a number of perspectives (or may be beneficial from only one perspective). For example, a quit-smoking program may be beneficial from a number of perspectives: the

person who quits smoking may have improved health; from a payer perspective, improvement in health may lead to a reduction in health resource use; from a societal perspective, benefits from productivity (e.g., reduction in sick days) and improved health in others from reduced effects of passive smoking may result. The benefits, however, are hard to measure and, usually, considered as unobserved potential consequences (Alayli-Goebbels et al., 2013). Furthermore, as psychological factors affect individuals’ decision-making, evaluating the “true” economic effect of behavior change interventions is difficult to measure; thus, the outcome or benefit selection is extremely important. For example, people might not fully adhere to guidelines during or after a quit-smoking intervention, perhaps because it is difficult to see immediate benefits as opposed to the more long-term benefits (e.g., reduced cancer or cardiovascular disease risk) (Kelly & Barker, 2016). Even though empirical evidence supports a strong correlation between smoking and lung cancer, people

Cost-Effectiveness Evaluations of Behavior Change Interventions

continue to smoke. Thus, owing to the loss-aversion and status quo bias in behavior economic theory (Thaler & Sunstein, 2009; see also Chapter 42, this volume), the positive effects of behavior change interventions may be underestimated if the benefits of these interventions will only be realized at a future point in time. The message framing strategy and choice architecture can also lead to changes in individuals’ preferred choices without the imposition of strict regulation (see Chapter 42, this volume).

26.3.3 Collection of Data on Costs and Benefits Data collection of costs and benefits should also be considered in the economic evaluation process. The source of data on which the economic evaluation depends is a primary issue and is highly dependent on the type of study design used (e.g., experimental, quasi-experimental, and nonexperimental). For a further discussion on study designs, see Chapter 22, this volume. Experimental designs in the form of randomized control trials (RCTs) are regarded as the “gold standard” for evaluating intervention effectiveness (Gray et al., 2011). RCTs provide the most effective data (Gray et al., 2011) with minimal confounds (Fox-Rushby & Cairns, 2005) and their results focus on resolving the primary questions with which the researchers are concerned in a systematic way, as well as reducing uncertainty (Morris, Devlin, & Parkin, 2007). For these reasons, RCTs are considered the most appropriate study design to provide data on economic evaluation. The strengths of data gained from studies using RCT designs, however, can also create limitations, especially when considering implementation of the intervention in real-world contexts (see Chapter 23, this volume). For example, financial restraints (e.g., RCTs are unlikely to detect benefits across a life course or across generations) and issues of generalizability (i.e., lack of external validity) are persistent problems

377

(Gray et al., 2011). Furthermore, although the clinical evidence may be precise, evidence from one clinical trial may not be sufficient to inform economic evaluation and, thus, inform decisionmaking on intervention rollout. Therefore, seeking other information and data for CEA from additional sources is often needed. Another key source of data on costs and benefits that can inform an economic analysis is decision-analytic models (Gray et al., 2011). Decision models enable the developer to visualize the sequences of events in a practical pathway following an intervention or action in a logical framework and allow costs and outcomes associated with that action to be applied in the analysis (Gray et al., 2011). This approach can incorporate a wider range of evidence and be able to capture the long-term effects of interventions (Drummond et al., 2015). According to the study needs, different types of models have been developed, such as decision tree, Markov model, and discrete choice simulation. Most commonly used are decision tree and Markov modeling (Gray et al., 2011). Researchers recommend applying the simplest model that can achieve the goals of the evaluation, although more complex models exist and may be necessary (Brennan, Chick, & Davies, 2006). Decisionanalytic models can utilize data from multiple sources and these can include trials, observational studies, experts’ opinions, official guidelines, or other resources. A key disadvantage of modeling is potentially unreliable data sources, and the associated uncertainty in any point estimate can lead to limitations in the overall evaluation (Gray et al., 2011). Dealing with uncertainty issues is discussed in Section 26.3.4. In summary, it should be acknowledged that the two sources of data that inform analyses of costs and benefits in economic evaluation (trialbased data and decision-analytic models) are not necessarily mutually exclusive. The economic evaluation based on clinical trials is mainly concerned with the impacts of different

TIANJIAO WANG, MARTIN DOWNES, JOSHUA BYRNES, AND PAUL SCUFFHAM

IV

λ

Incremental costs

378

I

•B Incremental benefits A

III

II

Figure 26.1 Cost-effectiveness plane

treatments associated with relative costs and benefits (Drummond et al., 2015). The modeled approach focuses on informing specific decisions based on a real-world scenario (Drummond et al., 2015). Therefore, using a combination of clinical trial and model approaches is preferable in CEA.

26.3.4 Drawing Conclusions from a CEA It is important to consider the conclusions that are drawn from a CEA. For example, the result of a CEA that compares interventions often investigates the difference in cost of the “active” intervention to its comparator and how that relates to the difference in the outcome of interest (e.g., investigating the difference in cost between the status quo and implementing a nationwide environmental behavioral change intervention and how this relates to the difference in CO2 emissions). This is the definition of an ICER and calculation formula is displayed as below: Þ ICER ¼ ððC1C0 E1E0Þ

where C1 and E1 = the cost and effect in the intervention group C0 and E0 = the cost and effect in the control care group. The cost-effectiveness plane, as a tool to draw conclusions from a CEA, comprises four quadrants (see Figure 26.1) (Drummond et al., 2015), and the point at which the ICER lies determines whether an intervention is cost-effective or not. Point A in the graph usually represents the control group of the study, such as “usual practice” or “doing nothing” (Gray et al., 2011). Compared to a standard intervention (Point A), the active intervention is considered more effective and more expensive if located in quadrant I (northeast), but more effective and less costly if located in quadrant II (southeast). On the other hand, the active intervention is considered less effective and less costly if located in quadrant III (southwest), but less effective and more expensive if located in quadrant IV (northwest) (Drummond et al., 2015). Therefore, interventions located in quadrant II are regarded as dominant strategies and should be implemented and those located in quadrant IV are regarded as dominated strategies

Cost-Effectiveness Evaluations of Behavior Change Interventions

and should not be implemented. Interventions located in quadrant I and quadrant III are considered uncertain and, as such, a threshold is required to draw conclusions. As a result, a maximum acceptable threshold (generally referred to using the Greek letter λ) is often applied (Gray et al., 2011); this is the maximum amount of money a decision maker is prepared to spend for the outcome (e.g., $50,000 for a year of life at full health). For example, if the result of the CEA for a new intervention is located at point B in the graph, then the intervention would be considered cost-effective since, given the λ, it will cost less than the acceptable threshold while obtaining the same amount of benefit. As there is often uncertainty over the data inputs in the analysis, as outlined in Section 26.3.3, with regards to quality and rigor, it is often necessary to test the uncertainty in the result of the CEA (Gray et al., 2011). This is achieved by performing appropriate sensitivity analysis to evaluate the reliability in a range of scenarios (Mpofu, 2014). There are several approaches that could be considered when undertaking a sensitivity analysis. Deterministic sensitivity analysis can be used to investigate the sensitivity of the results from a CEA to variations in specific parameters (or multiple parameters). One (univariate) or more (multivariate) parameters are changed in the CEA using uncertainty data, usually from the source of the parameter (e.g., 95% confidence interval) to develop a range of ICERs to help assess the impact of change on the overall result for that specific parameter. The results of deterministic sensitivity analysis provide some idea as to the uncertainty within the model and are useful when there is a lot of uncertainty for one specific parameter. However, it is often suggested that deterministic sensitivity analysis (especially univariate) need to be used with caution, as its ability to identify the true uncertainty is limited. The assumption of the analysis is that all the parameters are independent, which does not take into account the correlation between parameters. As a result, the uncertainty

379

may be underestimated (Gray et al., 2011). Owing to its straightforward and simple nature, however, the approach is still frequently used in the literature (Briggs, Sculpher, & Claxton, 2006). Oneway sensitivity analysis can be illustrated is by “tornado” graph, which present the impact of parameters on ICERs. Probabilistic sensitivity analysis is another approach to sensitivity analysis. It is often regarded as the standard approach to presenting uncertainty jointly in economic evaluation modeling across all parameters simultaneously (Gray et al., 2011). Each parameter will be assigned a specific statistical distribution depending on its characteristics and the data used for sourcing the parameter. In the end, results of this sensitivity analysis will usually be presented visually in scatterplots along the cost-effectiveness plane and cost-effectiveness acceptability curve. As in the previous graph, the cost-effectiveness plane represents the probabilities of being cost-effective based on λ. The costeffectiveness acceptability curve indicates the probabilities of being cost-effective with the variation of λ. This probabilistic analysis also emphasizes that the purpose of economic evaluation is to provide evidence to support or inform, rather than make, decisions. This section has provided an overview of the theory and methodology of economic evaluation, including outlining the assumptions of neoclassical economic theory, discussing the forms of economic evaluation, describing and discussing issues in measurement and data collection of costs and benefits, and drawing conclusions from CEA. Given the existing uncertainties, the assessment of the quality of CEA is essential to avoid making inappropriate decisions. Next, the chapter outlines the appraisal of CEA.

26.4 Appraisal of CEA for Behavior Change Interventions Since economic evaluation studies require the reporting of different types of data, including

380

TIANJIAO WANG, MARTIN DOWNES, JOSHUA BYRNES, AND PAUL SCUFFHAM

costs, benefits, and cost-effectiveness results, considerable challenges are faced when scrutinizing the findings of an evaluation (Husereau et al., 2013). In addition, as the conclusions of economic evaluation studies are often used to provide evidence for decision-making around intervention implementation, the quality of these studies is extremely important in order to avoid making adverse and costly decisions. Appraisal tools have, therefore, been developed to optimize reporting of economic evaluation and improve the quality of these studies, such as the CHEERS statement (Husereau et al., 2013), and Drummond’s (2015) checklist. These two checklists have been used widely as guidelines for researchers when conducting economic evaluation studies, as well as for the appraisal of economic evaluation quality (Husereau et al., 2013). The function of these tools is to guide the reporting of results of economic evaluations and help researchers, reviewers, and other stakeholders check whether evaluation items included are reasonably described in studies. Researchers need to make decisions on which method is appropriate depending on the research question and desired outcome. For example, for specific medical products, it is appropriate to regard the patients directly receiving the treatment as the target population (Item 4 in the CHEERS checklist). However, when a study investigates the benefits of reducing cigarettes through a behavior change intervention (e.g., a quit-smoking program), measuring the benefits to participants in the program alone might not be sufficient because this does not capture the benefits to people not directly targeted, such as preventing the participant’s family from passive smoking (Alayli-Goebbels et al., 2013). The setting and location (Item 5 in the CHEERS checklist) also need to be considered, as behavior change interventions can be implemented at three levels, consisting of the individual, community, and national level (Cutler, 2004; see also Chapters 17 and 18, this volume). Even though medical interventions can have broader effects, behavior change interventions are most likely to

have broader social impact. In addition, although RCTs are often considered the “gold standard” for conducting economic evaluation (Gray et al., 2011), sometimes such designs are not feasible or ethical for behavior change interventions (see Chapter 21, this volume). Natural experiments, for example, might be more appropriate in some cases (AlayliGoebbels et al., 2013). In addition, it is also important to consider the time horizon of economic evaluation studies. The time frame can be short-term or long-term, usually depending on the nature of the disease in health care. However, long-term impact is often of utmost salience when evaluating behavior change interventions (Alayli-Goebbels et al., 2013). This section has discussed the appraisal of CEA for behavior change interventions. It selects some of the items from checklists used in the appraisal of these interventions (Drummond, 2015; Husereau et al., 2013), which need to be considered carefully when applying CEA in behavior change interventions. Some implications and recommendations for further research are discussed in Section 26.5.

26.5 Implications for Further Research 26.5.1 Implications for Measuring Benefits CEA, as a form of economic evaluation, provides an approach to inform decision-making about setting priorities for the funding of behavior change interventions. However, the theoretical foundation of economic evaluation, that is, neoclassical economic theory, might be criticized when applied to behavior change interventions. Some may argue that the aim of behavior change interventions is not to maximize individuals’ utility (see Chapter 42, this volume) but to benefit the individual and the surrounding community. For example, a quitsmoking program may not bring about pleasure for the individual who evaluates greater pleasure

Cost-Effectiveness Evaluations of Behavior Change Interventions

from smoking over their risk in loss of health but may benefit other members of the household. Another example is placing restrictions on fishing as an intervention for sustainable development, but fishing enthusiasts and their families may need to sacrifice some of the pleasure they derive from fishing. Consequently, the societal perspective may need to be taken into consideration when conducting and interpreting CEA for behavior change interventions. This perspective aims to capture all possible costs and benefits for society as a whole rather than an exclusive focus on the costs and benefits to a particular group. The benefits of behavior change interventions may not be solely on improvement of one focused outcome but may also improve welfare more broadly and may be beneficial to the current and future generations, such as the improvement of sustainable development and adaptations to climate change. Moreover, the implementation of behavior change interventions may completely change human behavior in the long run, which might lead to the reduction of the budget allocation in other fields, such as reductions in health care budgets due to the change in individuals’ physical activity participation. Therefore, measuring the benefits of behavior change interventions from a societal perspective is complex and challenging for CEA. Another possible critique related to neoclassical economic theory is that, if the intervention can successfully change individuals’ behaviors, their preferences may be changed in the long run. However, in the neoclassical economic world, people’s preferences are assumed to be stable. They assume people’s decisions are always rational and that they always know the exact ranking of their preferences against various available options. However, if individuals’ preferences can be changed from being stable to more dynamic over time, the assumptions underpinning economic evaluation may need to be relaxed or reviewed. CEA is a dominant framework applied in national guidelines for economic evaluations of treatments and interventions (Drummond et al.,

381

2015; NICE, 2013) because it can provide integral evidence for the government to make decisions on resource allocation. The monetary values of cost units are consistent across different forms of economic evaluation; however, this is not the case when measuring benefits. If the units of benefits can be consistent as well, it would be straightforward for policy makers to set priorities across multiple fields, such as health care and education. For example, in cost utility analysis, researchers try to unify the benefits from different interventions of different diseases into one type of unit (i.e., the QALY). In cost-benefit analysis, both costs and benefits are measured by monetary values. However, assigning monetary values for less tangible benefits (e.g., psychological wellbeing) represents a challenge. Thus, the appropriate benefit units for the economic evaluation behavior change interventions need to be addressed in future studies. Overall, the issues related to the selection of perspectives, dynamic preferences, and benefit units of behavior change interventions should be comprehensively considered for CEA, since this would influence the decision on the funding of behavior change interventions.

26.5.2 Implications for Modeling Appropriate statistical models are required for behavior change interventions. These models are the core of economic evaluation, which can explore the long-term effects of an intervention on the participants or on overall society. A model usually describes the progress of a disease or problem with or without the intervention in place. Different states of a disease or status can be defined according to some outcome indicators. For example, the status of Alzheimer’s disease can be defined as mild, moderate, or severe according to scores on an evaluation tool (e.g., the Mini Mental State Examination; Perneczky et al., 2006). However, there is little consensus on the definitions of behavior change states and little

382

TIANJIAO WANG, MARTIN DOWNES, JOSHUA BYRNES, AND PAUL SCUFFHAM

consensus on the measures and tools used across studies. Therefore, statistical models used for the economic evaluation of behavior change interventions need to take the situation for evaluation into account and design the model accordingly.

26.5.3 Political Viewpoint Many nations are governed by a democratic political system wherein the public votes for representatives to form a government based on their preferences, with everyone of voting age having “equal say.” One of the features of the system is that elected officials have a mandate to implement beneficial policies that are popular with the electorate; otherwise, their chances of maintaining office may be subsequently harmed at the ballot box. In most democratic countries, the typical time between mandatory elections is seldom longer than five years, leaving a relatively short “window” for governments to implement policies. These time periods are also curtailed because new governments need some time to deal with the handover when first elected and also need some time to prepare as the next round of elections approaches. As a result of this relatively short policy implementation window, governments are more likely to prioritize interventions or programs that are likely to have relatively short-term benefits, which they can then present to the electorate in advance of election periods. As a consequence, such policies are unlikely to be designed with long-term benefits in mind and may even have detrimental long-term consequences. For example, a government may stimulate the national economy and create jobs by giving tax breaks to multinationals, but those companies may not have a strong record on environmental protection or sustainability. The short-termism brought about by political systems presents considerable challenges to commissioning funding for behavior change interventions whose outcomes will only be realized in the long term, even in the face of advocacy arguments that have strong evidence to

support their societal benefits and cost-effectiveness. What is needed are strategic bodies with long-term funding that operate independently of government, or are enshrined in law, that make evidence-based policy decisions that will bring about change resulting in long-term effectiveness; the UK National Institute for Health and Care Excellence is one such strategic body.

26.6 Conclusion This chapter has provided a brief overview of the economic theory, methods, and practice of CEA for behavior change interventions. CEA, as a form of economic evaluation, is a compelling method to inform decisions on setting priorities for funding interventions. Alongside the efficacy, effectiveness, process, and acceptability components, economic evaluation essential is an aspect of the evaluation of behavior change interventions and is often the centerpiece of evidence presentations to governments, policy makers, and other stakeholders to persuade them to invest in behavior change interventions.

References AIHW (Australian Institute of Health and Welfare). (2013). Health Expenditure Australia 2011–12. www.aihw.gov.au/reports/health-welfare-expendi ture/hea-2011-12/report-editions Alayli-Goebbels, A. F., Evers, S. M., Alexeeva, D. et al. (2013). A review of economic evaluations of behavior change interventions: Setting an agenda for research methods and practice. Journal of Public Health, 36, 336–344. https://doi.org/ 10.1093/pubmed/fdt080 Baker, L., Birnbaum, H., Geppert, J., Mishol, D., & Moyneur, E. (2003). The relationship between technology availability and health care spending: Attempts to address technology availability and rising costs could end up badly misguided if implications for quality are not considered. Health Affairs, 22, W3–537. https://doi.org/10.1377/ hlthaff.w3.537

Cost-Effectiveness Evaluations of Behavior Change Interventions

Battista, W., Romero-Canyas, R., Smith, S. L. et al. (2018). Behavior change interventions to reduce illegal fishing. Frontiers in Marine Science, 5, 403. https://doi.org/10.3389/fmars.2018.00403 Bhattacharya, J., Hyde, T., & Tu, P. (2013). Health Economics. New York: Palgrave Macmillan. Binger, B. R., & Hoffman, E. (1988). Microeconomics with Calculus. Glenview, IL: Scott Foresman. Bodenheimer, T. (2005). High and rising health care costs. Part 2: Technologic innovation. Annals of Internal Medicine, 142, 932–937. https://doi.org/ 10.7326/0003-4819-142-11-200506070-00012 Brennan, A., Chick, S. E., & Davies, R. (2006). A taxonomy of model structures for economic evaluation of health technologies. Health Economics, 15, 1295–1310. https://doi.org/ 10.1002/hec.1148 Briggs, A., Sculpher, M., & Claxton, K. (2006). Decision Modelling for Health Economic Evaluation. New York: Oxford University Press. Cutler, D. M. (2004). Behavioral health interventions: What works and why. In N. B. Anderson, R. A. Bulatao, & B. Cohen (Eds.), Critical Perspectives on Racial and Ethnic Differences in Health in Late Life (pp. 643–674). Washington, DC: National Academies Press. https://doi.org/10.17226/11086 Drummond, M. F., Sculpher, M. J., Claxton, K., Stoddart, G. L., & Torrance, G. W. (2015). Methods for the Economic Evaluation of Health Care Programmes (4th ed.). New York: Oxford University Press. Fertig, A., Lefkowitz, J., & Fishbane, A. (2015). Using Behavioral Science to Increase Retirement Savings: A New Look at Voluntary Pension Contributions in Mexico. Ideas42 report. www .ideas42.org/wp-content/uploads/2015/11/ I42_571_MexicoPensionsReport_ENG_final_dig ital.pdf Folland, S., Goodman, A. C., & Stano, M. (2016). The Economics of Health and Health Care (7th ed.). Abingdon: Routledge. Fox-Rushby, J., & Cairns, J. (2005). Economic Evaluation. Maidenhead: Open University Press. Fragoulakis, V., Mitropoulou, C., Williams, M., & Patrinos, G. P. (2015). Economic Evaluation in Genomic Medicine. Burlington, CA: Academic Press.

383

Gray, A. M., Clarke, P. M., Wolstenholme, J. L., & Wordsworth, S. (2011). Applied Methods of CostEffectiveness Analysis in Healthcare. New York: Oxford University Press. Hallsworth, M., Snijders, V., Burd, H. et al. (2016). Applying Behavioral Insights: Simple Ways to Improve Health Outcomes. Report of the WISH Behavioral Insights Forum 2016. www.imperial .ac.uk/media/imperial-college/institute-of-globalhealth-innovation/Behavioral_Insights_Report(1).pdf Huang, X., Lin, J., & Demner-Fushman, D. (2006). Evaluation of PICO as a knowledge representation for clinical questions. AMIA Annual Symposium Proceedings, 2006, 359–363. https://www.ncbi .nlm.nih.gov/pmc/articles/PMC1839740/ Husereau, D., Drummond, M., Petrou, S. et al. (2013). Consolidated health economic evaluation reporting standards (CHEERS) statement. Cost Effectiveness and Resource Allocation, 11, 6. https://doi.org/10.1186/14787547-11-6 Kelly, M. P., & Barker, M. (2016). Why is changing health-related behaviour so difficult? Public Health, 136, 109–116. https://doi.org/10.1016/j .puhe.2016.03.030 Morris, S., Devlin, N., & Parkin, D. (2007). Economic Analysis in Health Care. Chichester: Wiley. Mpofu, E. (2014). Community-Oriented Health Services: Practices Across Disciplines. New York: Springer. Neri, D., Leifer, J., & Barrows, A. (2016). Graduating Students into Voters. Overcoming the Psychological Barriers Faced by Student Voters: A Behavioral Science Approach. Ideas42 report. www.ideas42.org/wp-content/uploads/2017/05/ Students_into_Voters.pdf NICE (National Institute for Health and Care Excellence). (2013). Guide to the Methods of Technology Appraisal 2013. www.nice.org.uk/ process/pmg9/chapter/evidence OECD (Organisation for Economic Co-operation and Development). (2017). Health spending (indicator). https://data.oecd.org/healthres/healthspending.htm Perloff, J. (2015). Microeconomics (7th ed.). Boston: Pearson.

384

TIANJIAO WANG, MARTIN DOWNES, JOSHUA BYRNES, AND PAUL SCUFFHAM

Perneczky, R., Wagenpfeil, S., Komossa, K., Grimmer, T., Diehl, J., & Kurz, A. (2006). Mapping scores onto stages: Mini-mental state examination and clinical dementia rating. The American Journal of Geriatric Psychiatry, 14, 139–144. https://doi.org/ 10.1097/01.jgp.0000192478.82189.a8 Schardt, C., Adams, M. B., Owens, T., Keitz, S., & Fontelo, P. (2007). Utilization of the PICO framework to improve searching PubMed for clinical questions. BMC Medical Informatics and

Decision Making, 7, 16. https://doi.org/10.1186/ 1472-6947-7-16 Thaler, R. H., & Sunstein, C. R. (2009). Nudge: Improving Decisions about Health, Wealth, and Happiness. New Haven, CT: Yale University Press. WHO (World Health Organization). (2016). Health System Efficiency: How to Make Measurement Matter for Policy and Management. Geneva: WHO.

27 Addressing Underserved Populations and Disparities in Behavior Change Benjamin Schüz and Monica Webb Hooper

Practical Summary Social disparities in health-related behaviors and other behaviors critical to well-being are some of the main reasons for social disparities in health and other outcomes. Developing interventions that effectively deliver changes in behaviors in underserved and “minority” populations is challenging, as current research and theoretical frameworks have only recently begun to explore the factors underlying disparities in behaviors. This chapter summarizes the state of the science on behavioral disparities, discusses how current models of behavior can help to understand and modify disparities in health and other behaviors, and highlights how barriers to behavior change are determined by social and cultural factors. Better reporting and analysis of behavioral disparities is recommended to allow more targeted and effective intervention development.

27.1 Introduction Changing behavior in underserved and racial and ethnic minority populations presents considerable challenges. Consider the following research example. In 2018, a community-engaged research study was conducted to understand the needs of underserved communities in a Midwest US city (Webb Hooper et al., 2019). Residents of economically disadvantaged sectors of the city shared healthrelated lived experiences that affected their use of tobacco products, dietary habits, physical activity engagement, health care seeking, and the use of preventive health services (e.g., annual check-ups and cancer screenings). Participants self-identified as African American (80 percent), non–Hispanic White (17 percent), or “other” (3 percent). They were mostly older (between sixty and eighty years of age), female (68 percent), and had health insurance (97 percent); 10 percent were deaf/hearingimpaired, and 71 percent were diagnosed with a

chronic health condition. The shared experiences revealed the depth of multilevel determinants of behavior (organizational, community, interpersonal, and intrapersonal) and disparities by socioeconomic status (SES), race, and disability. Participants identified many instances where they found behavior change particularly difficult to accomplish, and these were attributed to a range of difficulties. Changing tobacco consumption provides an excellent example: Among participants in low SES neighborhoods, both the awareness of and access to culturally appropriate, evidence-based tobacco cessation programs were low. Consistent with the literature (e.g., Webb Hooper, Payne, & Parkinson, 2017), medical provider counseling and prescriptions for Federal Drug Administration (FDA)– approved medications were lower among African American and deaf/hearing-impaired participants compared to non–Hispanic Whites. Further, https://doi.org/10.1017/9781108677318.027

386

¨ Z AND MONICA WEBB HOOPER BENJAMIN SCHU

African American participants described working conditions that resulted in frequent exposures to secondhand smoke, which was not a notable theme observed among non–Hispanic Whites. Themes at the community level revealed dramatic differences in lived experiences, which impacted health behavior. The physical environment in low SES neighborhoods included significant exposure to smoking cues (e.g., cigarette butts on the ground, seeing others smoking) and tobacco marketing (e.g., at corner stores and on signs posted throughout the area) as well as the widespread ability to purchase single cigarettes (“loosies”). Interpersonal factors, including themes of smoking as a socially normative behavior, peer influences, and communication challenges, emerged as barriers to cessation among African Americans and deaf/ hearing-impaired participants, even when readiness to quit was high. Compared to higher SES and non– Hispanic White participants, the expressed attitudes, concerns, and needs of the underserved participants revealed individual-level barriers to cessation, such as mental health concerns, maladaptive coping strategies, the belief that tobacco product modification reduces cancer risk, and medical distrust. Taken together, findings from this example study illustrate some of the key challenges for behavior change in underserved populations. This chapter discusses and illustrates the structural inequalities that contribute to socioeconomic differences in behavior and behavior change. Even though the approach to predicting, understanding, and changing behavior in behavioral science is mainly focused on exploring and explaining determinants on an individual level, or at least as filtered through an individual perspective, there are structural and societal factors affecting behavior that are beyond the control of the individual (see also Chapters 16, 17, and 28, this volume). In many respects, understanding and changing behaviors such as tobacco use are not an individual problem but represent structural issues and inequities created and maintained by society, history, and policy.

While the study of behavioral contributions to health disparities is a relatively new area of study, there has been a surge over the past two decades in studies designed to understand the complexity of factors that give rise to and facilitate inequities and to appropriately design behavior change interventions to address them.

27.2 Disparities and Inequities: Definitions and Dimensions There remains considerable debate on the definitions of the terms used for key constructs and variables identified in this chapter. However, it is important that consistent language is used among scientists, practitioners, and policy makers when describing differences in behaviors and outcomes across populations. Determinants of behavior are closely interlinked and include social psychological and biological (e.g., age, gender) factors, physical environment, and access to facilitating environments. The next sections outline some of the key concepts and structural variables important to understanding behavior change in underserved populations and economically disadvantaged populations.

27.2.1 Inequities, Disparities, and Equity Inequities and disparities are concepts that, in their most abstract form, describe the unequal distribution of goods and resources within a society. Most research on and applications of these concepts concern health-related topics. The focus of the chapter, therefore, will be on work in this domain, noting that these concepts can be applied to describing the unequal distribution of other relevant outcomes such as educational attainment (e.g., Mickelson, 2003). Health inequities, which are synonymous with health disparities, refer to a particular category of health differences that are rooted in systematic social disadvantage and can be seen as a prototypical

Addressing Underserved Populations and Disparities

example of how individual-level outcomes are affected by social inequality. Health inequities are caused through a complex set of interrelated factors at the environmental, societal, socioeconomic, individual, and behavioral levels. The causes include, but are not limited to, differential exposure to environmental health agents, differential access to health services, work conditions, and psychosocial implications of poverty (Marmot, 2015). Health disparities adversely affect multiple populations, characterized by SES, race and ethnicity, sexual orientation, gender, disability status, and geographic location (U.S. Department of Health and Human Services, 2010). Health equity refers to the aspirational goal of optimal health for the entire population (U.S. Department of Health and Human Services, 2008). The implications of these inequities and disparities for the design of behavior change interventions will be discussed in section 27.6.

27.2.2 Dimensions of Disparities: Intersectionality There are multiple dimensions along which individuals and groups can be stratified socially within a society. Behaviors and outcomes might be distributed similarly or differentially along these dimensions, which is one key challenge in research on inequities and differences in behaviors. As illustrated by the multilevel influences on health and health behaviors in the example in the introduction (see Section 27.1; Webb Hooper et al., 2019), inequities can be observed and examined at the level of an individual, a social group, an environmental (geographical) area, or any combination thereof. This has implications for the measurement and the conceptual distinction of dimensions of inequity. One key publication on inequality and health (Shaw et al., 2007) outlines more than thirty-two indicators along which the patterning of health outcomes has been observed. Further, as illustrated in the introductory example, these dimensions of inequity

387

comprise potentially different mechanisms of action that translate social patterning into health outcomes. For example, lower individual educational attainment might be associated with lower health literacy, leading to a lower capacity to process and act on health-related information, thus affecting health negatively – while arealevel deprivation could imply less access to health services and thus potential delays in helpseeking. To address this heterogeneity in aspects and effects of inequity, a framework to report and analyze equity effects of interventions has been proposed – the PROGRESS-Plus framework (O’Neill et al., 2014; see Sidebar 27.1). To complicate matters further, there are considerable interactions between inequality dimensions in affecting health; for example, the health effects of low educational attainment and low income are rarely experienced separately from each other. This phenomenon has been described as intersectionality (Bauer, 2014), indicating that persons who experience discrimination or hardship based on one indicator of inequality are also very likely to face discrimination based on another.

27.3 Frameworks of Health Inequities Several models have been applied to or are relevant to the study of health behavior among underserved populations. In this section, three relevant frameworks are reviewed: the social ecological model, the biopsychosocial model, and the minority stress model.

27.3.1 Social Ecological Model The social ecological model of health promotion recognizes that individual behavior is a function of multilevel factors, including intrapersonal, interpersonal (e.g., family and friends), organizational (e.g., health care systems), community, and public policy (e.g., health insurance access), each

388

¨ Z AND MONICA WEBB HOOPER BENJAMIN SCHU

Sidebar 27.1 PROGRESS-Plus: a framework addressing heterogeneity in socioeconomic status measures

For the evaluation of socioeconomic differences in the effects of health-promotion interventions, the Cochrane Collaboration has proposed a framework (PROGRESSPlus; O’Neill et al., 2014) that comprises some of the most relevant dimensions of social stratification, according to the research group. The PROGRESS acronym stands for “Place of Residence,” “Race/Ethnicity/Culture/Language,” “Occupation,” “Gender/Sex,” “Religion,” “Education,” “Socioeconomic status,” and “Social capital,” with “plus” indicating additional factors such as subjective experiences regarding disadvantage and personal relationships. This framework has been instrumental in stimulating research on differential outcomes of health interventions according to these indicators, so-called equity effects of interventions. However, such differential outcomes are still not routinely reported in research. For example, an equity-focused systematic review on physical activity interventions (Attwood, van Sluijs, & Sutton, 2016) found that, out of 171 intervention studies that assessed PROGRESS-Plus factors, only 25 actually reported analyses stratified by at least one of these dimensions. Follow-up reviews of physical activity interventions, such as those in older adults (Lehne & Bolte, 2017), yielded even lower proportions of effect sizes stratified by PROGRESS-Plus dimensions. While PROGRESS-Plus has at least seen some application in research on health behavior change interventions, no such systematic work on differential effects of behavior change interventions in general exists so far. This suggests that, in order to make evidence-based decisions on equity effects of behavioral interventions, analyses of research on interventions stratified by these dimensions need to be routinely reported.

of which contribute to unhealthy behaviors (McLeroy et al., 1988; see also Chapter 17, this volume). In underserved groups, in particular, the influence of upstream (e.g., health care system) structural factors may negatively affect individual-level potential for adaptive health behavior change. For example, racial differences in human papilloma virus (HPV) vaccine uptake can be best understood within the context of the social ecological framework applied to health behavior in underserved populations. Hispanic youth (Burdette et al., 2017) and African American women aged nineteen to twenty-six are less likely to initiate the vaccination series compared to

non–Hispanic Whites, even after adjusting for demographics and health care access (Williams et al., 2017). The reasons for this disparity in uptake are unclear but it is plausible that intrapersonal (e.g., risk perceptions), interpersonal (e.g., parental beliefs), and health care (e.g., lack of provider recommendation; Burdette et al., 2017) factors play a role.

27.3.2 Biopsychosocial Model The biopsychosocial model extends the traditional biomedical model with a more holistic conceptualization of factors that contribute to

Addressing Underserved Populations and Disparities

health, wellness, and disease risk (Engel, 1977). The model highlights the unique and overlapping roles of psychological (e.g., cognitions and emotions), biological (e.g., heredity), and social factors (e.g., income and education). Applied to the study of health behaviors among underserved populations, Myers’s (2009) lifespan biopsychosocial model builds on the original model and proposes that health behavior is predicted indirectly by race/ethnicity and SES and mediated by psychosocial adversities, psychosocial reserve capacity, cognitive processing and emotional regulation, biological vulnerabilities, and biological stress processes. For example, a person with lower educational attainment might encounter specific psychosocial adversities in everyday life (e.g., daily hassles and job problems) and possess lower psychosocial reserve capacity such as access to supportive social networks, which in turn results in less adaptive attempts at coping with stress, such as risky alcohol consumption. In comparison to the social ecological model, the lifespan biopsychosocial model specifies direct and mediational effects and offers greater specificity of hypothesized individual-level mechanisms underlying behavior.

27.3.3 Minority Stress Model The minority stress model (Meyer, 2003) is a framework for understanding and prioritizing strategies for addressing behavior change in sexual and gender minority populations. The foundation of the model is that minority populations face unique socially based, chronic stressors that are not commonly experienced by nonstigmatized groups. It asserts that, for example, sexual minorities face mistreatment, discrimination, and victimization across the lifespan, which in turn affect access to quality medical care, career opportunities, and health behaviors (Marshal et al., 2008). The internalization of these stressful experiences may increase the likelihood of mental health problems (Williams, Neighbors, &

389

Jackson, 2003) and engagement in maladaptive behaviors such as tobacco smoking (Wheldon et al., 2018), problematic alcohol use (Wilson et al., 2016), polysubstance use (Kashubeck-West & Szymanski, 2008), or sexual risk behavior (Bimbi et al., 2006). Similar to the social ecological model, the minority stress model recognizes the role of upstream stressors (e.g., societal marginalization) on group and individual behavior and health.

27.4 Inequities in Behavior and Behavior Outcomes Even though inequity is principally generated at the structural and societal level, one mechanism by which it translates into, for example, health outcomes is through individual behavior (Petrovic et al., 2018; Sidebar 27.2). This means that the structurally unequal distribution of resources, barriers, and discrimination causes similarly unequal distributions of health-related behaviors. Health behaviors, in turn, are proximal mechanisms by which unequal health outcomes are cultivated and distributed, as engaging or refraining from health-related behaviors such as smoking or adhering to a medication schedule interact directly with the physiological processes in that manifest in health outcomes (for a generic causal model, see, e.g., Hardeman et al., 2005). The key point here is that a person’s SES is related to the kinds of behavior they engage in and how often. A number of high-profile studies based on nationally representative data show that health-related behaviors are segregated by SES both in high-income countries (e.g., Meader et al., 2016) and in low-income countries (Allen et al., 2017). In these studies, a pattern emerges by which the more economically disadvantaged individuals are, the higher the likelihood that they engage in riskier behaviors and have barriers to engaging in healthpromoting behaviors.

390

¨ Z AND MONICA WEBB HOOPER BENJAMIN SCHU

Sidebar 27.2 Health behaviors as causal links between inequity and health

A recent systematic review and meta-analysis (Petrovic et al., 2018) reveals that the relationship between structural and social inequity and health outcomes is mediated through health behaviors. In this review, 114 studies were assessed for the degree to which the four health behaviors of smoking, alcohol consumption, physical activity, and diet contributed to socioeconomic inequities in a range of health outcomes – all-cause mortality, cardiovascular disease, and metabolic disorders. Put simply, the authors tried to identify how much of the socioeconomic gradient in these outcomes was due to differences in health behaviors. The authors recalculated socioeconomic effects from the reviewed studies and found that all four health behaviors contributed substantially to explaining health inequities (see Figure 27.1). Smoking contributed most to health inequities, and the influences of the different health behaviors were analyzed separately by SES indicators. Overall, this systematic review and metaanalysis underscores the importance of health behaviors for explaining health inequities and suggests that behavior change interventions are a viable method to reduce health inequities – if these interventions work regardless of socioeconomic background.

Diet

Physical Activity

Alcohol

Smoking

–5

0

5

10

15

20

25

30

Percentage Contribution of Each Variable to All-Cause Mortality by Health Behavior Computed According to the Absolute Scale Difference Method Other SEP Indicators

Occupation

Education

Figure 27.1 Contribution of health behaviors to inequities in all-cause mortality by SES indicator in Petrovic et al. (2018)

35

Addressing Underserved Populations and Disparities

This general pattern of more economic disadvantage being associated with higher levels of risky behaviors is also evident for other maladaptive behaviors such as criminal behaviors (Odgers et al., 2008) or school absenteeism (Ready, 2010). However, the opposite seems to be true for prosocial behaviors: Individuals with lower SES are more likely to help someone in need. For example, Piff et al. (2010) showed in a series of experiments that individuals with lower SES were more generous, more charitable, more trusting, and more helpful than individuals with higher SES. What makes this research particularly interesting is the fact that all studies controlled for potential alternative explanations such as religiosity or ethnicity. What could underlie this particular pattern? Manstead (2018) argues that individuals with lower SES rely more on kin support, which would imply a more prosocial norm toward people requiring help. Manstead further argues that individuals from a lower socioeconomic background (“working class”) show more empathy based on a more collectivist normative orientation, which in turn makes it less likely that they profit from behavior change interventions that promote independence and personal control factors and, as such, rely on personal agency. At the same time, such a collectivist normative orientation would make maintaining common practices (such as unhealthy behaviors) or distrust in staterun health practices (such as mandatory medical examinations) more likely (e.g., Lindbladh et al., 1996). The issue of intervention-generated inequities in outcomes (Lorenc et al., 2013) is revisited in Section 27.7 after reviewing the importance of differential effects of mechanisms and processes in interventions.

27.4.1 Mechanisms Underlying Differences in Health Behaviors and Health Behavior Change As outlined in the models of inequity reviewed in Section 27.3, the roles, status, and expectations associated with being a member of a particular

391

social group determine access to the resources and opportunities or the barriers to health behaviors. These resources and barriers exist at the level of the individual (e.g., knowledge) and the environment someone lives in (e.g., walkability of neighborhood; see also Chapter 17, this volume). Disadvantage and discrimination likely increase stress, potentially leading to maladaptive behavioral coping strategies such as smoking (Jahnel et al., 2019). From a behavioral science perspective, the effects of inequity can be understood either through the mediating effects of proximal individual determinants (Hankonen et al., 2017) or through effects of inequity as effect modifiers of proximal determinants of behavior such as those outlined in the theory of planned behavior (Schüz, 2017; Schüz et al., 2017; see also Chapter 2, this volume). While socioeconomic influences on behavior are acknowledged in at least some of these theories, the shape and strength of this influence are poorly specified (Schüz, 2017). Typically, research based on health behavior theories either ignores SES or treats it as a “nuisance” factor (control variable) or indirectly addresses SES influences in variables such as “environmental constraints” (Fishbein et al., 2001). If indicators of SES feature in research at all, it is mostly in relation to the question of whether SES effects are mediated by the health cognitions specified in the theories and, more often than not, direct SES effects remain. Moderating effects of SES in the relationship between cognitions and behavior are rarely addressed. However, some studies find that relationships between determinants and health behaviors tend to be closer in people with higher SES (e.g., Schüz et al., 2019) and for those living in districts with higher financial resources (Schüz et al., 2012). Larger effects of these determinants on health behavior in higher SES individuals also imply that theories building on these determinants explain more variance in health behavior in individuals with higher as compared to lower SES – that is, the theories fit better for higher SES

392

¨ Z AND MONICA WEBB HOOPER BENJAMIN SCHU

individuals. This suggests that these theories might be less suited to explain behavior in individuals with lower SES. Moreover, interventions that use behavior change techniques (BCTs; Michie et al., 2013; see Chapters 19 and 20, this volume) based on these theories would be less effective if the determinants addressed in BCTs potentially affect behavior differentially depending on SES. Apart from differential efficacy and working mechanisms of interventions, there are also specific barriers to intervention effectiveness in underserved populations.

27.5 Barriers to Intervention Effectiveness for Behavior Change Among Underserved Populations Facilitating behavior change among underserved populations has been a challenge for clinicians and researchers. For instance, racial and ethnic minorities are less likely to utilize psychotherapy services (e.g., Doran et al., 2017) and are more likely to drop out of ongoing psychotherapy treatment (e.g., Spoont et al., 2015). Populations that fall under the umbrella of “underserved” have also been described as “hard to reach,” “difficult to treat,” and “distrustful.” Greater representation in basic behavioral science and intervention research, and ensuring that studies are adequately powered to detect potential differences by social factors, is important to ascertain the generalizability of research findings and to reduce, and ultimately eliminate, behavioral disparities. Accordingly, the National Institutes of Health requires the inclusion of women and racial and ethnic minorities in research (Freedman et al., 1995). However, despite this federal mandate, research suggests that progress has been limited (Geller et al., 2011; Kwiatkowski et al., 2013). A review examined racial and ethnic minority representation in US weight-loss intervention trials and found that enrollment was comparable to the population proportion for African Americans, Asians, and Native

Americans, although it was low among Hispanics (Haughton et al., 2018). These patterns suggest that these groups are not hard to reach per se; however, scientists and clinicians have not fully identified effective strategies to engage underrepresented subpopulations. The barriers to clinical trial enrollment in both the general population and among underserved groups exist on multiple levels (see the example in the introduction, Section 27.1; Webb Hooper et al., 2019). A review of cancer clinical trials and underrepresented groups identified awareness (provider and potential enrollees), opportunity (study design, strict eligibility criteria, and provider communication and biases), and acceptance (participant concern about harm, costs, transportation, and time) as barriers to enrollment (Ford et al., 2008). These organizational, provider, and participant-level barriers exist across populations; however, they disproportionately impact underserved groups. Webb Hooper et al. (2019) found that African Americans and Hispanics were more likely to be excluded from a behavioral tobacco cessation trial compared to non– Hispanic whites, with African Americans meeting more exclusion criteria than the other population groups. Community and user engagement is essential to increasing the enrollment of underserved populations in behavioral research (see also Chapter 25, this volume). The qualitative study described in the introduction, for example, found that underserved racial/ethnic minority and deaf/hearing-impaired communities were more willing to participate in behavioral, compared to medical, research. Additionally, participants emphasized the importance of being “served” and “not simply studied.” Community-engaged research (see also Chapters 18 and 28, this volume) has emerged as a strategy to improve population health, increase trust, and promote health equity (e.g., Erves et al., 2017). Community-engaged research builds partnerships with key community stakeholders, designs interventions that are responsive to the target population,

Addressing Underserved Populations and Disparities

and expands opportunities to participate (see also Chapter 24, this volume). Again, a multilevel framework can be applied to this approach, as it often includes organizational-academic-community partnerships and has the potential to increase reach and reduce acceptance barriers. Key aspects of community-engaged research include community involvement at the inception of the research and coownership of the work (see also Chapter 28, this volume). Stakeholders work alongside researchers and interventionists at each stage of the research process, including data interpretation and dissemination. Trusting relationships will develop as communities gather evidence that completion of the project and ultimate community-level benefit are not dependent on grant funding or one particular investigator. These components are essential for community buy-in and future implementation of evidence-based interventions into community settings. Another potential benefit of engaging community partners early in the scientific process is that these stakeholders have critical insider knowledge and perspectives on theoretical precursors to behavior change – such as cultural beliefs, practices, and competing life concerns that may affect the potential for meaningful and sustained change. The ultimate goal is to develop culturally and community competent recruitment and retention strategies (Otado et al., 2015).

27.6 Culturally Specific Behavior Change Interventions With the recognition that intervention components such as BCTs and intervention engagement might work differently in different social groups, groupspecific culture becomes an issue for designing effective behavior change interventions (Smith, Rodríguez, & Bernal, 2011). However, the definition of culture and the methods used to develop culturally specific protocols are often unclear and there is no standard approach (Kagawa Singer, 2012). A particularly important consideration in thinking about culturally specific behavior change

393

interventions is that the research questions, study design, inclusion/exclusion criteria, intervention characteristics, operationalization of outcomes, selection of measures, data analysis and interpretation, and directions for future research are grounded within the worldview of the researcher. The implicit assumption that evidence-based interventions can be generalized or adapted to fit different populations suggests that determinants of behavior change and underlying mechanisms are largely universal. While this may or may not be accurate, there are also distinguishing characteristics, such as differences in social, psychological, biological, and cultural factors that affect behavior and behavior change between and within defined groups. Cultural considerations theoretically underlie many approaches in behavioral science, yet are not often explicitly tested and represented in intervention development. For example, the common-sense model of self-regulation (Leventhal, Diefenbach, & Leventhal, 1992) explicitly considers social background and culture as sources of individual representations of health and illness. However, most research applications of this theory have examined individual differences in illness-related constructs without further specifying the underlying cultural influences on illness perceptions. Interventions based on this and other models of subjective illness perceptions still mainly target individual beliefs without necessarily examining the underlying cultural influences on the belief structure (for more discussion on the utility of applying the common-sense model to develop culturally specific interventions for behavior change, see Chapter 5, this volume). Models of culturally specific interventions include distinct, but overlapping and complementary, elements. A model established by Resnicow et al. (1999) specifies surface and deep structure in the development of culturally specific behavior change interventions. Surface structure refers to the ostensible, explicit indicators of cultural specificity, such as photographs of persons who appear to be members of a specific target population, language, food,

394

¨ Z AND MONICA WEBB HOOPER BENJAMIN SCHU

and race-matched clinicians. Deep structure, on the other hand, refers to the fit of an intervention within a cultural group and the embedded historical, cultural, social, psychological, and environmental factors that may be meaningful to influence health behavior in the target population. The assumption here is that surface structure increases acceptability and engagement, whereas deep structure content determines intervention efficacy. Expanding this idea, Kreuter et al. (2003) suggest five strategies for developing culturally specific interventions, including peripheral (packaging with overt references to a given group), evidential (presentation of evidence of a health-related issue for a group), linguistic (messages in the native language of a specific group), constituent-involving (indigenous staff members), and sociocultural (incorporation of cultural values and characteristics). The application of these models within the context of groupbased cognitive behavioral therapy for tobacco cessation that was specifically designed for African Americans has demonstrated efficacy (Webb Hooper et al., 2017; see Sidebar 27.3). Similarly, a meta-analysis of sixty-five experimental and quasiexperimental intervention trials supported the use of culturally specific treatments and found a positive association between the extent of cultural elements and clinical outcomes (Smith et al., 2011). This implies that interventions to change any individual behavior, not only health-related behavior, need to take into account culturally specific factors, as both the determinants of behaviors and the function and implications of behaviors could differ across cultures.

27.7 Implications and Conclusions The challenges in understanding how and why social, economic, and political structures impact individual behavior and behavior change go beyond the mere description and examination of individual factors. Focusing on health behavior change as an exemplar, this chapter has shown that there are structures responsible for social

inequities on many levels, starting from individual differences up to broader societal and policy factors. Furthermore, these factors have been shown to interact with more proximal, individual determinants of behavior and behavior change in complex patterns. Solely focusing on individual determinants of behavior change in individuals from underserved populations puts both the onus for behavior change and the blame for the likely lower success rate on the individual and ignores the multitude of external factors that impact on behavior and behavior change. Understanding and effectively targeting the multilevel influences on behavior and behavior change require acquiring a deep understanding of the broad category of “underserved” and the complexities of the communities whose behavioral choices may be limited by larger societal structures. Many underserved groups, such as sexual and gender minorities, often rely on support from their own communities in their efforts to cope, develop resilience, and progress in multiple domains (Meyer, 2003; Ouellette & DiPlacido, 2001). On the other hand, effective and sustainable behavior change relies on identifying and applying efficacious, effective, and evidencebased BCTs. However, the degree to which these techniques are universally effective across different socioeconomic and cultural groups is both unknown and under-researched. From both a theoretical and a practical perspective, more systematic research on how socioeconomic factors influence the effects of behavioral determinants and the effectiveness of interventions is needed. More research with, rather than merely about, underserved populations is needed, in particular randomized controlled trials of behavior change interventions that factor in cultural and socioeconomic factors. More interventions incorporating equity effects (i.e., differential effects according to socioeconomic subgroup; e.g., Lorenc et al., 2013), along with better reporting of equity influences and subgroup analyses, are needed

Addressing Underserved Populations and Disparities

395

Sidebar 27.3 Developing culturally specific behavior change interventions

Table 27.1 summarizes an example of the process of developing a culturally specific tobacco cessation intervention for African Americans in accord with the stage model of behavioral therapies research (Rounsaville, Carroll, & Onken, 2001). Table 27.1 Steps to developing an evidence-based tobacco cessation intervention for African American adults Step and Description

Evidence

1. Needs assessment.

Epidemiological data indicate racial disparities in tobacco cessation, such that African American smokers are less likely to quit compared to non–Hispanic whites (Trinidad et al., 2011). This population also suffers a disproportionate burden of tobacco-attributable diseases, including cancer and heart disease, despite smoking fewer cigarettes per day (Trinidad et al., 2011). In addition, few evidence-based interventions options are designed to address the needs of this group.

2. Formative research. a. Qualitative research in a sample of African a. Use qualitative methods to generate American tobacco smokers provided hypotheses and to learn about facilitators empirical data on subjective norms for and challenges that are unique (or smoking, knowledge of racial tobacco elevated) within populations and disparities, expectancies for culturally subgroups, preferred intervention specific interventions, key intervention formats, and translation potential. factors, and methods of recruitment (Webb et al., 2007). b. Test intervention components using experimental designs to understand key antecedents of behavior change in the target population. The constructs and intervention components tested can be driven by existing, adapted, or new theoretical models of behavior change that are appropriate for the population.

b. Webb, Baker, and de Ybarra (2010) investigated cultural specificity as a potential framing approach through which intervention messages influence smokingrelated risk perceptions and cognitive factors associated with tobacco smoking. They found that both message content and culturally specific framing play important roles in eliciting tobacco-related cognitive changes (e.g., greater personal risk perceptions, intentions to quit smoking, and motivation to quit smoking) among low-income African Americans.

3. Apply culturally appropriate theoretical model(s). Consider theoretical models of behavior change, particularly those that include cultural components. Examples

Hooper et al. (2013) applied the Resnicow, et al. (1999) and Kreuter et al. (2003) models to develop a tobacco cessation cognitive behavioral therapy (CBT)

Continued

396

¨ Z AND MONICA WEBB HOOPER BENJAMIN SCHU

Table 27.1 (cont.) Step and Description include the Resnicow et al. (1999) model of cultural sensitivity and the Kreuter et al. (2003) model of culturally tailored health communications. Well-established models can also be adapted to account for the role of cultural factors in behavior.

Evidence among African Americans. The intervention incorporated key model dimensions, including surface and deep structure (Resnicow et al., 1999), and constituent-involving strategies (Kreuter et al., 2003).

4. Develop the behavior change intervention. Hooper et al. (2013) developed a culturally specific group CBT for smoking cessation, Develop the content, framing, and format which was an adaptation of a standard CBT for the intervention. Consider the possible program with an emphasis on African intervention format (e.g., individual, American culture. The culturally specific group, family) and mode(s) of intervention framing and content were developed delivery (e.g., face-to-face, telephone, based on previous qualitative (Webb et al., internet-based, or mobile health). The 2007) and lab-based experimental research content should include evidence-based (Webb et al., 2010). Session content information and techniques, which can be included race/ethnicity and smoking, adapted from existing interventions (e.g., concerns regarding nicotine replacement/ cognitive behavioral therapy) or generated medication, religion/spirituality, and using culturally derived evidence. racism/discrimination as a specific stressor. Hooper et al. (2013) pilot-tested the culturally 5. Pilot-test the intervention protocol. specific CBT among treatment-seeking Before finalization, conduct a pilot test of African American tobacco smokers in a the intervention, specifically focusing on group format. Based on intervention feasibility and acceptability. Deliver the evaluations and session completion rates, intervention in a small sample within the the intervention was refined. intended population. Seek the feedback of end users and key stakeholders and refine the intervention accordingly. Webb Hooper et al. (2017) randomly assigned 6. Conduct a randomized controlled trial African American tobacco smokers to (RCT). RCTs remain the gold standard for receive a group-format culturally specific evaluating the efficacy of behavioral CBT versus a time and attention matched interventions. It is important to compare standard CBT. Results demonstrated that the effects of culturally specific smoking abstinence rates were two times interventions to their nonculturally greater in the culturally specific CBT group specific (i.e., standard) counterparts. compared to standard CBT. Hooper, Baker, and Robinson (2014) 7. Effectiveness research and initial translated the group-based culturally dissemination. Larger studies, delivered in specific tobacco intervention into a videoclinical and/or community settings, are based format, which is being evaluated in needed to evaluate culturally specific the context of a state tobacco quit line. intervention effects on behavior change in the real world. This step is important to ensure that these interventions are reaching the intended communities, with the goal of achieving health equity.

Addressing Underserved Populations and Disparities

to understand which interventions, and which behavior change techniques, work in which populations. The current body of theories and application frameworks needs to be extended with both culturally specific and socioeconomic influence points in order to improve understanding of the mechanisms underlying disparities in adaptive behaviors and, ultimately, inequities in well-being. Continuing to treat disparities in the effects of interventions as “nuisance” variables will not foster progress toward developing truly equitable behavior change interventions.

References Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50, 179–211. https://doi.org/10.1016/ 0749-5978(91)90020-T Allen, L., Williams, J., Townsend, N. et al. (2017). Socioeconomic status and non-communicable disease behavioural risk factors in low-income and lower-middle-income countries: A systematic review. The Lancet Global Health, 5, e277–e289. https://doi.org/10.1016/S2214109X(17)30058-X Attwood, S., van Sluijs, E., & Sutton, S. (2016). Exploring equity in primary-care-based physical activity interventions using PROGRESS-Plus: A systematic review and evidence synthesis. International Journal of Behavioral Nutrition and Physical Activity, 13, 60. https://doi.org/10.1186/ s12966-016-0384-8 Bauer, G. R. (2014). Incorporating intersectionality theory into population health research methodology: Challenges and the potential to advance health equity. Social Science and Medicine, 110, S10–S17. https://doi.org/10.1016/ j.socscimed.2014.03.022 Bimbi, D. S., Nanin, J. E., Parsons, J. T., Vicioso, K. J., Missildine, W., & Frost, D. (2006). Assessing gay and bisexual men’s outcome expectancies for sexual risk under the influence of alcohol and drugs. Substance Use and Misuse, 41, 643–665. https://doi.org/10.1080/10826080500411080

397

Burdette, A. M., Webb, N. S., Hill, T. D., & JokinenGordon, H. (2017). Race-specific trends in HPV vaccinations and provider recommendations: Persistent disparities or social progress? Public Health, 142, 167–176. https://doi.org/10.1016/j .puhe.2016.07.009 Doran, J. M., Pietrzak, R. H., Hoff, R., & HarpazRotem, I. (2017). Psychotherapy utilization and retention in a national sample of veterans With PTSD. Journal of Clinical Psychology, 73, 1259– 1279. https://doi.org/10.1002/jclp.22445 Engel, G. L. (1977). The need for a new medical model: A challenge for biomedicine. Science, 196(4286), 129–136. https://doi.org/10.1126/science.847460 Erves, J. C., Mayo-Gamble, T. L., Malin-Fair, A. et al. (2017). Needs, priorities, and recommendations for engaging underrepresented populations in clinical research: A community perspective. Journal of Community Health, 42, 472–480. https://doi.org/10.1007/s10900-016-0279-2 Fishbein, M., Triandis, H. C., Kanfer, F. H., Becker, M. H., Middlestadt, S. E., & Eichler, A. (2001). Factors influencing behavior and behavior change. In A. Baum, T. A. Revenson, & J. E. Singer (Eds.), Handbook of Health Psychology (pp. 3–17). Mahwah, NJ: Lawrence Erlbaum Associates. Ford, J. G., Howerton, M. W., Lai, G. Y. et al. (2008). Barriers to recruiting underrepresented populations to cancer clinical trials: A systematic review. Cancer, 112, 228–242. https://doi.org/ 10.1002/cncr.23157 Freedman, L. S., Simon, R., Foulkes, M. A. et al. 1995). Inclusion of women and minorities in clinical trials and the NIH Revitalization Act of 1993: The perspective of NIH clinical trialists. Controlled Clinical Trials, 16, 277–285; discussion 286–279, 293–309. https://doi.org/10.1016/0197-2456(95) 00048-8 Geller, S. E., Koch, A., Pellettieri, B., & Carnes, M. (2011). Inclusion, analysis, and reporting of sex and race/ethnicity in clinical trials: Have we made progress? Journal of Women’s Health, 20, 315– 320. https://doi.org/10.1089/jwh.2010.2469 Hankonen, N., Heino, M. T. J., Kujala, E. et al. (2017). What explains the socioeconomic status gap in activity? Educational differences in determinants of physical activity and screentime. BMC Public

398

¨ Z AND MONICA WEBB HOOPER BENJAMIN SCHU

Health, 17, 1–15. https://doi.org/10.1186/s12889016-3880-5 Hardeman, W., Sutton, S., Griffin, S. et al. (2005). A causal modelling approach to the development of theory-based behaviour change programmes for trial evaluation. Health Education Research, 20, 676–687. https://doi.org/10.1093/her/cyh022 Haughton, C. F., Silfee, V. J., Wang, M. L. et al. (2018). Racial/ethnic representation in lifestyle weight loss intervention studies in the United States: A systematic review. Preventive Medicine Reports, 9, 131–137. https://doi.org/10.1016/j .pmedr.2018.01.012 Hooper, M. W., Baker, E. A., & Robinson, R. G. (2014). Efficacy of a DVD-based smoking cessation intervention for African Americans. Nicotine and Tobacco Research, 16, 1327–1335. https://doi .org/10.1093/ntr/ntu079 Hooper, M. W., Larry, R., Okuyemi, K. et al. (2013). Culturally specific versus standard group cognitive behavioral therapy for smoking cessation among African Americans: An RCT protocol. BMC Psychol, 1, 15. https://doi.org/ 10.1186/2050-7283-1-15 Jahnel, T., Ferguson, S. G., Shiffman, S., & Schüz, B. (2019). Daily stress as link between disadvantage and smoking: An ecological momentary assessment study. BMC Public Health, 19, 1284. https://doi.org/10.1186/s12889-019-7631-2 Kagawa Singer, M. (2012). Applying the concept of culture to reduce health disparities through health behavior research. Preventive Medicine, 55, 356– 361. https://doi.org/10.1016/j.ypmed.2012.02.011 Kashubeck-West, S., & Szymanski, D. M. (2008). Risky sexual behavior in gay and bisexual men: Internalized heterosexism, sensation seeking, and substance use. Counseling Psychologist, 36, 595–614. https://doi.org/10.1177/ 0011000007309633 Kreuter, M. W., Lukwago, S. N., Bucholtz, D. C., Clark, E. M., & Sanders-Thompson, V. (2003). Achieving cultural appropriateness in health promotion programs: Targeted and tailored approaches. Health Education and Behavior, 30, 133–146. https://doi.org/10.1177/1090198102251021 Kwiatkowski, K., Coe, K., Bailar, J. C., & Swanson, G. M. (2013). Inclusion of minorities and women in cancer clinical trials, a decade later: Have we

improved? Cancer, 119, 2956–2963. https://doi .org/10.1002/cncr.28168 Laidley, T. M. (2011). The influence of social class and cultural variables on environmental behaviors: Municipal-level evidence from Massachusetts. Environment and Behavior, 45, 170–197. https:// doi.org/10.1177/0013916511416647 Lehne, G., & Bolte, G. (2017). Impact of universal interventions on social inequalities in physical activity among older adults: An equity-focused systematic review. International Journal of Behavioral Nutrition and Physical Activity, 14, 20. https://doi.org/10.1186/s12966-017-0472-4 Leventhal, H., Diefenbach, M., & Leventhal, E. A. (1992). Illness cognition: Using common sense to understand treatment adherence and affect cognition interactions. Cognitive Therapy and Research, 16, 143–163. https://doi.org/10.1007/ bf01173486 Lindbladh, E., Lyttkens, C. H., Hanson, B. S., Östergren, P., Isacsson, S. O., & Lindgren, B. (1996). An economic and sociological interpretation of social differences in health-related behaviour: An encounter as a guide to social epidemiology. Social Science and Medicine, 43, 1817–1827. https://doi .org/10.1016/S0277-9536(96)00087-1 Lorenc, T., Petticrew, M., Welch, V., & Tugwell, P. (2013). What types of interventions generate inequalities? Evidence from systematic reviews. Journal of Epidemiology and Community Health, 67, 190. https://doi.org/10.1136/jech-2012-201257 Manstead, A. S. R. (2018). The psychology of social class: How socioeconomic status impacts thought, feelings, and behaviour. British Journal of Social Psychology, 57, 267–291. https://doi.org/10.1111/ bjso.12251 Marmot, M. (2015). The Health Gap: The Challenge of an Unequal World. London: Bloomsbury. Marshal, M. P., Friedman, M. S., Stall, R. et al. (2008). Sexual orientation and adolescent substance use: A meta-analysis and methodological review. Addiction, 103, 546–556. https://doi.org/10.1111/ j.1360-0443.2008.02149.x McLeroy, K. R., Bibeau, D., Steckler, A., & Glanz, K. (1988). An ecological perspective on health promotion programs. Health Education and Behavior, 15, 351–377. https://doi.org/10.1177/ 109019818801500401

Addressing Underserved Populations and Disparities

Meader, N., King, K., Moe-Byrne, T. et al. (2016). A systematic review on the clustering and cooccurrence of multiple risk behaviours. BMC Public Health, 16. https://doi.org/10.1186/ s12889-016-3373-6 Meyer, I. H. (2003). Prejudice, social stress, and mental health in lesbian, gay and bisexual populations: Conceptual issues and research evidence. Psychological Bulletin, 129, 674–697. https://doi .org/10.1037/0033-2909.129.5.674 Michie, S., Richardson, M., Johnston, M. et al. (2013). The behavior change technique taxonomy (v1) of 93 hierarchically clustered techniques: Building an international consensus for the reporting of behavior change interventions. Annals of Behavioral Medicine, 46, 81–95. https://doi.org/ 10.1007/s12160-013-9486-6 Mickelson, R. A. (2003). When are racial disparities in education the result of racial discrimination? A social science perspective. Teachers College Record, 105, 1052–1086. https://doi.org/10.1111/ 1467-9620.00277 Myers, H. F. (2009). Ethnicity- and socio-economic status-related stresses in context: An integrative review and conceptual model. Journal of Behavioral Medicine, 32, 9–19. https://doi.org/ 10.1007/s10865-008-9181-4 O’Neill, J., Tabish, H., Welch, V. et al. (2014). Applying an equity lens to interventions: using PROGRESS ensures consideration of socially stratifying factors to illuminate inequities in health. Journal of Clinical Epidemiology, 67, 56– 64. https://doi.org/10.1016/j.jclinepi.2013.08.005 Odgers, C. L., Moffitt, T. E., Broadbent, J. M. et al. (2008). Female and male antisocial trajectories: From childhood origins to adult outcomes. Development and Psychopathology, 20, 673–716. https://doi.org/10.1017/S0954579408000333 Otado, J., Kwagyan, J., Edwards, D., Ukaegbu, A., Rockcliffe, F., & Osafo, N. (2015). Culturally competent strategies for recruitment and retention of African American populations into clinical trials. Clinical and Translational Science, 8, 460– 466. https://doi.org/10.1111/cts.12285 Ouellette, S. C., & DiPlacido, J. (2001). Personality’s role in the protection and enhancement of health: Where the research has been, where it is stuck,

399

how it might move. In A. Baum, T. A. Revenson, & J. E. Singer (Eds.), Handbook of Health Psychology (pp. 175–193). Mahwah, NJ: Lawrence Erlbaum Associates. Petrovic, D., de Mestral, C., Bochud, M. et al. (2018). The contribution of health behaviors to socioeconomic inequalities in health: A systematic review. Preventive Medicine, 113, 15– 31. https://doi.org/10.1016/j.ypmed.2018.05.003 Piff, P. K., Kraus, M. W., Côté, S., Cheng, B. H., & Keltner, D. (2010). Having less, giving more: The influence of social class on prosocial behavior. Journal of Personality and Social Psychology, 99, 771–784. https://doi.org/10.1037/a0020092 Ready, D. D. (2010). Socioeconomic disadvantage, school attendance, and early cognitive development: The differential effects of school exposure. Sociology of Education, 83, 271–286. https://doi.org/10.1177/0038040710383520 Resnicow, K., Baranowski, T., Ahluwalia, J. S., & Braithwaite, R. L. (1999). Cultural sensitivity in public health: Defined and demystified. Ethnicity and Disease, 9, 10–21. Rounsaville, B. J., Carroll, K. M., & Onken, L. S. (2001). A stage model of behavioral therapies research: Getting started and moving on from Stage I. Clinical Psychology: Science and Practice, 8, 133–142. https://doi.org/10.1093/clipsy.8.2.133 Schüz, B. (2017). Socio-economic status and theories of health behaviour: Time to upgrade a control variable. British Journal of Health Psychology, 22, 1–7. https://doi.org/10.1111/bjhp.12205 Schüz, B., Brick, C., Wilding, S., & Conner, M. (2019). Socioeconomic status moderates the effects of health cognitions on health behaviors within participants: Two multibehavior studies. Annals of Behavioral Medicine. https://doi.org/10.1093/abm/ kaz023 Schüz, B., Li, A. S. W., Hardinge, A., McEachan, R. R. C., & Conner, M. (2017). Socioeconomic status as a moderator between social cognitions and physical activity: Systematic review and metaanalysis based on the Theory of Planned Behavior. Psychology of Sport and Exercise, 30, 186–195. https://doi.org/10.1016/j.psychsport.2017.03.004 Schüz, B., Wurm, S., Ziegelmann, J. P. et al. (2012). Contextual and individual predictors of physical

400

¨ Z AND MONICA WEBB HOOPER BENJAMIN SCHU

activity: Interactions between environmental factors and health cognitions. Health Psychology, 31, 714–723. https://doi.org/10.1037/a0027596 Shaw, M., Galobardes, B., Lawlor, D. A., Lynch, J., Wheeler, B., & Smith, G. D. (2007). The Handbook of Inequality and Socioeconomic Position. Bristol: Policy Press. Smith, T. B., Rodríguez, M. D., & Bernal, G. (2011). Culture. Journal of Clinical Psychology, 67, 166– 175. https://doi.org/10.1002/jclp.20757 Spoont, M. R., Nelson, D. B., Murdoch, M. et al. (2015). Are there racial/ethnic disparities in VA PTSD treatment retention? Depression and Anxiety, 32, 415–425. https://doi.org/10.1002/ da.22295 Trinidad, D. R., Perez-Stable, E. J., White, M. M., Emery, S. L., & Messer, K. (2011). A nationwide analysis of US racial/ethnic disparities in smoking behaviors, smoking cessation, and cessationrelated factors. American Journal of Public Health, 101, 699–706. https://doi.org/ 10.2105/ Ajph.2010.191668 U.S. Department of Health and Human Services. (2008). The Secretary’s Advisory Committee on National Health Promotion and Disease Prevention Objectives for 2020. Phase I Report: Recommendations for the Framework and Format of Healthy People 2020. Retrieved from Section IV: Advisory Committee Findings and Recommendations. www.healthypeople .gov/sites/default/files/PhaseI_0.pdf U.S. Department of Health and Human Services. (2010). National Partnership for Action to End Health Disparities: The National Plan for Action. Atlanta, GA: U.S. Department of Health and Human Services. https://minorityhealth.hhs.gov/ npa/files/plans/toolkit/npa_toolkit.pdf Webb Hooper, M., Antoni, M. H., Okuyemi, K., Dietz, N. A., & Resnicow, K. (2017). Randomized controlled trial of group-based culturally specific cognitive behavioral therapy among African American smokers. Nicotine and Tobacco Research, 19, 333–341. https://doi.org/10.1093/ ntr/ntw181 Webb Hooper, M., Asfar, T., Unrod, M. et al. (2019). Reasons for exclusion from a smoking cessation trial: An analysis by race/ethnicity. Ethnicity and

Disease, 29, 23–30. https://doi.org/10.18865/ ed.29.1.23 Webb Hooper, M., Mitchell, C., Marshall, V. et al. (2019). Understanding multilevel factors related to urban community trust in healthcare and research. Unpublished manuscript, Case Comprehensive Cancer Center, Cleveland, OH. Webb Hooper, M., Payne, M., & Parkinson, K. A. (2017). Tobacco cessation pharmacotherapy use among racial/ethnic minorities in the United States: Considerations for primary care. Family Medicine and Community Health, 5, 193–203. https://doi.org/10.15212/ FMCH.2017.0138 Webb, M. S., Baker, E. A., & de Ybarra, D. R. (2010). Effects of culturally specific cessation messages on theoretical antecedents of behavior among lowincome African American smokers. Psychology of Addictive Behaviors, 24, 333–341. https://doi.org/ 10.1037/A0018700 Webb, M. S., Francis, J., Hines, B. C., & Guarles, F. B. (2007). Health disparities and culturally specific treatment: Perspectives and expectancies of African American smokers. Journal of Clinical Psychology, 63, 1247–1263. https://doi.org/ 10.1002/Jclp.20437 Wheldon, C. W., Kaufman, A. R., Kasza, K. A., & Moser, R. P. (2018). Tobacco use among adults by sexual orientation: Findings from the Population Assessment of Tobacco and Health Study. LGBT Health, 5, 33–44. https://doi.org/10.1089/ lgbt.2017.0175 Williams, D. R., Neighbors, H. W., & Jackson, J. S. (2003). Racial/ethnic discrimination and health: Findings from community studies. American Journal of Public Health, 93, 200–208. https://doi .org/10.2105/ajph.93.2.200 Williams, W. W., Lu, P., O’Halloran, A. et al. (2017). Surveillance of vaccination coverage among adult populations – United States, 2015. Morbidity and Mortality Weekly Report, 66, 1–28. https://doi.org/ 10.15585/mmwr.ss6611a1 Wilson, S. M., Gilmore, A. K., Rhew, I. C., Hodge, K. A., & Kaysen, D. L. (2016). Minority stress is longitudinally associated with alcohol-related problems among sexual minority women. Addictive Behaviors, 61, 80–83.

28 Behavior Change in Community Contexts Edison Trickett and Susan Paterson

Practical Summary People’s behavior depends not only on who “we” are but where “we” are. This chapter focuses on different ways that efforts have been made to change aspects of the communities in which we live to improve the choices and resources for members of the community. The concept of ecology is used to describe communities and their influence on our behavior. Some of these influences are direct, such as the influence of schools on children attending them; some indirectly affect community members, such as state or federal regulating of the drinking age that is reflected in local community settings. The chapter highlights three areas where changes in community life may be expected to promote potentially positive changes in individual behavior: (1) social norms or shared expectations; (2) creating new social settings in the community to meet previously unmet needs; and (3) local involvement in the policy-making process.

28.1 Introduction The present chapter provides an ecological framework and select examples of efforts to change the community context and, in so doing, provides new and ongoing resources for individuals living in the community. Unlike many community interventions directed toward changing individual behavior, the focus here is thus on the ecological context of communities. The chapter is organized as follows: First, some of the varied meanings of the concept of ecology for understanding the community context and individual behavior in it are reviewed. Perspectives prevalent in public health and described in other chapters of this volume are identified and briefly contrasted (see Chapters 17 and 18, this volume). The chapter then briefly highlights three efforts to alter aspects of the community context: (1) changing local social norms; (2) creating new local social settings; and (3) changing local

social policies. The chapter concludes with a series of summary comments and recommendations for future research on changing individual behavior through altering local ecology.

28.2 Ecological Perspectives on Conceptualizing and Changing the Community Context Ecology has been historically defined as the study of the relationship between organisms and their environments. Many differing perspectives on the term have been employed over time. Within the social sciences, sociologist Park (1952) is often credited with the popularization of human ecology as a perspective for understanding urban environments. Lewin (1951), Barker (1968), and https://doi.org/10.1017/9781108677318.028

402

EDISON TRICKETT AND SUSAN PATERSON

Moos (1974) likewise provided influential work placing individuals in context and in describing aspects of the community context itself. Lewin, for example, elaborated on the environment as a force field of influences on individual behavior. Barker (1968) viewed the environment as defined by its behavior settings, each of which elicits differing behaviors consistent with the agendas of the setting. Moos (1974) focused on the perceived social climate of high-impact settings such as school classrooms and work environments and how differing social climates affect people in them. Each has contributed to a larger cognitive map of how to conceptualize aspects of local ecology. Both Lewin and Moos similarly provide participatory strategies for changing environments – Lewin through what later became known as “action research” (Sanford, 1970) and Moos through an organizational survey feedback process (Pierce, Trickett, & Moos, 1972). More recently, Bronfenbrenner (1979) provided a heuristically useful systems theory perspective on the ecological context. Initially, the framework described four nested systems that exert influence on individual behavior – micro-, meso-, exo-, and macrosystems. For example, the family serves as an example of a microsystem influencing individual behavior, while culture is viewed as a macrosystem where influence on individual development is mediated through more local systems such as neighborhoods and schools as well as family. His later addition of the “chronosystem” added a time dimension to acknowledge cohort development in which individuals are nested as well. Bronfenbrenner’s conception of systems has been frequently translated as forces acting at different ecological “levels” to influence individual behavior, such as home and school influence over child development. The “levels” notion has become enormously influential in conceptualizing community interventions through an ecological lens. Within public health, early work by McLeroy et al. (1988), Simons-Morton et al. (1988), and Stokols et al. (1996) acknowledges

Bronfenbrenner’s notion of levels and their reciprocity as well as the active role of individuals in modifying their own contexts. Such frameworks guide the process of developing interventions intended to affect multiple levels and segments of the community of concern. McLeroy et al. (1988) describe potential interventions corresponding to the varied levels of the ecological context, while Stokols et al. (1996) outline two types of community interventions directed toward changing local ecology: environmental change strategies addressing specific environmental changes related to health (e.g., exercise programs) or safety (e.g., seat belts) and multilevel social ecological approaches that address multiple aspects and levels of the community context to affect change (e.g., a program for injection drug users targeting policies around access to nonprescription syringes in pharmacies and pharmacist training as well as individual counseling; Fuller et al., 2007). Within community psychology, an ecological perspective was initiated in the 1960s (Kelly, 1966, 2006; Trickett, Kelly, & Todd, 1972) to guide the development of the then new field of community psychology. This community development framework incorporates the multilevel environmental assessment and systems perspective of the other ecological perspectives mentioned earlier in this section. Guiding the environmental assessment process are a series of four ecological principles drawn from field biology and applied metaphorically to human communities: (1) interdependence of parts of the community as a system (fundamental aspect of systems theory); (2) cycling of resources, focusing on the community as a potential resource pool to be identified, engaged, and developed in the intervention process; (3) adaptation or demands for survival in the community (e.g., how local ecology of lives reflects responses to local conditions, opportunities, and constraints); and (4) succession, the time dimension of communities or

Behavior Change in Community Contexts

specific groups in communities focusing on both past experiences and future hopes. For social justice, scientific, and pragmatic reasons (Kelly, 1970), great attention is paid to the role of collaboration with local groups and organizations in developing, implementing, and evaluating change efforts. Together, these varied meanings of ecology have yielded data on the critical role of context in understanding individual behavior and, in so doing, have identified multiple ways and places in the community context to influence it. The next section describes three topics reflecting differing aspects of the community context: changing local social norms, creating local social settings, and changing/creating local social policy. Throughout the research on these topics is a dual emphasis on both the processes involved in changing the ecological context and the outcomes of such changes.

28.3 Community Social Norms Social norms represent shared assumptions, beliefs, and expectations about what constitutes appropriate behavior (see also Chapter 16, this volume). They reflect cultural products (values, customs, and traditions) and assumptive structures for understanding how individuals actually behave (descriptive norms) as well as how acceptable a behavior is in the community, regardless of one’s actual behavior (injunctive norms). Norms may be unspoken and are sometimes unable to be rationalized (“we just do it that way, that’s all”). Norms are found across multiple domains of the ecology of everyday community life. They are reflected in everyday interpersonal negotiations at home or work, the crafting of community celebrations, how elderly parents are cared for by their adult children, the degree of adolescent autonomy granted by parents, and how sexual encounters are negotiated. They exercise great influence over behavior by providing both a worldview and sanctions or rewards for enacting or transgressing them (Linos et al., 2013).

403

One major focus in the United States involves the influence of norms on such social problem behaviors as drinking, smoking, and sexual violence. For example, multiple studies of the influence of alcohol-related norms on US college student drinking support the influence of peer norms over those of faculty or parents (Berkowitz, 2004) in predicting where, with whom, and how much drinking occurs. Complex pathway models of how norms influence behavior have been developed. In an ecological study on underage drinking, Lipperman-Kreda et al. (2010) found direct and indirect relationships among adolescents’ perception of community alcohol norms and multiple aspects of local ecology, such as enforcement of underage drinking laws, perceived alcohol availability, and perceived drinking by peers. With respect to sexual violence, the World Health Organization (WHO, 2009) outlines multiple ways in which social and cultural norms in varied countries contribute to violence broadly defined (e.g., child maltreatment, intimate partner violence, community violence related to cultural intolerance). This same report outlines preventive programs across these areas. In a study of contraceptive behavior among Kenya and Ethiopian men and women, Dynes et al. (2012) found that cultural norms involving the gendered nature of power relations (male dominance in household decision-making, preference for sons, and decisions to have another child) and fertility preferences (number and gender of children) affect critical family planning decisions. For example, the association of condom use for women with infertility and promiscuity constrains use of family planning services. Banyard et al. (2004) document the prevalence of sexual violence in the United States in particular and embed its origins in the context of an ecological model that includes community and cultural norms as well as interpersonal factors. Together, this body of research provides the rationale for changing community ecology through normative change in

404

EDISON TRICKETT AND SUSAN PATERSON

varied groups in the community or the community itself.

28.3.1 Norm-Related Community Interventions One significant focus of intervention research on normative change has addressed sexual violence on college campuses. For example, Banyard et al. (2004) provide a detailed review of literature involving attempts to change college norms around sexual violence by utilizing a bystander intervention model to increase awareness and responsiveness of college students and college campuses more generally to sexual harassment. These efforts explicitly move toward a campus-wide process of “creating new norms where it is clear to all that harassment is unacceptable” (Bond, 1995, p. 168 cited in Banyard et al., 2004, p. 66). Perkins and Berkowitz (1986) focus on interventions designed to alter perceived norms with respect to excessive alcohol drinking on college campuses. The ecological emphasis involves multiple levels of a community, what is normatively acceptable in that environment, and where in the community leverage points for altering norms are found. In a subsequent review, Perkins (2002) reported that one significant source of drinking behavior was the misperception and exaggeration of how much drinking peers view as permissible and how much they actually drink. Multilevel interventions, including varied forms of in-person and media communication about actual drinking patterns, have been reported as promising in reducing high-risk drinking rates. There are also examples of how normative change may be integrated into a larger organizational change effort. Griffith et al. (2007) describe dismantling racism in a health care institution. By assessing the system’s practices, procedures, and policies, organizational norms were addressed in four ways:

(a) helping people who are committed to equity and anti-racist organizational values establish norms and a culture where people hold each other accountable for their behavior and the impact of their actions; (b) creating a culture where decisions about the allocation and use of money and resources consider their implications for social equity; (c) fostering organizational norms where decisions about how and what work gets done consider the racial equity; and (d) prioritizing of organizational goals and objectives is congruent with anti-racist organizational values (Griffith et al., 2007, p. 387).

At the community level, Schensul et al. (2015) studied gender normative change as part of a larger project targeted toward sexual risk reduction and improving sexual health among married couples in Mumbai, India. The larger intervention consisted of individual and couples counseling, the development of a women’s health center, and a media campaign that promoted gender equality. Women were randomly assigned to one of the two counseling conditions or a control group that received only medical care. A gender equity scale was administered to men and women in the community, nongovernmental organization (NGO) staff involved in the community, and local Imams (Muslim religious leaders) who were tasked with disseminating gender messages to their congregations. As a result of the three-year intervention, NGO staff, Imams, and male community members showed a marked increase in positive attitudes toward gender equality, whereas women did not show a significant change. The authors cite two possible explanations. The first involved who delivered the intervention. Men received gender equity messaging from Imams in Friday prayers, whereas women received these messages at five big events across the course of the intervention. The second involved the possibility for women that taking on new equitable norms could have negative ripple effects that could lead to violence or health issues.

Behavior Change in Community Contexts

As suggested in the previous study, the process issue of who implements relevant normative change activities is critical. For example, Kelly and Kalichman (2002) review numerous successful interventions among gay men at risk of HIV/ AIDS utilizing popular opinion leaders (POLs) in gay bars as disseminators of information regarding risk-reduction behaviors. The work is based on Rogers’s diffusion of innovations theory, “which postulates that new behavioral trends in a population can be initiated when sufficient number of natural popular opinion leaders within the population are observed to model new innovative behaviors” (Kelly & Kalichman, 2002, p. 629). A similar logic in the school context was used by Atkins et al. (2008). Here, key opinion leader teachers were selected to disseminate school and classroom practices for aiding children with attention deficit hyperactivity disorder (ADHD). In both instances, resources indigenous to the settings involved were drawn on in carrying out activities intended to create normative change. More generally, this literature suggests that targeting descriptive norms is particularly effective in reinforcing and changing individual behavior. For example, in a meta-analysis on college drinking behaviors, Borsari and Carey (2003) found that teaching students about the actual drinking norms of their peers was successful not only in changing students’ perceptions of their peers’ drinking but in decreasing their own drinking as well. Similarly, Sheeran, Abraham, and Orbell’s (1999) meta-analysis of 121 studies on condom use among heterosexual individuals found that individual perception of both descriptive and prescriptive peer group norms was correlated with an individual’s own condom use.

28.4 Creating Social Settings Responsive to Community Concerns Creating social settings responsive to community concerns represents another means of changing

405

the community context to address community needs and issues. Sarason (1972) defines settings as any instance in which “two or more people come together in new relationships over a sustained period of time in order to achieve certain goals.” His focus was on how to create humane and effective social settings that provide value both to those working in them and to those served by the setting. Decisions about what kinds of settings to create are based on community needs assessments, local initiatives of individuals or organizations energized by a social cause, and/ or external funding opportunities congruent with the mission of existing local organizations. Over time, descriptions of the creation and impact of a wide variety of community settings have emerged. Within the realm of education, for example, public schools have created Newcomer Centers to aid the transition of immigrant and refugee children and adolescents to school and life in the new country (Short, 2002). Often structured as schools within a larger public school, these settings provide a separate academic curriculum for children, including but transcending English language acquisition, acculturative experiences, and often services for immigrant families linking them to other relevant social or legal resources. Other kinds of community settings have been created to provide peer support and mutual assistance for individuals with serious mental illness (Toro, Rappaport, & Seidman, 1987) or those dealing with addiction (Jason et al., 2006). Settings designed to promote specific social changes or health issues have also been described. For example, Brodsky (2004) describes the structure and functioning of the Revolutionary Association of the Women of Afghanistan (RAWA). This group advocates for women’s rights while simultaneously promoting the broader goals of community development and positive social change. Campbell, Nair, and Maimane (2007) provide a process description of creating safe social spaces for dialogue about

406

EDISON TRICKETT AND SUSAN PATERSON

Sidebar 28.1 New settings affect communities: community gardens as an example

Community gardens are plots of land that are used to grow and harvest food for and by local residents. They have been shown to impact communities by creating both intended and unintended changes at multiple levels of analysis. In a review of community gardens intervention research, Okvat and Zautra (2011) describe psychological, community, and environmental outcomes, as well as the ripple effects created by these spaces. At the individual level, community gardens have increased education on sustainability and food production in and for the community (Corkery, 2004; Walter, 2013). They have also made people feel less isolated (Milligan, Gatrell, & Bingley, 2004). At the community level, gardens have engendered a sense of community pride (Kaplan & Kaplan, 1989). One major ripple effect from the development of community gardens is increased community activism on the part of local residents on issues unrelated to gardening itself (Armstrong, 2000). Through regular meaningful contact with other people in their communities, community gardeners became more active in local politics. In one instance, they successfully fought to keep a local supermarket in their community (Armstrong, 2000). In another, community gardens became a catalyst for activism among Puerto Ricans fighting against the gentrification of their neighborhood by increasing a sense of ownership and control over their community (Martinez, 2010).

promoting a health-enabling community response to HIV/AIDS. Finally, the creation of community gardens provides an example of the impact of such settings (see Sidebar 28.1). Sidebar 28.1 provides an overview of some of the kinds of settings created to improve community well-being. To take a more focused look at the kinds of issues and questions related to the creation of community settings, the creation of food pantries is selected as an extended example.

28.4.1 Food Pantries The issue of food insecurity affects one in eight families (Coleman-Jensen et al., 2017), particularly among low-income households. Food insecurity involves the need to figure out how “to acquire food in socially acceptable ways … without resorting to emergency food supplies, scavenging, stealing, or other coping strategies” (Hamelin et al., 1999, p. 528S). Food insecurity highlights the many relationships between

nutrition and health, including (1) physical manifestations (fatigue, illness); (2) psychological consequences (stress, fear of losing child); (3) family disturbances related to food acquisition involving parent-child relations; and (4) social network implications such as inviting friends to dinner. In response to this community need, many communities have created food pantries, generally not-for-profit community organizations that distribute food to people who suffer from hunger and food insecurity. Who do food pantries serve? Descriptive data suggest that rates of food insecurity of those using food pantries exceed its frequency in the general US population and can approach 90 percent (Greder et al., 2007). They include a diverse group of clients depending on location, from migrant workers in North Carolina (Quandt et al., 2004) to the homeless, the working poor, undocumented immigrants, and a small proportion of elderly in Pomona, California (Algert et al., 2006). Client food pantry needs may differ

Behavior Change in Community Contexts

depending on their rural/urban location, with rural clients more likely to develop longer-term involvement in food pantries and pantries themselves being more likely to know clients and be less structured than those in urban areas. While originally designed to provide emergency stopgap food resources, food pantries increasingly serve longer-term clients (Remley et al., 2013). This changing clientele has caused some food pantries to add additional services responsive to the longer-range concerns of clients, including nutrition programs, health referrals, food stamp outreach, and home energy assistance (Remley et al., 2013). What organizational issues do food pantries face? An ecological perspective focuses on settings as social systems facing challenges affecting their effectiveness and sustainability. With respect to clients of food pantries, for example, cultural and language issues affect those from diverse backgrounds who want more culturally appropriate foods and who have difficulty participating in services such as nutrition workshops because of language issues (Remley et al., 2013). Staff concerns include uncertainty about how to include enough appropriately nutritious food because of inconsistent donations and higher costs of such foods (Bazerghi et al., 2016). In

407

addition, because of the nutritional goal of food pantries, how to decide on what foods to include or exclude becomes a policy issue with implications for both clients and relationships with food donors and community partners (Handforth et al., 2013). One particular organizational policy issue that has been discussed in the literature involves the degree to which pantries allow clients to choose foods versus being given food preselected by the pantry. Traditionally, clients are given preselected bags of food where selection is controlled by food pantry staff and nutritionists. The empowerment logic behind client food selection, however, purports that client choice will result in greater client dignity (Remley et al., 2013) and less food being wasted since it is client-chosen. Sidebar 28.2 describes a study of the importance of choice for food pantry clients. What other functions can the new setting serve? Finally, research has shown that those who frequent food pantries are also at elevated risk for other health-related outcomes. Thus, food pantries also provide access to clients who may benefit from other services. For example, Bencivenga et al. (2008) carried out a mammography screening intervention in eighteen food pantries in rural Indiana. Seligman et al. (2015)

Sidebar 28.2 Effects of client choice in food pantries

Martin et al. (2012) reported a randomized trial comparing a client-choice food pantry with traditional ones not offering choice. Freshplace, a collaboration between university and community partners, began with an assessment of local poverty and developed a food pantry program based on a stages of change model. The model included fresh food, case management, services, and referrals and offered food pantry clients choice about which foods they could take. Clients also met with a project manager to develop a personal plan for becoming food-secure and selfsufficient. A randomized control trial (N = 227 overall) comparing Freshplace clients with clients of other food pantries not offering such a program found overwhelming support for the positive effects of the Freshplace program on subsequent client diet, self-efficacy, and food security (Martin et al., 2013, 2016). Such research, along with other reports (e.g., Remley et al., 2013), strongly supports the value of client choice in promoting increased nutrition as well as confidence to make choices on one’s own.

408

EDISON TRICKETT AND SUSAN PATERSON

developed a diabetes-prevention intervention in food pantries that included diabetic-appropriate food, blood sugar monitoring, primary care referral, and self-management support.

28.5 Social Policy and Community Change In addition to changing social norms and creating social settings, creating or amending social policy represents another strategy to change community ecology. Two general kinds of policies are prominent in the literature: those that affect regulations, laws, and procedures and those that involve environmental change. Regulating laws governing the sale of alcohol or smoking in the workplace represent examples of the former; creating hiking trails or community developments whose physical design promotes walking represent examples of the latter (McLeroy et al., 1988; Sallis et al., 2006). The range of specific policy topics is broad and supported by empirical literature linking policy targets with behavioral outcomes, often related to physical or mental health. For example, Sallis et al. (2006) reviewed research linking physical activity to varied outcomes such as reduced obesity and outline the role of policy in promoting physically active living in four life domains: recreation, transport, occupation, and household. Schmid et al. (1995) and Matson-Kaufman et al. (2005) reviewed policy literature regarding cardiovascular health in terms of regulatory policies and organizational changes to promote health such as the effect of food labeling in grocery stores on the sale of low-fat foods. Alcohol consumption was the focus of Elder et al.’s (2010) review, while local stakeholder involvement in health equity policy decisions has been stressed in the communitybased participatory research movement (CacariStone et al., 2014; Gong et al., 2009). Causal pathways and policy. The causal pathways of influence of policy on behavior differ

depending on the outcome of concern and the organizations through which policy is implemented. For example, Elder et al. (2010) suggest that increased alcohol taxes ultimately affect alcohol consumption by first increasing the price paid by consumers. This, in turn, deceases the demand for targeted alcoholic beverages while also changing the demand for nontargeted ones, leading to a deceased consumption of such beverages overall. This decreased consumption leads to less harmful consequences of excessive drinking. Story et. al. (2008) provide a thorough overview of an ecological perspective on creating healthy eating environments where the potential impact of policy is mediated by discrete social settings in the community, including home, childcare, schools, workplaces, retail food stores, and restaurants. For example, both availability and accessibility of healthy foods in the home have been associated with increased healthy food intake for adolescents, while workplace health-promotion programs have been found to increase fruit and vegetable intake. Thus, one perspective on assessing policy impact is to isolate a specific policy change and follow its impact on important health outcomes. Another set of policy-related interventions embeds specific policy changes in a larger strategy of community change involving multiple levels of ecological influence (McLeroy et al., 1988). This multilevel strategy is supported by research showing the independent and interactive contributions of factors at differing ecological levels on individual behavioral outcomes of local concern. For example, in a multilevel analysis of literature on influenza vaccination, Kumar et al. (2012) reported significant contributions across levels of the ecological model from intrapersonal to community, with 8 percent of the variance in outcomes attributable to policy-level influences. Prototypical of these studies is the COMMIT study (Glasgow et al., 1997), a multi-community trial to reduce cigarette smoking. Sidebar 28.3

Behavior Change in Community Contexts

409

Sidebar 28.3 Spotlight on policy: the COMMIT study

The Community Intervention Trial for Smoking Cessation (COMMIT) was a longitudinal randomized control trial testing a smoking cessation program in eleven communities. Overall, the intervention involved was directed toward several different levels of the community context: In the workplace, it (1) encouraged smoking cessation policies at the organizational level and (2) offered greater work site access to smoking cessation resources; in the community, it (1) worked with health care providers and (2) designed a mass media campaign to educate the public about smoking and tobacco use (Fisher, 1995). Glasgow et al. (1997) used this data to look specifically at the relationship between stringent anti-smoking policies in the workplace and success with smoking cessation. Results showed that people in smoke-free workplaces were significantly more likely to attempt quitting and also more likely to sustain their smoking cessation. Those who continued to smoke were more likely to smoke less overall if their workplace was smoke-free than those workplaces that had less stringent anti-smoking policies. The authors estimate that, if all workplaces in the United States had stringent smoke-free policies, an additional 178,000 smokers would stop smoking.

provides more in-depth information about the multiple levels and effects of policy change from the COMMIT study.

28.5.1 Empowering Local Processes to Influence Policy One final aspect of policy as a means of affecting local behavior involves mobilizing local communities to influence the policy-making process. Here, the community-based participatory research movement (CBPR) (Wallerstein et al., 2018) has generated multiple examples of creating empowering local processes to influence policy. Cacari-Stone et al. (2014) provide an example of how “locality-based research strategies” influenced policies involving issues of health equity. Here, local stakeholder groups were brought together to develop policy-relevant evidence relevant to the effects of long-term air pollution. The diverse local policy ideas resulting from this process of civic engagement and data gathering “were instrumental in facilitating policy formulation (the focus of these examples)

through systematic problem identification, setting the agenda by bringing legitimate attention to community issues, constructing policy alternatives, and adopting politically feasible policy objectives” (p. 1620). Sidebar 28.4 provides a description of how community members were engaged in another such project (Gong et al., 2009).

28.6 Summary and Conclusions The present chapter has offered an ecological perspective on ways of conceptualizing and altering aspects of the community context to affect individual behavior change. It has provided diverse examples of research focusing on social norms, social settings, and social policy as examples of affecting individual behavior through altering aspects of local ecology. In general, data are supportive of the critical role of the community context in affecting both immediate behaviors and longer-range physical and psychological outcomes. Multiple levels of the ecological community context affect individual behavior

410

EDISON TRICKETT AND SUSAN PATERSON

Sidebar 28.4 Spotlight on policy: How can community members be engaged in the policy formation process?

Gong et al. (2009) discuss how community members can be engaged in the policy process and help shape future policy decisions. Using a community-based participatory research (CBPR) methodology, researchers worked with the United Long-Term Care Workers Union and other partners “in an intervention project targeting the intersection of the home and work environment for two economically marginalized populations – low-income elderly and disabled persons and the lowwage in-home care workers who help them live independently” (Gong et al., 2009, p. S531). The team developed intervention materials and identified barriers to help formulate policy at the organizational and public policy levels. Stakeholders also helped develop and analyze interview protocols with community members. After collecting data, collective brainstorming was used to develop organizational policy suggestions and public policy ideas responsive to the specific communities. These community partnership activities, involving a wide array of local stakeholders, led to a shared belief in the importance of worker safety and empowered community partners to formulate policy directions designed to improve safety in the process of providing care in the home.

and can be conceptual guides for research and leverage points for intervention. However, reviewing the literature on research and intervention on community context effects heightens a number of research questions that present conceptual, methodological, and funding challenges. Conceptually, ecology rests on a systems theory of interdependence involving the ecosystem as the unit of analysis with changes in any aspect of the ecosystem causing ripple effects in other parts of the system. While counter examples are evident (e.g., Hirsch, Levine, & Miller, 2007), the literature on changing aspects of local ecology still adopts a more program-focused perspective than “intervention as event in system” (Hawe, Shiell, & Riley, 2009) perspective and normatively does not address ripples in the community context (Trickett, Espino, & Hawe, 2011). In addition, theory about how differing levels of the ecological context interact is typically not integrated into this research, even when differing levels are sites for intervention. Methodological challenges flow in part from the conceptual ones. Some reflect the difficulties

pinpointing and tracking the complex sets of multilevel ecological influences on behavior in any community context. Others reflect the lack of context-relevant information in research designs focusing on establishing causal relationships between inputs and outcomes through randomization or quasi-experimental designs. Part of the context-relevant information involves descriptive data about (1) the specific places involved and (2) the kinds of research relationships formed between the data gatherers and those local settings and individuals through whom and/or on whom data are gathered. Ecological perspectives are inclusive of such aspects of community research as relevant to theory building, interpretation, and generalizability of findings. The inclusion of qualitative and mixed method designs are potential antidotes here (see also Chapter 30, this volume). In addition, while many efforts to alter local ecology address issues of disparities such as gender norms, poverty, and policies affecting marginalized individuals (see also Chapter 27, this

Behavior Change in Community Contexts

volume), additional research could focus more explicitly on the social justice implications of the topics addressed. Such work might address such questions as (1) which community members are most at risk if they attempt to change discriminatory norms and practices and (2) who benefits most and least in communities from the creation of new settings. While alluded to by Schensul et al. (2015), issues of how power, justice, and structural forces more broadly differentially affect groups in the community can inform our understanding of how changing local ecology may affect the just allocation of resources and for whom. Finally, while there are obvious exceptions, efforts at influencing individual behavior indirectly through changing community ecology typically assume a longer time frame to accomplish than efforts directly oriented toward individuallevel change. Further, as the CBPR literature makes clear (e.g., Wallerstein et al., 2018) such efforts may involve a series of discrete and emerging efforts over time rather than a specific set of preordained programmatic activities. Such a perspective requires great flexibility in how resources are generated, used, and redeployed as circumstances in communities change (Head & Alford, 2015). The role of external funding in such circumstances can inadvertently constrain efforts to learn about the complexities and time frames for understanding and addressing community change.

References Algert, S. J., Agrawal, A., & Lewis, D. S. (2006). Disparities in access to fresh produce in lowincome neighborhoods in Los Angeles. American Journal of Preventive Medicine, 30, 365–370. https://doi.org/10.1016/j.amepre.2006.01.009 Armstrong, D. (2000). A survey of community gardens in upstate New York: Implications for health promotion and community development. Health and Place, 6, 319–327. https://doi.org/10.1016/ S1353-8292(00)00013-7

411

Atkins, M. S., Frazier, S. L., Leathers, S. J. et al. (2008). Teacher key opinion leaders and mental health consultation in low-income urban schools. Journal of Consulting and Clinical Psychology, 76, 905–908. https://doi.org/10.1037/a0013036 Banyard, V. L., Plante, E. G., & Moynihan, M. M. (2004). Bystander education: Bringing a broader community perspective to sexual violence prevention. Journal of Community Psychology, 32, 61–79. https://doi.org/10.1002/jcop.10078 Barker, R. (1968). Ecological Psychology: Concepts and Methods for Studying the Environment of Human Behavior. Stanford, CA: Stanford University Press. Bazerghi, C., McKay, F. H., & Dunn, M. (2016). The role of food banks in addressing food insecurity: A systematic review. Journal of Community Health, 41, 732–740. https://doi.org/10.1007/s10900-0150147-5 Bencivenga, M., DeRubis, S., Leach, P., Lotito, L., Shoemaker, C., & Lengerich, E. J. (2008). Community partnerships, food pantries, and an evidence-based intervention to increase mammography among rural women. The Journal of Rural Health, 24, 91–95. https://doi.org/ 10.1111/j.1748-0361.2008.00142.x Berkowitz, A. D. (2004). The social norms approach: Theory, research, and annotated bibliography. www.alanberkowitz.com/articles/social_norms .pdf Borsari, B., & Carey, K. B. (2003). Descriptive and injunctive norms in college drinking: A metaanalytic integration. Journal of Studies on Alcohol, 64, 331–341. Brodsky, A. E. (2004). With All Our Strength: The Revolutionary Association of the Women of Afghanistan. New York: Routledge. Bronfenbrenner, U. (1979). The Ecology of Human Development: Experiments by Nature and Design. Cambridge, MA: Harvard University Press. Cacari-Stone, L., Wallerstein, N., Garcia, A. P., & Minkler, M. (2014). The promise of communitybased participatory research for health equity: A conceptual model for bridging evidence with policy. American Journal of Public Health, 104, 1615–1623. https://doi.org/10.15288/ jsa.2003.64.331

412

EDISON TRICKETT AND SUSAN PATERSON

Campbell, C., Nair, Y., & Maimane, S. (2007). Building contexts that support effective community responses to HIV/AIDS: A South African case study. American Journal of Community Psychology, 39, 347–363. https://doi.org/10.1007/ s10464-007-9116-1 Coleman-Jensen, A., Rabbitt, M. P., Gregory, C. A., & Singh, A. (2018). Household Food Security in the United States in 2017 (Report No. ERR-256). Washington, DC: U.S. Department of Agriculture, Economic Research Service. Corkery, L. (2004). Community gardens as a platform for education for sustainability. Australian Journal of Environmental Education, 20, 69–75. https://doi.org/10.1017/S0814062600002317 Dynes, M., Stephenson, R., Rubardt, M., & Bartel, D. (2012). The influence of perceptions of community norms on current contraceptive use among men and women in Ethiopia and Kenya. Health and Place, 18, 766–773. https://doi.org/ 10.1016/j.healthplace.2012.04.006 Elder, R. W., Lawrence, B., Ferguson, A. et al. (2010). The effectiveness of tax policy interventions for reducing excessive alcohol consumption and related harms. American Journal of Preventive Medicine, 38, 217–229. https://doi.org/10.1016/ j.amepre.2009.11.005 Fisher, E. B., Jr. (1995). The results of the COMMIT trial. Community Intervention Trial for Smoking Cessation. American Journal of Public Health, 85, 159–160. https://doi.org/10.2105/AJPH.85.2.159 Fuller, C. M., Galea, S., Caceres, W. Blaney, S., Sisco, S., & Vlahov, D. (2007). Multilevel communitybased intervention to increase access to sterile syringes among injection drug users through pharmacy sales in New York City. American Journal of Public Health, 97, 117–124. https://doi .org/10.2105/AJPH.2005.069591 Glasgow, R. E., Cummings, K. M., & Hyland, A. (1997). Relationship of worksite smoking policy to changes in employee tobacco use: Findings from COMMIT. Community Intervention Trial for Smoking Cessation. Tobacco Control, 6, S44. https://doi.org/10.1136/tc.6.suppl_2.S44 Gong, F., Baron, S., Ayala, L., Stock, L., McDevitt, S., & Heaney, C. (2009). The role for communitybased participatory research in formulating policy

initiatives: Promoting safety and health for inhome care workers and their consumers. American Journal of Public Health, 99, S531–S538. https:// doi.org/10.2105/AJPH.2008.152405 Greder, K. A., Garasky, S., & Klein, S. (2007). Research to action: A campus-community partnership to address health issues of the food insecure. Journal of Extension, 45, 1–12. https:// doi.org/hdfs_pubs/7/ Griffith, D. M., Mason, M., Yonas, M., Eng, E., Jeffries, V., Plihcik, S., & Parks, B. (2007). Dismantling institutional racism: Theory and action. American Journal of Community Psychology, 39, 381–392. https://doi.org/ 10.1007/s10464-007-9117-0 Hamelin, A. M., Habicht, J. P., & Beaudry, M. (1999). Food insecurity: Consequences for the household and broader social implications. The Journal of Nutrition, 129, 525S–528S. https://doi.org/ 10.1093/jn/129.2.525S Handforth, B., Hennink, M., & Schwartz, M. B. (2013). A qualitative study of nutrition- based initiatives at selected food banks in the feeding America network. Journal of the Academy of Nutrition and Dietetics, 113, 411–415. https://doi.org/10.1016/ j.jand.2012.11.001 Hawe, P., Shiell, A., & Riley, T. (2009). Theorizing interventions as events in systems. American Journal of Community Psychology, 43, 267–276. https://doi.org/10.1007/s10464-009-9229-9 Head, B., & Alford, J. (2015). Wicked problems: Implications for policy and management. Administration and Society, 47, 711–739. https:// doi.org/10.1177/0095399713481601 Hirsch, G., Levine, R., & Miller, R. (2007). Using systems dynamics modeling to understand the impact of social change initiatives. American Journal of Community Psychology, 39, 239–253. https://doi.org/10.1007/s10464-007-9114-3 Jason, L. A., Olson, B. D., Ferrari, J. R., & Lo Sasso, A. T. (2006). Communal housing settings enhance substance abuse recovery. American Journal of Public Health, 96, 1727–1729. https://doi.org/ 10.2105/AJPH.2005.070839 Kaplan, R., & Kaplan, S. (1989). The Experience of Nature: A Psychological Perspective. New York: Cambridge University Press.

Behavior Change in Community Contexts

Kelly, J. G. (1966). Ecological constraints on mental health services. American Psychologist, 21, 535. https://doi.org/10.1037/h0023598 Kelly, J. G. (1970). Toward an ecological conception of preventive interventions. In D. Adelson & B. Kalis (Eds.), Community Psychology and Mental Health (pp. 126–145). Scranton, PA: Chandler. Kelly, J. G. (Ed.). (2006). Becoming Ecological: An Expedition into Community Psychology. Oxford: Oxford University Press. Kelly, J. A., & Kalichman, S. C. (2002). Behavioral research in HIV/AIDS primary and secondary prevention: Recent advances and future directions. Journal of Consulting and Clinical Psychology, 70, 626. https://doi.org/10.1037/ 0022-006X.70.3.626 Kumar, S., Quinn, S. C., Kim, K. H., Musa, D., Hilyard, K. M., & Freimuth, V. S. (2012). The social ecological model as a framework for determinants of 2009 H1N1 influenza vaccine uptake in the United States. Health Education and Behavior, 39, 229–243. https://doi.org/10.2105/ AJPH.2011.300307 Lewin, K. (1951). Field Theory in Social Science: Selected Theoretical Papers (Ed. Dorwin Cartwright). Oxford: Harpers. Linos, N., Slopen, N., Subramanian, S. V., Berkman, L., & Kawachi, I. (2013). Influence of community social norms on spousal violence: A populationbased multilevel study of Nigerian women. American Journal of Public Health, 103, 148– 155. https://doi.org/10.2105/AJPH.2012.300829 Lipperman-Kreda, S., Grube, J. W., & Paschall, M. J. (2010). Community norms, enforcement of minimum legal drinking age laws, personal beliefs and underage drinking: An explanatory model. Journal of Community Health, 35, 249–257. https://doi.org/10.1007/s10900-010-9229-6 Martin, K., Shuckerow, M., O’Rourke, C., & Schmitz, A. (2012). Changing the conversation about hunger: The process of developing Freshplace. Progress in Community Health Partnerships: Research, Education, and Action, 6, 429–434. https://doi.org/10.1353/cpr.2012.0056 Martin, K. S., Colantonio, A. G., Picho, K., & Boyle, K. E. (2016). Self-efficacy is associated with increased food security in novel food pantry

413

program. SSM- Population Health, 2, 62–67. https://doi.org/10.1016/j.ssmph.2016.01.005 Martin, K. S., Wu, R., Wolff, M., Colantonio, A. G., & Grady, J. (2013). A novel food pantry program: Food security, self-sufficiency, and diet-quality outcomes. American Journal of Preventive Medicine, 45, 569–575. https://doi.org/10.1016/ j.amepre.2013.06.012 Martinez, M. J. (2010). Power at the Roots: Gentrification, Community Gardens, and the Puerto Ricans of the Lower East Side. New York: Lexington Books. Matson-Kaufman, D. M., Brownstein, J. N., Neiner, J. A., & Greaney, M. L. (2005). A site-specific literature review of policy and environmental interventions that promote physical activity and nutrition for cardiovascular health: What works? American Journal of Health Promotion, 19, 167– 193. https://doi.org/10.4278/0890-1171-19.3.167 McLeroy, K. R., Bibeau, D., Steckler, A., & Glanz, K. (1988). An ecological perspective on health promotion programs. Health Education and Behavior, 15, 351–377. https://doi.org/10.1177/ 109019818801500401 Milligan, C., Gatrell, A., & Bingley, A. (2004). “Cultivating health”: Therapeutic landscapes and older people in northern England. Social Science and Medicine, 58, 1781–1793. https://doi.org/ 10.1016/S0277-9536(03)00397-6 Moos, R. H. (1974). Conceptualizations of human environments. American Psychologist, 38, 652– 665. https://doi.org/10.1037/h0035722 Okvat, H. A., & Zautra, A. J. (2011). Community gardening: A parsimonious path to individual, community, and environmental resilience. American Journal of Community Psychology, 47, 374–387. https://doi.org/10.1007/s10464-0109404-z Park, R. (1952). Human Communities: The City and Human Ecology. Glencoe, IL: Free Press. Perkins, H. W. (2002). Social norms and the prevention of alcohol misuse in collegiate contexts. Journal of Studies on Alcohol, 14, S164–S172. https://doi .org/10.15288/jsas.2002.s14.164 Perkins, H. W., & Berkowitz, A. D. (1986). Perceiving the community norms of alcohol use among students: Some research implications for campus

414

EDISON TRICKETT AND SUSAN PATERSON

alcohol education programming. International Journal of the Addictions, 21, 961–976. https:// doi.org/10.3109/10826088609077249 Pierce, W. D., Trickett, E. J., & Moos, R. H. (1972). Changing ward atmosphere through staff discussion of the perceived ward environment. Archives of General Psychiatry, 26, 35–41. https:// doi.org/10.1001/archpsyc.1972.01750190037008 Quandt, S. A., Arcury, T. A., Early, J., Tapia, J., & Davis, J. D. (2004). Household food security among migrant and seasonal Latino farmworkers in North Carolina. Public Health Reports, 119, 568–576. https://doi.org/10.1016/ j.phr.2004.09.006 Remley, D. T., Kaiser, M. L., & Osso, T. (2013). A case study of promoting nutrition and long-term food security through choice pantry development. Journal of Hunger and Environmental Nutrition, 8, 324–336. https://doi.org/10.1080/ 19320248.2013.819475 Sallis, J. F., Cervero, R. B., Ascher, W., Henderson, K. A., Kraft, M. K., & Kerr, J. (2006). An ecological approach to creating active living communities. Annual Review of Public Health, 27, 297–322. https://doi.org/10.1146/annurev .publhealth.27.021405.102100 Sanford, N. (1970). Whatever happened to action research? Journal of Social Issues, 26, 3–23. https://doi.org/10.1111/j.1540-4560.1970 .tb01740.x Sarason, S. (1972). The Creation of Settings and the Future Societies. Boston, MA: Brookline Books. Schensul, S. L., Singh, R., Schensul, J. J., Verma, R. K., Burleson, J. A., & Nastasi, B. K. (2015). Community gender norms change as a part of a multilevel approach to sexual health among married women in Mumbai, India. American Journal of Community Psychology, 56, 57–68. https://doi.org/10.1007/s10464-015-9731-1 Schmid, T. L., Pratt, M., & Howze, E. (1995). Policy as intervention: Environmental and policy approaches to the prevention of cardiovascular disease. American Journal of Public Health, 85, 1207–1211. https://doi.org/10.2105/ AJPH.85.9.1207 Seligman, H. K., Lyles, C., Marshall, M. B. et al. (2015). A pilot food bank intervention featuring

diabetes-appropriate food improved glycemic control among clients in three states. Health Affairs, 34, 1956–1963. https://doi.org/10.1377/ hlthaff.2015.0641 Sheeran, P., Abraham, C., & Orbell, S. (1999). Psychosocial correlates of heterosexual condom use: A meta-analysis. Psychological Bulletin, 125, 90– 132. https://doi.org/10.1037/0033-2909.125.1.90 Short, D. J. (2002). Newcomer programs: An educational alternative for secondary immigrant students. Education and Urban Society, 34, 173–198. https:// doi.org/10.1177/0013124502034002004 Simons-Morton, D. G., Simons-Morton, B. G., Parcel, G. S., & Bunker, J. F. (1988). Influencing personal and environmental conditions for community health: A multilevel intervention model. Family and Community Health: The Journal of Promotion and Maintenance, 11, 25–35. https://doi.org/ 10.1097/00003727-198808000-00006 Stokols, D., Allen, J., & Bellingham, R. (1996). The social ecology of health promotion: Implications for research and practice. American Journal of Health Promotion, 10, 247–251. https://doi.org/ 10.4278/0890-1171-10.4.247 Story, M., Kaphingst, K. M., Robinson-O’Brien, R., & Glanz, K. (2008). Creating healthy food and eating environments: Policy and environmental approaches. Annual Review of Public Health, 29, 253–272. https://doi.org/10.1146/annurev .publihealth.29.020907.090926 Toro, P. A., Rappaport, J., & Seidman, E. (1987). Social climate comparison of mutual help and psychotherapy groups. Journal of Consulting and Clinical Psychology, 55, 430–431. https://doi.org/ 10.1037/0022-006X.55.3.430 Trickett, E., Espino, S. R., & Hawe, P. (2011). How are community interventions conceptualized and conducted? An analysis of published accounts. Journal of Community Psychology. 39, 576–591. https://doi.org/10.1002/jcop.20455 Trickett, E. J., Kelly, J. G., & Todd, D. M. (1972). The social environment of the high school: Guidelines for individual change and organizational development. In S. Golann & C. Eisdorfer (Eds.), Handbook of Community Mental Health (pp. 331– 406). New York: Appleton-Century-Crofts.

Behavior Change in Community Contexts

Wallerstein, N., Duran, B., Oetzel, J., & Minkler, M. (Eds.). (2018). Community-Based Participatory Research for Health. San Francisco: John Wiley & Sons. Walter, P. (2013). Theorising community gardens as pedagogical sites in the food movement.

415

Environmental Education Research, 19, 521–539. https://doi.org/10.1080/13504622.2012.709824 WHO (World Health Organization). (2009). Changing Cultural and Social Norms that Support Violence. http://apps.who.int/iris/bitstream/handle/10665/ 44147/?sequence=1

29 Changing Behavior in the Digital Age David J. Kavanagh

Practical Summary The world is increasingly connected to the internet, and the advent of smartphones and tablets means that most people can now have a therapist or coach “in their pocket.” This offers exciting opportunities to get support for important behavioral changes whenever they want. Not only can people gain information but they can anonymously access the same help they would get from a coach or therapist, without seeing them. Service users can also work with a practitioner or peer group on their device or use digital tools between face-toface sessions. There is now substantial evidence that some digital interventions can have strong effects, although their pace of development is far outstripping research, and many have no supporting evidence. This proliferation also makes it hard to know what to choose, although sites that give advice are emerging. Risks to privacy and data security remain challenges and quality varies, but digital support for behavior change is rapidly gaining acceptance.

29.1 Introduction In the past, behavior change has usually been delivered in face-to-face training or therapy sessions, although there are precedents for delivery of segments or whole interventions by phone, mail, audios, or videos (e.g., Kavanagh et al., 1999). However, the digital revolution that has occurred in our everyday lives is now increasingly impacting the way behavior change is supported. This chapter unpacks this phenomenon and evaluates its potential. It focuses on the rationale for using digital resources to foster and support behavior change, their acceptance and use, evidence on self-guided and coached interventions using digital resources, and how their quality can be assessed. The chapter concludes

with implications for the use of digital resources for research and promotion of behavior change.

29.1.1 Definitions In this chapter, digital resources, tools, programs, or services for behavior change include any that use a digital device, including computers, mobile phones, tablets, wearable devices, web-linked appliances and robots, although work on behavior change using the latter devices is still in its infancy. In the health domain, an encompassing term is eHealth, although this sometimes also includes crisis or counseling services using audio phones. Types of

https://doi.org/10.1017/9781108677318.029

Changing Behavior in the Digital Age

417

Sidebar 29.1 Types of digital resources and tools for behavior change • • • •

Informational resources (e.g., on websites). Self-guided programs or mobile applications (“apps”) to support self-management. Coached programs or “apps,” where a trainer assists users to use the resource. Digital interventions by trainers or peers, where sessions are delivered online (e.g., by audio/video links, chatlines, emails, or text messages). • Digital tools to guide face-to-face training or therapy (e.g., where a trainer obtains specialist support during sessions).

resources and interventions are summarized in Sidebar 29.1. Websites or apps may offer information about problems or issues or about the nature and benefits of potential goals but may also include specific advice on what the person could do. So, a website may offer information and advice on quitting smoking or addressing depression.1 Other digital approaches simulate all or part of an intervention, where users’ input creates individualized content such as action plans (see also Chapters 6, 7, and 39, this volume). Some provide a scaffold to support self-management, giving no additional human assistance. These may focus on assessment or selfmonitoring (see Chapter 37, this volume), provide reminders to initiate behaviors (see Chapter 14, this volume), or offer personalized training or treatment. Examples include the Fitz app, described in Sidebar 29.2 and, in mental health, the mood gym web program for depression.2 Other interventions offer digital, audio/video, or face-to-face coaching or reminders by a trainer or peer, which focus on optimizing use of the digital program or tool to change behavior. In some cases, digital tools are core to the intervention.3 In others, they are adjunctive to other interventions (e.g., tracking progress). The coaching or reminders may be by audio or video, web chat, emails, or text messages, and responses may be immediate or delayed. Alternatively, digital interventions may comprise human interventions that are fully conducted in digital formats (e.g., online training, counseling, or peer support) or are face-to-face but guided by

digital tools (e.g., the Indigenous well-being app Stay Strong).

29.1.2 Why Should Digital Tools Be Used in Behavior Change Interventions? Currently, face-to-face services that promote changes in some important domains do not have sufficient population access. For example, only 43 percent of adult US residents and 46 percent of adult Australians with a twelve-month mental disorder report receiving treatment for it (SAMHSA, 2018; Whiteford et al., 2014). The problem is due to both limited supply of appropriate services (e.g., insufficient trained practitioners; location and opening hours) and insufficient uptake (e.g., because of cost or perceived stigma; Brown et al., 2016; Kerridge et al., 2017). Access is particularly low in males, young people, nonurban residents, groups at socioeconomic disadvantage, Indigenous people, and ethnic minorities (Brown et al., 2016; SAMHSA, 2018). When people do receive an intervention, its delivery in routine services often has low fidelity to procedures with established efficacy (Kavanagh et al., 1993; Randall, Wakefield, & 1

2 3

See, for example, https://smokefree.gov (on quitting smoking) and https://bluepages.anu.edu.au (on addressing depression). See https://moodgym.com.au. For example, the digital courses at https://mindspot.org .au.

418

DAVID J. KAVANAGH

Sidebar 29.2 Example of a standalone digital behavior change tool: The Fitz app

Fitz is a mobile phone app that applies an imagery-based motivational intervention to any behavior change (Functional Imagery Training, FIT; Kavanagh et al., 2014). FIT uses the spirit of motivational interviewing (explained further in Chapter 45, this volume) but encourages participants to use mental imagery at each step (see also Chapter 33, this volume). If they are committed to change, they are shown how to practice imagery about their plans and expected outcomes to strengthen motivation whenever it is needed. Research has demonstrated that FIT for weight reduction resulted in 5.8 kg greater weight loss than motivational interviewing (Solbrig et al., 2018). Fitz uses a chatbot to deliver a simplified version of FIT (for a screenshot of the app, see Figure 29.1). Users record audios throughout the conversation, which act as summaries and later reminders. Other audios assist them to create motivational imagery, and they can add photos to help further. They can set reminders to practice imagery and take action, record achievements, and see their progress. Further outcomes using Fitz are currently being tested.

Figure 29.1 Screenshot of the Fitz app

Changing Behavior in the Digital Age

Richards, 2012). Establishment of sustained behavior change also requires its recall and implementation of the action in the natural environment and that also is difficult to attain (Richard, Glaser, & Lussier, 2017). Digital delivery can address some of these challenges. Digital devices are potentially accessible almost anywhere – especially when they are carried or worn and the web can be accessed or an “app” is used offline. If they are self-guided and used anonymously, they reduce perceived risk of stigma about help-seeking and fulfill some users’ preferences for self-reliance (Gulliver, Griffiths, & Christensen, 2010). Digital resources or tools are typically free of charge or impose low user costs and can have high cost-effectiveness, especially when selfguided and user numbers are substantial (Lintvedt et al., 2013). While devices and web access incur costs, these issues can potentially be addressed for low-income users by direct subsidy (Marasinghe et al., 2012) or free web access hubs. In contrast to face-to-face interventions, digital ones have the same potential content every time, as long as the device, software, and web access are reliable. Digital delivery also allows users to review content, obtain reminders to engage in behaviors, or access support whenever and wherever it is needed. These advantages are accentuated by a rapid trend for the world to be digitally connected. In January 2018, there were an estimated 4 billion internet users (53 percent of the world population; We Are Social & Hootsuite, 2018). In twenty-one highly developed countries, more than 80 percent of the population now has internet access, with many nations, including UAE, Sweden, the Netherlands, and the UK, at or near market saturation (We are Social & Hootsuite, 2018). Differences between national rates are progressively reducing – for example, rapid growth is occurring in Egypt, Kenya, and Vietnam, with increases in internet availability over 2017–2018 of 41 percent, 35 percent, and 28 percent, respectively (We Are Social & Hootsuite, 2018). Age differences within

419

countries are also reducing. Internet access is nearly universal in younger US groups, with 98 percent of those aged eighteen to twenty-nine years having access in 2018. However, access by older groups is also high, with 87 percent access in the fifty-to-sixty-four-year-old age group and 66 percent in Americans aged sixty-five and older, and rates across age groups are converging (Pew Research Center, 2018a). As a consequence, demographic inequities of access are rapidly disappearing, although web coverage, slow speed, and download costs remain issues for some users. Digital connections are also becoming more portable. Worldwide, smartphones are now the most common digital devices (52 percent vs. 43 percent for laptops and desktops; We Are Social & Hootsuite, 2018). As yet, worldwide uptake of other devices is small by comparison (e.g., tablets, 4 percent; We Are Social & Hootsuite, 2018), although market penetration for tablets has been rising rapidly and is already very high in some countries (e.g., 53 percent in the United States; 43 percent and 27 percent, respectively, for large and small tablets in Australia; Deloitte, 2018; Pew Research Center, 2018b).

29.2 Acceptability and Engagement Digital tools and resources for health have high acceptance by respondents to online surveys. For example, 58 percent of a 2015 sample of US mobile phone users had downloaded a health app – most commonly to promote physical activity or nutrition (Krebs & Duncan, 2015). However, even among young users, the internet is primarily used to obtain information or community connection rather than interventions (Park & Kwon, 2018). In older users who are dealing with a chronic disorder such as diabetes that requires complex behavioral regimens, digital interventions may face particular acceptance challenges (Parham et al., 2018; Schleibe et al., 2015) and may be seen as an additional task of

420

DAVID J. KAVANAGH

insufficient benefit, unless the tool is particularly intuitive and entertaining and users quickly see benefits from its use. Furthermore, most users tend to be women, and users are skewed toward groups with higher education (Titov et al., 2018). Engagement of men and users with lower education remains a challenge. Digital providers allow users to make informed decisions about whether they need to make behavior changes and about potential sources of support. For example, in Australia, of the 458,921 individuals who visited the MindSpot mental health website over its first thirty months of operation, only 25,469 completed a mental health assessment (Titov et al., 2017). While 90 percent of these were above screening cutoffs for anxiety or depression, only 37 percent accessed the site to seek treatment and just 24 percent started a digital course; the remainder received information about other treatment options. However, acceptance of digital support for behavior change is growing. Internationally, digital interventions have become a substantial component of mental health services, reaching many who have never sought treatment previously (Titov et al., 2018). Fidelity of digital interventions to evidencebased practice does not guarantee every user will receive an effective intervention. Most users typically do not complete self-guided interventions with multiple components (Batterham et al., 2008). Ongoing use of self-guided interventions and the implementation of related actions require significant self-control (see Chapter 11, this volume), and users say they miss receiving the reminders, encouragement, and rewards a coach would give (Darvell, Kavanagh, & Connolly, 2015). Some users may believe they have already obtained sufficient benefit from the segments they completed (Darvell et al., 2015), but greater average changes are obtained by users who complete more of an evidence-based intervention (Zeng et al., 2016). As shown in Sidebar 29.4, retention rates increase sharply when brief online or phone sessions are provided alongside the digital

intervention to review progress and show participants how to optimally use the digital resources. Researchers have also examined other factors associated with longer use of digital interventions. Baumel and Yom-Tov (2018) found that “therapeutic persuasiveness” (inclusion of persuasive design and behavior change principles) was the strongest predictor of the use of health apps, with aspects such as visual design having only weak effects. Similarly, intentions to continue using an e-learning tool for university students were most strongly associated with user satisfaction, followed by its perceived usefulness (Cho, Cheng, & Lai, 2009). Functionality and ease of use were only indirectly associated with intentions, through their influence on perceived usefulness. “Gamification” (e.g., inclusion of points, leaderboards, or unlocking more challenging or entertaining features of digital interventions) has increasingly received attention. Gamification can increase interest and cognitive involvement in a well-being tool (Kelders, Sommers-Spijkerman, & Goldberg, 2018), and should influence usage and outcomes. In Allam et al. (2015), patients with rheumatoid arthritis were randomly allocated to a control group or to programs that provided information, social support, gamification (points for engagement in features such as quizzes or posts), or support plus gamification, with the aims of increasing physical activity and reducing health care utilization and overuse of prescribed medication. Gamification increased website use, and participants who also received social support increased their physical activity and had less health care utilization. Similarly, embedding normative feedback about alcohol consumption within a social game increased the impact of the feedback on alcohol consumption two weeks later (Boyle et al., 2017). In principle, participatory codesign and development by potential users should increase a tool’s attractiveness, engagement, and efficacy (Sidebar 29.3), although evidence that it has these effects is limited.

Changing Behavior in the Digital Age

421

Sidebar 29.3 Participatory codesign and development

Genuine codesign with potential users of digital resources capitalizes on their experience in attempting behavior change, identifying content and functionality that may have addressed their perceived needs, together with designs and other features that would entice them to use the tool (Clemenson et al., 2017; see also Chapter 25, this volume). This expansion of the creative team also potentially improves the tool’s innovation. Participatory codesign begins at the start of the development process (see Chapter 21, this volume), rather than just comprising feedback on developed tools, and has multiple phases of input as the tool’s development proceeds. Expert content and design development occurs in parallel, and each phase involves mutual influence and negotiation between users and experts. Participatory codesign does typically increase the time and cost of producing digital tools, and, even with the inclusion of different user informants at different stages, their input may not be representative of the wider user community. While codesign does not guarantee greater use of a tool, its advantages and its consistency with increasing user involvement in services are leading to its increasing acceptance as best practice.

Usage and engagement with a digital resource is itself a behavior change target, and preceding its use by a brief intervention with that target should in principle increase it. Making the required usage easier (e.g., by minimizing intervention length and other demands) is also likely to improve use and increase benefits. Data entry time was the most common reason for discontinuing a health app use in Krebs and Duncan (2015), so an obvious solution involves automated data acquisition wherever possible. Common examples are recording of step counts, heart rate, or sleep duration and quality on fitness bands, but automated capture and analysis of other variables (e.g., location, facial affect) is progressively becoming available. Best development practice also minimizes required reading age and amount of text, although standard software typically now allows text to be read aloud.

29.3 Do Digital Interventions Work? There is now substantial evidence supporting the efficacy of digital interventions in changing

behavior. Research on web programs is currently more substantial than for apps and the extent of evidence varies widely across behavior domains, but these differences are rapidly diminishing. Systematic reviews of digital health interventions have documented positive effects for a wide range of targets, including increased physical activity or improved diet (Schoeppe et al., 2016), glycemic control in Type 2 diabetes (Kebede et al., 2018), HIV testing (Conserve et al., 2017), workplace health and well-being (Howarth et al., 2018), and insomnia (Zachariae et al., 2016). Mental health targets that have shown positive effects include depression (Sztein et al., 2018), anxiety disorders (Domhardt et al., 2018), posttraumatic stress disorder (Sijbrandij, Kunovsk, & Cuijpers, 2016), stress (Heber et al., 2017), and substance use disorders (Nesvåg & McKay, 2018). Small but significant effects on prevention of anxiety and depression are also seen (Sander, Rausch, & Baumeister, 2016). In fact, trials comparing coached interventions with face-to-face treatment for mental health and somatic conditions currently

422

DAVID J. KAVANAGH

Sidebar 29.4 Sample digital behavior change intervention study

Kiluk et al. (2018) randomly allocated 191 people with substance use disorders to outpatient treatment as usual (TAU), clinician-delivered, manualized cognitive behavioral therapy (CBT), or CBT4CBT, a therapist-supported web program. In CBT4CBT, participants worked on seven core CBT skills, which were described and illustrated by animations and videos, and included quizzes, other interactive exercises, and recommended home practice. Brief weekly calls, each lasting about ten minutes, reviewed their status and program use. CBT4CBT had the best retention in treatment (76 percent completing all modules vs. 54 percent in TAU and 39 percent in clinician CBT) and had significantly superior reductions in days of substance use per week over monthly assessments that concluded at six months after the end of treatment (nine months post-baseline). At treatment termination, 82 percent of CBT4CBT participants said they were very satisfied with their treatment. show no average difference between the two formats (Carlbring et al., 2018), and, as shown in Sidebar 29.4, there is evidence of superior effects from digital formats in some trials. Remote coaching of digital courses for anxiety and depression is also highly effective in routine delivery (Titov et al., 2017), allowing this format to be available to large numbers of users across an entire country. Digital resources and tools have also been used to elicit behavior change in non-health domains. Training for specific sporting skills (e.g., vision training; Appelbaum & Erickson, 2018) and webbased skills development for health practitioners (Jackson et al., 2018) have each demonstrated significant effects across studies, and map-based visualizations have been used to encourage adaptations to climate change (Neset et al., 2016). Effect sizes of therapist-coached interventions are typically stronger than for self-guided interventions (e.g., in health interventions, Domhardt et al., 2018; Heber et al., 2017). Many users feel they need a coach (Darvell et al., 2015), although head-to-head comparisons do not always show greater effects (e.g. Dear et al., 2018). Research is now examining whether avatars or social robots may act as coaches or as sole intervention agents. While the current evidence on social

robots is slim and limited in range (Robinson, Cottier, & Kavanagh, 2019), it suggests that robots may be sufficiently humanoid to support behavior change and potentially may augment existing service capabilities. As important as randomized controlled trials are in evaluating whether a digital tool changes behavior, there are significant limitations to an exclusive reliance on them (see Sidebar 29.5). Furthermore, it is important to supplement evidence from these trials with evidence from the use of tools by people who have not volunteered for a trial and with ongoing evidence from digital services, including rates of exacerbation (where the existing behavior is problematic). Current evidence is that digital mental health services with robust governance and risk management can demonstrate similar effect sizes to randomized controlled trials, with low rates of deterioration (Titov et al., 2018). These results offer encouragement that similar results are also likely to be seen in other behavior domains.

29.4 Assessing the Quality of Digital Interventions A significant issue for both users and practitioners in developing digital behavior change

Changing Behavior in the Digital Age

423

Sidebar 29.5 Challenges in assessing the efficacy of digital interventions

The “gold standard” of outcome evaluation in clinical sciences involves replicated results from randomized controlled trials with strong methodology and extended follow-up (see also Chapter 22, this volume). However, those trials require large budgets and long completion periods. By then, a digital resource requires substantial revision and often is no longer available. The problem is confounded by many developers lacking funding for research or not valuing it sufficiently. This creates a vacuum for users seeking advice about efficacy and a challenge for agencies wanting to provide it. Suggested partial solutions have included the use of brief trials, ones with large factorial designs, and sequential, multiple assignment randomized controlled trials (SMART) (Kaker, Gustafson, & Shah, 2014). However, each of these has limitations – for example, brief trials cannot demonstrate maintained effects and, while routine follow-up of users can supplement these, it is difficult to obtain high long-term follow-up rates with users of self-guided programs and apps. Seeing some digital resources as applying established treatments in a new format rather than as new ones may help. If motivational interviewing has strong support for engaging people in change, there is no need for new trials from each service that delivers it; instead, their evaluation focuses on fidelity and routine outcome assessment. When experts attest to digital tools having high fidelity to an established intervention or clinical guidelines, that advice can be provided to users, alongside any research specifically testing them.

interventions is how they choose which digital tools and resources to use. For example, 318,000 health apps were available in 2017, with 200 being added every day and the pace increasing over time (IQVIA Institute for Human Data Science, 2017). Search functions in app stores are currently suboptimal, most apps do not have evidence on their efficacy (Sucala et al., 2017), many do not cite the currency of information, and the only advice on quality often is a star rating on the app store that is subject to manipulation. While some digital interventions are regulated as medical devices, others currently have no certification. Privacy, data security, safety, and protection against malware for apps and web programs that deliver behavioral interventions are not assured. To address these issues, sites by respected providers that list resources and programs and

provide search functions are emerging.4 However, listing requires a judgment of which ones should be included, implying a form of certification – as does their use in services.5 Instruments to assess the quality of digital tools are emerging and variants of these may assist with aspects of quality assessment. Examples include the Mobile Apps Rating Scale, which has both an expert (Stoyanov et al., 2015) and a user version (Stoyanov et al., 2016) and offers ratings of both digital quality and content. Standards applying to digital services involving human providers would need to incorporate

4

5

See, for example, https://apps.beta.nhs.uk, https://head tohealth.gov.au. For example, www.england.nhs.uk/mental-health/ adults/iapt/digital-therapy-selection.

424

DAVID J. KAVANAGH

additional criteria (e.g., relating to governance, training, supervision, and quality assurance). While current efforts internationally to develop certification standards will provide additional assurance and advice to users, there is a risk that some will delay or impose additional costs to developers, discouraging them from seeking certification unless it provides significant benefits (e.g., increased revenue). As a result, it is likely that most digital resources and interventions will continue to lie outside any certification system. In the absence of laws prohibiting uncertified digital resources and interventions (which are extremely unlikely), they will continue to be available, and developers will need to see a cost benefit in voluntarily seeking certification. The onus will continue to be on users to assess whether to download and use a digital resource. That is not to say that digital resources will be entirely unregulated: If they are held to be medical devices, they will be subject to regulation and will need to meet other legislative requirements (e.g., consumer protection and privacy – as in recent EU regulations).6 Furthermore, their use in some services (e.g., health and education) will be subject to regulation of professional practice.

29.5 Implications for Research and Practice Support for the efficacy of digital interventions is already strong, although some behavioral domains or delivery modes (e.g., social robots) remain under-researched, and relatively few trials unpack the effects of specific intervention elements. Other important questions that continue to demand attention include the marketing of digital interventions, eliciting and maintaining their use, and increasing their impact on everyday behaviors. Emerging questions include the effects of improved social communication, machine 6

See https://eurlex.europa.eu/LexUriServ/LexUriServ. do?uri=CONSLEG:2002L0058:20091219:EN:html.

learning, and big data analysis on the individual tailoring of interventions and on the impact of digital coaching. However, it is becoming difficult for researchers to make a case for sufficient novelty to obtain grant funding simply to evaluate the efficacy of a new tool or service. Many of the early, evidence-based digital interventions were developed by university-based researchers, who relied on time-limited grants and tenders and have often struggled to maintain their resources over time. As digital interventions for behavior change have become more accepted and valued, they have gained commercial value and are now attracting interest from major companies (e.g., health insurers, internet providers, pharmaceutical companies). This growing interest will result in some tools having increased technical and functional sophistication, as well as improved security and maintenance, but may also make it harder for small enterprises to meet the demands of both consumers and regulators. Digital interventions may sometimes be in competition with other services, but, increasingly, they are transforming the way practice is conducted. The use of SMS reminders to increase attendance at appointments is already a very familiar practice. Digital tools can also be used by educators, coaches, or health practitioners to increase the impact of their work, by delivering education, monitoring behavior between sessions, or giving support on issues they lack skills, confidence, or time to address. Practitioners can become coaches of self-guided interventions, encouraging users to maintain engagement and maximize the impact of the tool. Moderated peer forums can offer a safe context where people can obtain social support and tips from others like them, supplementing individual assistance from the practitioner. While there remains suspicion or reluctance by some practitioners to adopt digitally informed practice (Hennemann, Beutel, & Zwerenz, 2017), there are now examples of largescale interventions to increase the integration of digital tools into routine practice. An example is

Changing Behavior in the Digital Age

eMental Health in Practice,7 a nationwide program in Australia to increase the use of digital mental health in primary care through marketing, training, mentoring, and organizational change in services. In addition, digital interventions are increasingly being integrated into national policies (e.g., Council of Australian Governments, 2017), which will accelerate their integration into funding models and practice.

29.6 Summary and Conclusion Digital tools and interventions for behavior change are rapidly increasing in number and their community acceptance is growing, although it remains stronger for information than for interventions and usage by some groups (e.g., men, those with less education, older users) remains a challenge. Engagement and maintained use are often suboptimal in self-guided interventions and represent important behavior change targets themselves. Some interventions have gathered a strong evidence base. However, the pace of tool development far outstrips the pace of efficacy research, and it often takes so long to obtain research support, that the tool has lost currency. A recent focus has been on more agile ways to obtain this support. The impact of digital tools may increase when coaching is provided, but trials that test effects of coaching do not always show superior effects and coaching imposes additional costs that reduce the intervention’s capacity for populationwide impact. There is currently an increased emphasis on assessment of the quality of digital tools and resources and on their regulation. Increasingly, prospective users can access expert online advice that can help them understand their behavioral issues, search for appropriate tools, and obtain advice on their effects and safety of the tools they do select. This emphasis on quality assessment and assistance in tool selection, together with a rapidly rising community use of digital

425

tools, will increase their integration into routine practice, as will the progressive inclusion of digitally informed practice in professional training. Full integration of digital resources, tools and interventions into policy, practice and recurrent funding has yet to occur but rapid progress toward it appears inevitable. This trend will not replace other models of support but it will change the way the support is delivered and will likely increase its cost-effectiveness and the reach of behavior change interventions.

References Allam, A., Kostova, Z., Nakamoto, K., & Schultz, P. J. (2015). The effect of social support features and gamification on a Web-based intervention for rheumatoid arthritis patients: Randomized controlled trial. Journal of Medical Internet Research, 17, e14. https://doi.org/10.2196/ jmir.3510 Appelbaum, L. G., & Erickson, G. (2018). Sports vision training: A review of the state-of-the-art in digital training techniques. International Review of Sport and Exercise Psychology, 11, 160–189. https://doi.org/10.1080/1750984x.2016.1266376 Batterham, P. J., Neil, A. L., Bennett, K., Griffiths, K. M., & Christensen, H. (2008). Predictors of adherence among community users of a cognitive behavior therapy website. Patient Preference and Adherence, 2, 97–105. www.dovepress.com/ predictors-of-adherence-among-communityusers-of-a-cognitive-behavior-peer-reviewed-arti cle-PPA Baumel, A., & Yom-Tov, E. (2018). Predicting user adherence to behavioral eHealth interventions in the real world: Examining which aspects of intervention design matter most. Translational Behavioral Medicine, 8, 793–798. https://doi.org/ 10.1093/tbm/ibx037 Boyle, S. C., Earle, A. M., LaBrie, J. W., & Smith, D. J. (2017). PNF 2.0? Initial evidence that gamification can increase the efficacy of brief, web-based personalized normative feedback 7

See www.emhprac.org.au/.

426

DAVID J. KAVANAGH

alcohol interventions. Addictive Behaviors, 67, 8– 17. https://doi.org/10.1016/j.addbeh.2016.11.024 Brown, A., Rice, S. M., Rickwood, D. J., & Parker, A. G. (2016). Systematic review of barriers and facilitators to accessing and engaging with mental health care among at-risk young people. AsiaPacific Psychiatry: Official Journal of the Pacific Rim College of Psychiatrists, 8, 3–22. https://doi .org/10.1111/appy.12199 Carlbring, P., Andersson, G., Cuijpers, P., Riper, H., & Hedman-Lagerlöf, E. (2018). Internet-based vs. face-to-face cognitive behavior therapy for psychiatric and somatic disorders: An updated systematic review and meta-analysis. Cognitive Behaviour Therapy, 47, 1–18. https://doi.org/ 10.1080/16506073.2017.1401115 Cho, V., Cheng, T. C. E., & Lai, W. M. (2009). The role of perceived user-interface design in continued usage intention of self-paced learning tools. Computers and Education, 53, 216–227. https:// doi.org/10.1016/j.comedu.2009.01.014 Clemenson, J., Rothman, M. J., Smoith, A. C., Caffery, L. J., & Danbjorg, D. B. (2017). Participatory design methods in telemedicine research. Journal of Telemedicine and Telecare, 23, 780–785. https://doi.org/10.1177/1357633X16686747 Conserve, D. F., Jennings, L., Aguiar, C., Shin, G., Handler, L., & Maman, S. (2017). Systematic review of mobile health behavioral interventions to improve uptake of HIV testing for vulnerable and key populations. Journal of Telehealth and Telecare, 23, 347–359. https://doi.org/10.1177/ 1357633X16639186 Council of Australian Governments. (2017). The Fifth National Mental Health and Suicide Plan. Canberra: Council of Australian Governments. www.coaghealthcouncil.gov.au/Portals/0/Fifth% 20National%20Mental%20Health%20and% 20Suicide%20Prevention%20Plan.pdf Darvell, M., Kavanagh, D. J., & Connolly, J. (2015). A qualitative exploration of Internet-based treatment for comorbid depression and alcohol misuse. Internet Interventions, 2, 174–182. https://doi.org/ 10.1016/j.invent.2015.03.003 Dear, B. F., Fogliati, V. J., Fogliati, R. et al. (2018). Treating anxiety and depression in young adults: A randomised controlled trial comparing

clinician-guided versus self-guided Internetdelivered cognitive behavioural therapy. Australian and New Zealand Journal of Psychiatry, 52, 668–679. https://doi.org/10.1177/ 0004867417738055 Deloitte. (2018). Behaviour Unlimited: Mobil Consumer Survey 2018: The Australian Cut. www2.deloitte.com/au/mobile-consumer-survey Domhardt, M., Geßlein, H., von Rezori, R., & Baumeister, H. (2018). Internet- and mobile-based interventions for anxiety disorders: A metaanalytic review of intervention components. Depression and Anxiety, 36, 213–224. https://doi .org/10.1002/da.22860 Gulliver, A., Griffiths, K. M., & Christensen, H. (2010). Perceived barriers and facilitators to mental health help-seeking in young people: A systematic review. BMC Psychiatry, 10, 113. www.biomed central.com/1471-244X/10/113 Heber, E., Ebert, D. D., Lehr, D. et al. (2017). The benefit of web- and computer-based interventions for stress: A systematic review and meta-analysis. Journal of Medical Internet 19, e32. https://doi .org/10.2196/jmir.5774 Hennemann, S., Beutel, M. E., & Zwerenz, R. (2017). Ready for eHealth? Health professionals’ acceptance and adoption of eHealth interventions in inpatient routine care. Journal of Health Communication, 22, 274–284. https://doi.org/ 10.1080/10810730.2017.1284286 Howarth, A., Quesada, J., Silva, J., Judycki, S., & Mills, P. (2018). The impact of digital health interventions on health-related outcomes in the workplace: A systematic review. Digital Health, 4, 1–18. https://doi.org/10.1177/ 2055207618770861 IQVIA Institute for Human Data Science. (2017). The Growing Value of Digital Health: Evidence and Impact on Human Health and the Healthcare System. www.iqvia.com/-/media/iqvia/pdfs/ institute-reports/the-growing-value-of-digitalhealth.pdf?_=1547778440731 Jackson, C. B., Quetsch, L. B., Brabson, L. A., & Herschell, A. D. (2018). Web-based training methods for behavioral health providers: A systematic review. Administration and Policy in Mental Health and Mental Health Services

Changing Behavior in the Digital Age

Research, 45(4), 587–610. https://doi.org/ 10.1007/s10488-018-0847-0 Kaker, T. B., Gustafson, D. H., & Shah, D. (2014). How can research keep up with eHealth? Ten strategies for increasing the timeliness and usefulness of eHealth research. Journal of Medical Internet Research, 16, e36. www.jmir.org/2014/2/e36 Kavanagh, D. J., Andrade, J., May, J., & Connor, J. P. (2014). Motivational interventions may have greater sustained impact if they trained imagerybased self-management. Addiction, 109, 1062– 1063. https://doi.org/10.1111/add.12507 Kavanagh, D. J., Clark, D., Manicavasagar, V., Piatkowska, O. H., Rosen, A., & Tennant, C. (1993). Application of a cognitive-behavioural family intervention for schizophrenia in multidisciplinary settings: What can the matter be? Australian Psychologist, 28, 181–188. https://doi .org/10.1080/00050069308258899 Kavanagh, D. J., Sitharthan, T., Spilsbury, G., & Vignaendra, S. (1999). An evaluation of brief correspondence programs for problem drinkers. Behavior Therapy, 30, 641–656. https://doi.org/ 10.1016/S0005-7894(99)80030-6 Kebede, M. M., Zeeb, H., Peters, M., Heise, T. L., & Pischke, C. R. (2018). Effectiveness of digital interventions for improving glycemic control in persons with poorly controlled type 2 diabetes: A systematic review, meta-analysis, and metaregression analysis. Diabetes Technology and Therapeutics, 20, 767–782. https://doi.org/ 10.1089/dia.2018.0216 Kelders, S. M., Sommers-Spijkerman, M., & Goldberg, J. (2018). Investigating the direct impact of a gamified versus nongamified well-being intervention: An exploratory experiment. Journal of Medical Internet Research, 20, e247. www .jmir.org/2018/7/e247 Kerridge, B. T., Mauro, P. M., Chou, S. P. et al. (2017). Predictors of treatment utilization and barriers to treatment utilization among individuals with lifetime cannabis use disorder in the United States. Drug and Alcohol Dependence, 181, 223–228. https://doi.org/10.1016/j.drugalcdep.2017.09.032 Kiluk, B. D., Nich, C., Buck, M. B. et al. (2018). Randomized clinical trial of computerized and clinician-delivered CBT in comparison with

427

standard outpatient treatment for substance use disorders: Primary within-treatment and followup outcomes. American Journal of Psychiatry, 175, 853–863. https://doi.org/10.1176/appi .ajp.2018.17090978 Krebs, P., & Duncan, D. T. (2015). Health app use among US mobile phone owners: A national survey. JMIR mHealth and uHealth, 3, e101. https://doi.org/10.2196/mhealth.4924 Lintvedt, O. K., Griffiths, K. M., Eisemann, M., & Waterloo, K. (2013). Evaluating the translation process of an Internet-based self-help intervention for prevention of depression: A cost-effectiveness analysis. Journal of Medical Internet Research 15, e18. www.jmir.org/2013/1/e18 Marasinghe, R. B., Edirippulige, S., Kavanagh, D., Smith, A., & Jiffry, M. T. M. (2012). Effect of mobile phone-based psychotherapy in suicide prevention: A randomised controlled trial in Sri Lanka. Journal of Telemedicine and Telecare, 18, 151–155. https://doi.org/10.1258/jtt.2012. SFT107 Neset, T. S., Opach, T., Lion, P., Lilja, A., & Johansson, J. (2016). Map-based web tools supporting climate change adaptation. Professional Geographer, 68, 103–114. https://doi.org/ 10.1080/00330124.2015.1033670 Nesvåg, S., & McKay, J. R. (2018). Feasibility and effects of digital interventions to support people in recovery from substance use disorders: Systematic Review. Journal of Medical Internet Research, 20, e255. www.jmir.org/2018/8/e255 Parham, S., Kavanagh, D. J., Shimada, M., May, J., & Andrade, J. (2018). Qualitative analysis of feedback on Functional Imagery Training: A novel motivational intervention for type 2 diabetes. Psychology and Health, 33, 416–429. https://doi.org/10.1080/ 08870446.2017.1360493 Park, E., & Kwon, M. (2018). Health-related Internet use by children and adolescents: Systematic review. Journal of Medical Internet Research, 20, e120. www.jmir.org/2018/4/e120/ Pew Research Center. (2018a). Internet use by age: Surveys conducted 2000–2018. www.pewinter .net.org/chart/internet-use-by-age/

428

DAVID J. KAVANAGH

Pew Research Center. (2018b). Mobile fact sheet: Other devices. www.pewinternet.org/fact-sheet/ mobile/ Randall, G. E., Wakefield, P. A., & Richards, D. A. (2012). Fidelity to assertive community treatment program standards: A regional survey of adherence to standards. Community Mental Health Journal, 48, 138–149. https://doi.org/ 10.1007/s10597-010-9353-x Richard, C., Glaser, E., & Lussier, M. T. (2017). Communication and patient participation influencing patient recall of treatment discussions. International Journal of Public Participation in Health Care and Health Policy, 20, 760–770. https://doi.org/10.1111/hex.12515 Robinson, N. L., Cottier, T. V., & Kavanagh, D. J. (2019). Psychosocial health interventions by social robots: Systematic review of randomized controlled trials. Journal of Medical Internet Research, 21, e13203. https://doi.org/10.2196/ 13203 SAMHSA (Substance Abuse and Mental Health Services Administration). (2018). Key Substance Use and Mental Health Indicators in the United States: Results from the 2017 National Survey on Drug Use and Health (HHS Publication No. SMA 18–5068, NSDUH Series H-53). Rockville, MD: SAMHSA. www.samhsa.gov/data/sites/default/ files/cbhsq-reports/NSDUHFFR2017/ NSDUHFFR2017.pdf Sander, L., Rausch, L., & Baumeister, H. (2016). Effectiveness of internet-based interventions for the prevention of mental disorders: A systematic review and meta-analysis. JMIR Mental Health, 3, e38. www.jmir.org/2016/3/e38/ Schleibe, M., Reichelt, J., Bellman, M., & Kirch, W. (2015). Acceptance factors of mobile apps for diabetes by patients aged 50 or older: A qualitative study. Medicine 2.0, 4, e1. www.medicine20.com/ 2015/1/e1/ Schoeppe, S., Alley, S., Van Lippevelde, W. et al. (2016). Efficacy of interventions that use apps to improve diet, physical activity and sedentary behaviour: A systematic review. The International Journal of Behavioral Nutrition and Physical Activity, 13, 127. https://doi.org/10.1186/s12966016-0454-y

Sijbrandij, M., Kunovsk, I., & Cuijpers, P. (2016). Effectiveness of internet-delivered cognitive behavioral therapy for posttraumatic stress disorder: A systematic review and meta-analysis. Depression and Anxiety, 33, 783–791. https://doi .org/10.1002/da.22533 Solbrig, L., Whalley, B., Kavanagh, D. J. et al. (2018). Functional imagery training versus motivational interviewing for weight loss: A randomised controlled trial of brief individual interventions for overweight and obesity. International Journal of Obesity, 43, 883–894. https://doi.org/10.1038/ s41366-018-0122-1 Stoyanov, S., Hides, L., Kavanagh, D. J., Tjondronegoro, D., Zelenko, O., & Mani, M. (2015). Mobile application rating scale (MARS). A new tool for assessing the quality of healthrelated mobile applications. JMIR mHealth and uHealth, 3, e27. https://mhealth.jmir.org/2015/1/ e27/ Stoyanov, S., Hides, L., Kavanagh, D. J., & Wilson, H. (2016). Development and validation of the user version of the mobile application rating scale (uMARS). JMIR mHealth and uHealth, 4, e72. https://doi.org/10.2196/mhealth.5849 Sucala, M., Cuijpers, P., Muench, F. et al. (2017). Anxiety: There is an app for that. A systematic review of anxiety apps. Depression and Anxiety, 34, 518–525. https://doi.org/10.1002/da.22654 Sztein, D., Koransky, C. E., Fegan, L., & Himelhoch, S. (2018). Efficacy of cognitive behavioural therapy delivered over the Internet for depressive symptoms: A systematic review and metaanalysis. Journal of Telemedicine and Telecare, 24, 527–539. https://doi.org/10.1177/ 1357633X17717402 Titov, N., Dear, B., Nielssen, O. et al. (2018). ICBT in routine care: A descriptive analysis of successful clinics in five countries. Internet Interventions, 13, 108–115. https://doi.org/10.1016/j. invent.2018.07.006 Titov, N., Dear, B., Staples, L. G. et al. (2017). The first 30 months of the MindSpot Clinic: Evaluation of a national e-mental health service against project objectives. Australian and New Zealand Journal of Psychiatry, 51, 1227–1239. https://doi.org/ 10.1177/0004867416671598

Changing Behavior in the Digital Age

We Are Social, & Hootsuite. (2018). Digital in 2018: Global Overview. https://wearesocial.com/blog/ 2018/01/global-digital-report-2018 Whiteford, H. A., Buckingham, W. J., Harris, M. G. et al. (2014). Estimating treatment rates for mental disorders in Australia. Australian Health Review, 38, 80–85. https://doi.org/10.1071/ AH13142 Zachariae, R., Lyby, M. S., Ritterband, L. M., & O’Toole, M. S. (2016). Efficacy of internet-

429

delivered cognitive-behavioral therapy for insomnia: A systematic review and meta-analysis of randomized controlled trials. Sleep Medicine Reviews, 30, 1–10. https://doi.org/10.1016/ j.smrv.2015.10.004 Zeng, E. Y., Heffner, J. L., Copeland, W. K., Mull, K. E., & Bricker, J. B. (2016). Get with the program: Adherence to a smartphone app for smoking cessation. Addictive Behaviors, 63, 120–124. https://doi.org/10.1016/j.addbeh.2016.07.007

30 Critical and Qualitative Approaches to Behavior Change Kerry Chamberlain and Antonia Lyons

Practical Summary Having robust and successful processes for promoting and achieving behavior change is desirable because such change can assist people to lead healthier and more successful and sustainable lives. However, behavior change is complex, and research and practice in the field struggle to consolidate agreed and successful methods and interventions for behavior change. The field primarily follows a dominant model of research, the scientific paradigm, to design, implement, and research interventions for behavior change. This chapter offers a critical examination of this approach and argues for the value of including more critical and qualitative approaches to behavior change research and practice. A critical perspective has benefit in promoting more critical considerations of how research is undertaken, how findings are used, and who stands to benefit from the interventions proposed and implemented. Qualitative research has benefit in producing different forms of knowledge, giving different insights into how behavior change works and fails. Such research provides a more contextualized understanding of how behaviors are shaped and play out in everyday life and thus how efforts might be better directed to achieve change. Taking a critical qualitative approach, informed by interdisciplinarity, can surmount some of the limitations of the dominant scientific approach and fruitfully extend behavior change research and practice.

30.1 Introduction Behavior change theories and models have a long history in health psychology and related disciplines (see the chapters in Part I). They have prospered because they have the potential to offer a major framework to achieve change in behaviors and to provide the means for developing interventions to ensure people can lead better lives. However, as is widely recognized within the behavior change research literature, the theories, models, and techniques for driving behavior change are highly variable, as are the methods used, the behaviors

targeted, the groups and settings in which they are applied, and the delivery systems used for implementation (Davis et al., 2015; Norris et al., 2018). This means the field is complex, difficult to synthesize, and problematic in terms of developing agreed, functional interventions for change. This chapter provides a critical consideration of behavior change and an overview of major concerns that have been raised regarding behavior change theories and models. It then goes on to https://doi.org/10.1017/9781108677318.030

Critical and Qualitative Approaches to Behavior Change

discuss how critical and qualitative approaches can respond to some of these concerns and have value for extending practices in this field. Throughout the chapter, examples are drawn primarily from research on behavior change in the domains of health and environmental sustainability. The chapter provides arguments for using qualitative research in behavior change research, gives examples of where qualitative approaches have been employed, and outlines social practice theory as a means to address many of the concerns about dominant approaches to behavior change. In addition, critical perspectives and their value to the field of behavior change are considered. The chapter concludes by arguing the benefits for behavior change researchers and practitioners to work interdisciplinarily and by advocating the importance for behavior change researchers to understand and incorporate qualitative approaches into behavior change research, as well as the value of taking a critical perspective on their research.

30.2 Research Paradigms and Questioning Assumptions Approaches to investigating how to change people’s behavior, and developing interventions based on this research, have drawn heavily on the assumptions of behavioral science. Positivist assumptions of objective truth, of a world composed of real objects, a world independent of the researcher, and a view that generalizable causal laws are open to discovery, underly most behavioral science. This “received view” (Gergen, 2019) of science is largely taken for granted in research and theory on behavior change. The assumptions underlying this approach are given very limited, if any, attention by researchers and practitioners in this field. There is little acknowledgment that this work sits within a particular paradigm and that there are alternative paradigms of research and knowledge generation that have the potential to be revealing of the status quo of a field, to unsettle and add value to it. The process

431

of considering paradigms leads to questions about underlying assumptions, which shape the conduct of research and, consequently, also shape the forms of knowledge that are gained (Chamberlain, 2014). A critical examination of these assumptions can reveal how this occurs and how the paradigm adopted may limit the creative development of approaches for effective behavioral change. That said, each approach brings its particular benefits and limitations to behavior change research. Critical and qualitative approaches have the potential to bring unique perspectives to bear on a body of knowledge developed primarily within a positivist, scientific paradigm. “Critical” here means explicitly considering the nature of knowledge generation, the paradigms in which it was generated, and the consequences of this in terms of what knowledge is produced, with whom, and who benefits from this knowledge. “Qualitative” means research approaches that do not primarily employ quantitative data, or reduce experience to numbers, approaches that situate their data and analysis in context, work primarily through induction rather than deduction, and seek to interpret rather than describe data (a summary of key qualitative approaches used in research is provided in Table 30.1). These approaches tend to be more open and to sit within constructivist and interpretive paradigms, where the aims are to gain in-depth understandings of the nature of the social world (Gergen, Josselson & Freeman, 2015). However, tensions exist between these paradigms and the positivist paradigm utilized in the physical sciences in terms of epistemology. They generally have different aims (e.g., gaining insight and understanding as opposed to identifying causation), different methodologies, and different criteria for evaluation. For these reasons, qualitative approaches are often marginalized or dismissed as not relevant, or not providing “real” knowledge (Hagger & Chatzisarantis, 2011; Lyons, 2011). Critical and qualitative approaches have much to offer to the behavior change field.

432

KERRY CHAMBERLAIN AND ANTONIA LYONS

Table 30.1 Key qualitative approaches for research Approach

Description and Key References

Narrative inquiry

Identifying the narrative (storied) content of social and personal accounts of experience (Clandinin, 2013) Identifying how knowledge and power are constructed, circulated, and contested and how practices are located in ideologies (Wodak & Meyer, 2016) Identifying the essential nature of experience (Van Manen, 2016) Identifying, largely through participant methods, insights into the field of interest and ways in which it may be changed or altered to improve lives (Madison, 2019) Identifying ways in which social change can be affected in partnership between researchers and the people affected (Chevalier & Buckles, 2013) Identifying a local and specific theory of a phenomenon, usually focused on the process involved (Bryant & Charmaz, 2007)

Critical discourse analysis Phenomenological analysis Critical ethnographic methodology Participatory action research Grounded theory

In adopting the approaches and methods of the physical sciences, behavioral science researchers and practitioners assume that they can develop and use theories and models of change that are universally applicable. In other words, they adopt general assumptions that it is possible to identify a parsimonious set of fundamental laws that underlie and regulate all behavioral change approaches and solutions. These assumptions determine the directions, practices, and findings of the research (Carter & Little, 2007). This approach is highlighted, for example within health psychology, by the extensive recent attempts to systematize procedures for developing change interventions, indexed by attempts to develop systematic taxonomies of behavior change methods (e.g., Kok et al., 2016), attempts to identify the “active components” of behavior change (e.g., Michie et al., 2013), and procedures such as French et al.’s (2012) proposals for a standardized four-step process to developing interventions (see Chapter 20, this volume). These kinds of integration are admirable in their goals and have gained widespread acceptance and approval in health psychology but, as Ogden (2016) has argued, there is also need to consider whether they advance the discipline, are

appropriate for expanding research questions, and whether they promote innovation and new ways of thinking about people and the behaviors in which they engage. In attempting to develop universal theories and models of research, much behavior change research has been the subject of ongoing critique for sidelining the social contexts in which people live their lives (Crossley, 2000; Lyons & Chamberlain, 2017; Marks, 2013; Mielewczyk & Willig, 2007; Murray & Chamberlain, 2014). Psychology and behavioral science tend to take a strongly individualized approach to theory and research, locating change within the individual and giving little attention to contextual demands and constraints on change (but see, by contrast, Chapters 17 and 43, this volume). Behavior change is predominantly sought by seeking individual change and taking a strongly individualized approach to intervention. Psychological theory has been drawn on heavily in developing interventions, with relatively little consideration of concepts and theories from other disciplines and perspectives. For example, in the discipline of health psychology, Holman, Lynch, and Reeves (2017) employed bibliometric techniques to map the field of intervention research. They

Critical and Qualitative Approaches to Behavior Change

showed how, although publications on health behavior interventions have grown substantially across the last three decades, and especially since 2006, this growth has been dominated by psychological research, aided by epidemiological and health economics research, with other relevant social science disciplines, such as sociology, anthropology, and geography, making much smaller contributions by comparison. Another health-related example comes from the UK National Institute for Health and Care Excellence (NICE), which issued recommendations for behavior change at population, community, and individual levels (NICE, 2007) more than ten years ago, noting that “the most effective interventions are those that target several levels simultaneously and consistently” (Tombor & Michie, 2017). Although this guidance “provides a systematic, coherent and evidence-based approach, considering generic principles for changing people’s health-related knowledge, attitudes and behavior, at individual, community and population levels” (NICE, 2007), NICE issued a new behavior change guideline in 2014, focused specifically on individual change (NICE, 2014) (and, coincidentally, changed the title of the 2007 guideline from Behavior Change to Behavior Change: General Approaches). This reflects the disciplinary dominance of individualized psychological approaches in this field and further constrains the development and application of interventions at other levels. Disciplinaryspecific approaches, while enhancing careers and status, are unlikely to assist in achieving widespread changes in behavior across individuals, groups, communities, and populations. Furthermore, previous critiques of applying behavior change models and theories within health psychology have not necessarily led to changes in the ways in which key ideas are conceptualized, investigated, and implemented. More than a decade ago, Mielewczyk and Willig (2007) critiqued the ways that behavior change models were not fit for purpose in health

433

psychology, yet they still continue to be applied without addressing their contextual constraints – what precisely constitutes a health behavior, how and where is it exhibited, and by whom? This critique was extended more recently by Ogden (2016), who argued that the integration and systemizing of behavior change approaches limits both patient and theory variability, yet variability is something professionals working with people should celebrate rather than attempt to reduce. A recent discussion of how to advance the field from observation to intervention (Sheeran, Klein, & Rothman, 2017; see also Chapter 20, this volume) falls back to arguing for the adoption of new methods, taken from experimental medicine. This continues the prioritization of individualism and “scientific” research rather than examining the perspectives of people and the situations in which behaviors occur. Behavior change psychologists acknowledge that interventions are complex (Tombor & Michie, 2017), demonstrated through the ongoing search for categorization, identification of active ingredients and mechanisms of action, arguments for developing more unified theory (e.g., capabilities, opportunities, and motivation behavior-based theory of change model (COM-B), Michie & Wood, 2015; or single theory of preventive behavior (STOPB), Abraham et al., 2009), and pressure on correct reporting of interventions (e.g., Johnston et al., 2018). However, in these ways, the field continues to shape and frame both the development and the scope of interventions and, in doing so, limits the potential for successful behavior change. All of this research continues, even though many behavior change reviews show limited, mixed, or no effects of interventions (e.g., Bouton, 2014). For example, Stautz et al. (2018) report a systematic review of studies exploring the role of self-control in interventions to change alcohol, tobacco, and food consumption. They report that “Of 54 eligible studies, 22 reported no evidence of modification, 18 reported

434

KERRY CHAMBERLAIN AND ANTONIA LYONS

interventions to be less effective in those with low self-control, and 14 reported interventions to be more effective in those with low self-control. This pattern did not differ from chance” (p. 157). This is perhaps a result of accumulating studies across several different behavioral categories, each differing considerably both within and between categories, all occurring in quite different contexts. Further, although the authors did explore different aspects of self-control, this research remains focused on individual, internalized trait notions for the causes of, and routes to, intervention for behavior change. As Kelly and Barker (2016) argue, foregrounding individual behavior as the key focus of change “avoids having to think about the complexity of the social, political and economic factors which influence people’s health and sidesteps confronting the powerful vested commercial interests that may not want people to change their behavior to more healthy ways of living” (p. 110). Although they direct this argument at policy makers, many health psychologists and behavioral scientists also fall into this trap. Kelly and Barker (2016) also argue that, rather than trying to predict behavior change, a “regressive inference approach” (p. 113) is needed, which seeks to understand the preceding conditions that led to the behavior, the context in which it occurs, and the social patterning that holds the behavior in place (see also Holman & Borgstrom, 2016). This subject is discussed further in the following sections (see also Chapter 28, this volume).

30.3 Qualitative Approaches to Behavior Change Qualitative approaches can be valuable for exploring the meanings that individuals attach to their behavior, as well as their experiences of interventions related to changing their behavior. This can inform the development of interventions, ensuring that they are devised and implemented in the most effective ways. For example,

Lally and colleagues (2011) explored the experiences that adults had when they were enrolled on an eight-week weight-loss intervention, particularly in terms of how changes in behavior became a habit. Their thematic analysis of interview transcripts demonstrated that, while behavior change was initially cognitively effortful, these adults found the changes easier as they became more automatic. This automaticity was linked to workbased contexts and having specific cues associated with the habits (see Chapter 41, this volume). In a different context, Penn and colleagues (2008) examined the experiences of people who had maintained changes in their diet and exercise over 3–5 years to prevent diabetes. They conducted in-depth interviews with fifteen such people to identify key features that helped them maintain their changes in behavior, and also those that hindered changes, to inform future interventions. Key findings highlighted that having both social and professional support, and getting fitter, helped these people to maintain their changes in diet and exercise but that physical conditions (such as arthritis) and social demands (such as caring for others) hindered such changes. They concluded that holistic views of people implementing and maintaining behavior change are important and that people’s life contexts – and how these change over time – are key factors that should be taken into account. As they noted “it takes time for behavior change to be learnt and absorbed into routine” (p. 235) and people’s experiences can highlight the challenges of implementing long-lasting behavior change. Beyond gaining insight into people’s experiences of behavior change interventions, qualitative approaches also enable researchers to move beyond codified and deterministic positivist methods to create knowledge. On their own, these methods cannot capture the complex nature of human behavior (Hilton & Johnston, 2017). As discussed in Section 30.2, behaviors are often assumed to be unified in intervention research (Ogden, 2016) and often treated as devoid of

Critical and Qualitative Approaches to Behavior Change

context. However, many studies have documented that ignoring the contexts in which behaviors occur can lead to very limited understandings, which in turn make interventions ineffective. Qualitative research approaches can be used to explicitly investigate the contexts of behaviors and the meanings that people ascribe to them. Taking smoking cessation as an example, some years ago Laurier, McKie, and Goodwin (2000) documented in detail how smoking was a very complex, and contextually determined, behavior. By analyzing interviews with smokers, Laurier and colleagues illustrated how the simple act of smoking a cigarette varies considerably across the day, where the meaning of the first cigarette in the morning, the last cigarette of the day, the social cigarette in the bar, and the “snatched” smoke during work are all fundamentally different, both as behaviors and for the person so behaving. Hence smoking cannot be subsumed under a general rubric of a health behavior if it is to be understood fully and its incidence reduced. Further, Christakis and Fowler (2008) have demonstrated how smoking, and change in smoking behavior, is embedded within social networks, further reflecting the need to understand and address behavior in context when seeking to create interventions that can make change. Other researchers argue that smoking is a behavior that involves a whole set of actions that together make a “bundle” or practice (e.g., rolling a cigarette, lighting and inhaling it) (Blue et al., 2016). This practice is linked to other social practices, such as taking a tea break at work, drinking alcohol, relaxing in the evening, and so on. Research on vaping as a social practice highlights similar findings. Keane et al. (2017) showed how vaping is a practice that opens up space and time, allowing for nicotine to be consumed in the home and bundled with other important social practices. There was a positive web of meaning that vapers gave to their vaping, including “concepts of health, freedom, transformation and choice” (p. 473). Conceptualizing a behavior

435

as a practice that is coordinated with other social practices allows researchers, practitioners, and policy makers to consider the ways in which people “are recruited to, and how they come to defect from” specific behaviors within their everyday lives (Blue et al., 2016, p. 40).

30.3.1 Social Practice Theory Approaching the understanding of behavior as a social practice is counter to the widespread acceptance of rational choice theory that unpins individually oriented behavior change. Instead, the focus is on how social practices, such as eating healthily or engaging in environmentally sustaining behaviors, emerge, are sustained, evolve, and disappear. A social practice approach provides much greater insight into how practices go together and how they are routinized and synchronized in everyday life. It shifts attention away from individual decision-making moments and toward people negotiating and enacting social practices in their daily lives (Bouton, 2014; Hargreaves, 2011; Kelly & Barker, 2016; Shove & Warde, 2002). This approach forces, as Giddens (1984) has argued, the dissolution of structure and agency arguments, as it theorizes human activities as shaped by structured rules and understandings, which are simultaneously reproduced and reworked by the ongoing flow of human actions. This means that many human activities are conducted without voluntary or conscious direction (e.g., much smoking, drinking, and eating) but occur within the flow of everyday social living, ordered by space and time and dependent on practical or tacit knowledge (Shove, Pantzar, & Watson, 2012). For example, Twine (2018) examined the rise of vegan eating and documents how a materialist practice approach can explain the recent rise in veganism as well as arguing that such findings have value for informing policies and interventions to increase plant-based eating. Similarly, Kippax discusses the value and use of social practice approaches in

436

KERRY CHAMBERLAIN AND ANTONIA LYONS

the field of HIV interventions (Kippax, 2018; Kippax & Race, 2003). Similar arguments have been made for alcohol consumption, where researchers argue for a shift in the unit of analysis from individual drinkers to drinking practices and “cultures” (Meier, Warde, & Holmes, 2018). This shifts the focus “from alcohol consumption to drinking occasions; specifically how, when, where, why and with whom drinking and getting drunk occur and vary across time, place and population” (p. 212). Changing the nature of drinking occasions, rather than individuals’ behavior within those occasions, provides an alternative point of disruption. This raises possibilities for changing the practice (and its consequent bundles of actions, rhythms, and routines) and changing the meanings associated with the practice. This approach also allows an insight into how social practices vary across different social groups. Alcohol consumption is the act of drinking, but what this act means varies vastly across different groups and settings. Patterns of consumption (what, when, how much, who with), and their associated meanings, differ from teenage parties, to religious ceremonies, to craft beer tastings, to after-work drinks, to dinner parties, to weddings, to gay and lesbian holiday packages, and so on (Kelly & Barker, 2016). Kelly and Barker (2016) argue that designing interventions to reduce alcohol consumption based simply on age, social class, or an overarching behavior change model is very limited, or in their words “clearly pointless” (p. 114), as it is targeting interventions at people who are intoxicated. Rather, identifying an intervention point and approach based on setting and context (and examining the situated social practices) is more likely to be effective. As they note, when the practice of smoking changed (when its links and networks changed and it became less glamorous and more socially undesirable), people changed with it. They see “the breaking up of the links within the practice of smoking” (p. 114) as the key to public health success.

Hargreaves (2011) applied social practice theory to a behavior change initiative that sought to increase pro-environmental behaviors in the workplace, as described in detail in Sidebar 30.1. Such qualitative in-depth approaches to examining behavior and change processes are able to reveal far greater insight into the relevant aspects and complexities of daily life than is possible with the conventional and dominant theories and models utilized within psychology. While linear and individualistic models of behavior change are appealing because they “render policy responses straightforward” (Hargreaves, 2011, p. 82), they “fail to appreciate the ways in which, variously, social relations, material infrastructures and context are intrinsic to the performance of social practices” (p. 83). A social practice approach highlights ways of breaking existing unsustainable practices and replacing them with more sustainable patterns, providing knowledge that extends and improves both policy and interventions.

30.4 Critical Perspectives on Behavior Change Research A critical approach to behavior change research, at its broadest, asks simply who benefits from framing research about people’s behavior in particular ways? As has been argued throughout this chapter, what people do and how they behave is often neither straightforward nor rational and is embedded within a complex set of personal, social, political, and economic factors and situated within varying cultural and geographic environments. For example, the ease of getting people to recycle household waste products depends on council recycling schemes, the function of recycling centers, the ability to buy recyclable packaging, the costs of recycling (both time and money), not to mention people’s understanding of the environmental consequences of reducing waste (World Bank figures show that more than 2 billion metric tons of waste is being

Critical and Qualitative Approaches to Behavior Change

437

Sidebar 30.1 Case study of behavior change in the workplace using a social practice perspective (Hargreaves, 2011)

Hargreaves (2011), working in the area of pro-environmental behavior change, offers a detailed examination of a specific behavior change initiative. He used social practice theory in a critical ethnographic approach, which “de-centers individuals from analyses, and turns attention instead to the social and collective organization of practices” (p. 79). Hargreaves was involved in participant observation in the site for nine months, attending all meetings to plan, establish, complete, and evaluate the initiative and keeping an extensive field diary (also a data source). He also conducted thirty-eight in-depth interviews with key individuals, those running the initiative and a range of people involved and affected by the initiative. The findings provided a detailed, contextualized view of the ways that changes in behavior unfolded in the contexts of everyday working life within the organization, documenting the surprises, difficulties, and successes involved in challenging and changing people’s practices. Importantly, this research was able to identify mundane “footholds” for behavior change, above and beyond individual attitudes and values, to implicate how practices-as-entities (the goals of the initiative) were transformed into practices-asperformances (routine actions) and incorporated into the organization of everyday life within the setting.

produced every year and that figure is estimated to rise to 3.4 billion by 2050 – with major consequences for sustainable, healthy, and inclusive cities and communities; Kaza et al., 2018). Who benefits from the vast amount of plastic used in packaging foods and other goods? Overwhelmingly, commercial interests benefit – industries and companies that use the best (often cheapest) materials to move, distribute, and sell more products and thus increase their profits for their shareholders. More broadly still, Western economies benefit, as they depend on (over) consumption for their success (Nair, 2018). Who benefits from getting individual people to change their behavior to recycle waste materials? Market-driven economies and industries benefit, and the focus on changing the behavior of individuals obscures the role that more macro, upstream factors play in this issue (such as the production and use of the materials in the first place) (Acunzo et al., 2018; Evans, Welch, & Swaffield, 2017).

When considering changing behaviors that are related to health, similar patterns are observed. A focus on the individual to change their views and increase their motivation, to form intentions and habits to engage in healthier behavior, shifts the focus away from the complex social and economic environments in which they live. As Kelly and Barker (2016) put it, “the behaviors which need to be changed are sustained and nurtured by highly profitable industries selling goods which make people ill – sugar rich, energy dense fatty foods and alcoholic beverages as well, of course, as tobacco” (p. 110). The construction of health-related behaviors as primarily due to individual choice, as something totally within individuals’ control if they just choose to do it and had stronger motivation, benefits these commercial industries. This individual choice framework facilitates them to argue for deregulation to increase their sales, lobby for lower taxes and for increased product availability by seeking to sell in more places and have more outlets, and to

438

KERRY CHAMBERLAIN AND ANTONIA LYONS

campaign for increased freedom to market their products as widely as possible. All of this in turn affects the choices that people can and cannot make within the context and society in which they live. Focusing on individual behavior and choice means that the political landscape – where these industries often have powerful vested interests – is ignored or not fully considered. Such industries lobby governments to ensure that behavior change approaches (particularly education and increasing knowledge) are dominant because these approaches have not been highly effective at the population level and thus do not have a great impact on consumption and sales of unhealthy commodities.

30.5 Implications for Research, Practice, and Translation As argued in Section 30.2, utilizing qualitative research in behavior change research has considerable potential for enhancing the field of behavior change. However, qualitative research, in seeking to contribute to a field dominated by positivist understandings of what constitutes acceptable knowledge, faces considerable challenge. Knowledge gained through qualitative research is often evaluated by the wrong standards (Malterud, 2019), often regarded as subsidiary to “proper” evidence-based knowledge or required to be conducted in certain ways (e.g., Majid & Vanstone, 2018), especially if used in intervention research (e.g., Noyes et al., 2018). As Malterud (2019) argues, methodologies developed for research within a scientific evidence-based framework are not necessarily fit for purpose in seeking qualitative synthesis. Qualitative research not only produces different forms of knowledge, giving different insights into how behavior change works and often fails, but also offers, and to some degree forces, a different perspective: a much more contextual understanding of the location and function of behaviors and how efforts might be better

directed to achieve change. Adopting a qualitative perspective also helps to promote two further potential changes in this field, in promoting the need to seek more interdisciplinary practice and in forcing a more critical examination of research approaches, the use of findings, and consideration of who stands to gain from the interventions proposed and put into practice. This is not to argue that all behavior change researchers must undertake qualitative, interdisciplinary research but that the field would benefit if they took a broader perspective and adopted a more critical perspective on their research designs and the implications, and applications, of their research findings. As long as they continue to consider behaviors as unitary and individualized, and change as rational, then they will continue to perpetuate the very issues that Kelly and Barker (2016), among others, have identified as limiting and restricting the understanding, adoption, and application of behavior change interventions. Behavior change researchers need to reconsider these basic assumptions, work more collaboratively with social science and other relevant researchers and practitioners, such as those in public health and political and environmental studies, and take a more nuanced look at how behaviors are contextualized, complexly situated and socially and culturally enabled and patterned and how points of intervention at different levels of operation – individual, social, cultural, national – can be brought together to produce sustained change in behaviors. Fortunately, there is some indication that this is happening, with the recognition of the limitation of social cognition models (e.g., Sniehotta, Presseau, & AraujoSoares, 2014), the appearance of critiques from within psychology about the limitations of its approaches (e.g., Hilton & Johnston, 2017; Ogden, 2016), and, potentially, an increasing acceptance and use of qualitative methods for research within psychology (Gergen et al., 2015; Kidd, 2002; Levitt et al., 2018). This is reinforced by calls for the incorporation of qualitative

Critical and Qualitative Approaches to Behavior Change

research into medical and policy decision-making (e.g., Dohan et al., 2016; Greenhalgh et al., 2016; Marini, 2015).

30.6 Summary and Conclusions This chapter argues that behavior change research needs to extend its brief, to work more interdisciplinarily, to appreciate the contributions of qualitative research and incorporate them into change research agendas and to adopt a critical approach to researching behavior change and delivering interventions. This would lead to a substantial reconsideration of how research in this area should be framed and conducted. A critical perspective raises important questions about who the field seeks to serve and how intervention research can best work to serve the interests of particular groups in society. In overlooking or ignoring the wider societal interests and power structures that sustain behaviors, and in retaining a focus on individual change, the field often fails its mandate to work for the benefit of people on the ground.

References Abraham, C., Kelly, M., West, R., & Michie, S. (2009). The UK National Institute for Health and Clinical Excellence public health guidance on behaviour change: A brief introduction. Psychology, Health and Medicine, 14, 1–8. https://doi.org/10.1080/ 13548500802537903 Acunzo, D., Escher, G., Ottersen, O. et al. (2018). Framing planetary health: Arguing for resourcecentred science. The Lancet Planetary Health, 2, e101–e102. https://doi.org/10.1016/S2542-5196 (18)30023-8 Blue, S., Shove, E., Carmona, C., & Kelly, M. P. (2016). Theories of practice and public health: Understanding (un)healthy practices. Critical Public Health, 26, 36–50. https://doi.org/10.1080/ 09581596.2014.980396 Bouton, M. E. (2014). Why behavior change is difficult to sustain. Preventive Medicine, 68, 29–36. https:// doi.org/10.1016/j.ypmed.2014.06.010

439

Bryant, A., & Charmaz, K. (Eds.). (2007). The SAGE Handbook of Grounded Theory. Thousand Oaks, CA: SAGE. Carter, S. M., & Little, M. (2007). Justifying knowledge, justifying method, taking action: Epistemologies, methodologies, and methods in qualitative research. Qualitative Health Research, 17, 1316–1328. https://doi.org/10.1177/ 1049732307306927 Chamberlain, K. (2014). Epistemology and qualitative research. In P. Rohleder & A. Lyons (Eds.), Qualitative Research in Clinical and Health Psychology (pp. 9–28). Basingstoke: Palgrave Macmillan. Chevalier, J., & Buckles, D. (2013). Participatory Action Research: Theory and Methods for Engaged Inquiry. New York: Routledge. Christakis, N., & Fowler, J. (2008). The collective dynamics of smoking in a large social network. New England Journal of Medicine, 358, 2249–2258. https://doi.org/10.1056/ NEJMsa0706154 Clandinin, D. (2013). Engaging in Narrative Inquiry. New York: Routledge. Crossley, M. (2000). Rethinking Health Psychology. Buckingham: Open University Press. Davis, R., Campbell, R., Hildon, Z., Hobbs, L., & Michie, S. (2015). Theories of behaviour and behaviour change across the social and behavioural sciences: A scoping review. Health Psychology Review, 9, 323–344. https://doi.org/ 10.1080/17437199.2014.941722 Dohan, D., Garrett, S. B., Rendle, K. A., Halley, M., & Abramson, C. (2016). The importance of integrating narrative into health care decision making. Health Affairs, 35, 720–725. https://doi .org/10.1377/hlthaff.2015.1373 Evans, D., Welch, D., & Swaffield, J. (2017). Constructing and mobilizing the “consumer”: Responsibility, consumption and the politics of sustainability. Environment and Planning A, 49, 1396–1412. https://doi.org/10.1177/ 0308518X17694030 French, S., Green, S., O’Connor, D. et al. (2012). Developing theory-informed behavior change interventions to implement evidence into practice: A systematic approach using the Theoretical

440

KERRY CHAMBERLAIN AND ANTONIA LYONS

Domains Framework. Implementation Science, 7, 38. https://doi.org/10.1186/1748-5908-7-38 Gergen, K. J. (2019). Qualitative psychology and the new pluralism. In B. Schiff (Ed.), Situating Qualitative Methods in Psychological Science (pp. 61–71). New York: Routledge. Gergen, K. J., Josselson, R., & Freeman, M. (2015). The promises of qualitative inquiry. American Psychologist, 70, 1–9. https://doi.org/10.1037/ a0038597 Giddens, A. (1984). The Constitution of Society: Outline of the Structuration Theory. Cambridge: Polity. Greenhalgh, T., Annandale, E., Ashcroft, R. et al. (2016). An open letter to The BMJ editors on qualitative research. BMJ, 352, i563. https://doi .org/10.1136/bmj.i563 Hagger, M., & Chatzisarantis, N. (2011). Never the twain shall meet? Quantitative psychological researchers’ perspectives on qualitative research. Qualitative Research in Sport, Exercise and Health, 3, 266–277. https://doi.org/10.1080/ 2159676X.2011.607185 Hargreaves, T. (2011). Practice-ing behaviour change: Applying social practice theory to proenvironmental behaviour change. Journal of Consumer Culture, 11, 79–99. https://doi.org/ 10.1177/1469540510390500 Hilton, C. E., & Johnston, L. H. (2017). Health psychology: It’s not what you do, it’s the way that you do it. Health Psychology Open, 4. https://doi .org/10.1177/2055102917714910 Holman, D., & Borgstrom, E. (2016). Applying social theory to understand health-related behaviours. Medical Humanities, 42(2), 143–145. https://doi .org/10.1136/medhum-2015-010688 Holman, D., Lynch, R., & Reeves, A. (2017). How do health behaviour interventions take account of social context? A literature trend and co-citation analysis. Health, 22, 389–410. https://doi.org/ 10.1177/1363459317695630 Johnston, M., Johnston, D., Wood, C. E., Hardeman, W., Francis, J., & Michie, S. (2018). Communication of behaviour change interventions: Can they be recognised from written descriptions? Psychology and Health, 33, 713–723. https://doi.org/10.1080/ 08870446.2017.1385784

Kaza, S., Yao, L. C., Bhada-Tata, P., & Van Woerden, F. (2018). What a Waste 2.0: A Global Snapshot of Solid Waste Management to 2050. Washington, DC: World Bank. https://openknowledge.world bank.org/handle/10986/30317 Keane, H., Weier, M., Fraser, D., & Gartner, C. (2017). “Anytime, anywhere”: Vaping as social practice. Critical Public Health, 27, 465–476. https://doi .org/10.1080/09581596.2016.1250867 Kelly, M., & Barker, M. (2016). Why is changing health-related behaviour so difficult? Public Health, 136, 109–116. https://doi.org/10.1016/ j.puhe.2016.03.030 Kidd, S. (2002). The role of qualitative research in psychological journals. Psychological Methods, 7, 126–138. https://doi.org/10.1037/1082989X.7.1.126 Kippax, S. (2018). A journey to HIV prevention research: From social psychology to social health via multidisciplinarity. Journal of Health Psychology, 23, 442–456. https://doi.org/10.1177/ 1359105317707529 Kippax, S., & Race, K. (2003). Sustaining safe practice: Twenty years on. Social Science and Medicine, 57, 1–12. https://doi.org/10.1016/S0277-9536(02) 00303-9 Kok, G., Gottlieb, N., Peters, G.-J. et al. (2016). A taxonomy of behaviour change methods: An intervention mapping approach. Health Psychology Review, 10, 297–312. https://doi.org/ 10.1080/17437199.2015.1077155 Lally, P., Wardle, J., & Gardner, B. (2011). Experiences of habit formation: A qualitative study. Psychology, Health and Medicine, 16, 484–489. https://doi.org/10.1080/13548506.2011.555774 Laurier, E., McKie, L., & Goodwin, N. (2000). Daily and lifecourse contexts of smoking. Sociology of Health and Illness, 22, 289–309. https://doi.org/ 10.1111/1467-9566.00205 Levitt, H. M., Bamberg, M., Creswell, J. W., Frost, D. M., Josselson, R., & Suárez-Orozco, C. (2018). Journal article reporting standards for qualitative primary, qualitative meta-analytic, and mixed methods research in psychology: The APA Publications and Communications Board task force report. American Psychologist, 73, 26. https://doi.org/10.1037/amp0000151

Critical and Qualitative Approaches to Behavior Change

Lyons, A. C. (2011). Advancing and extending qualitative research in health psychology. Health Psychology Review, 5, 1–8. https://doi.org/ 10.1080/17437199.2010.544638 Lyons, A. C., & Chamberlain, K. (2017). Critical health psychology. In B. Gough, (Ed.), Palgrave Handbook of Critical Social Psychology (pp. 533–555). London: Palgrave Macmillan. https://doi.org/10.1057/978-1-137-51018-1_26 Madison, D. S. (2019). Critical Ethnography: Method, Ethics, and Performance. London: SAGE. Majid, U., & Vanstone, M. (2018). Appraising qualitative research for evidence syntheses: A compendium of quality appraisal tools. Qualitative Health Research, 28, 2115–2131. https://doi.org/10.1177/1049732318785358 Malterud, K. (2019). The impact of evidence-based medicine on qualitative metasynthesis: Benefits to be harvested and warnings to be given. Qualitative Health Research, 29, 7–17. https://doi.org/ 10.1177/1049732318795864 Marini, M. G. (2015). Narrative Medicine: Bridging the Gap Between Evidence-Based Care and Medical Humanities. Cham: Springer International. Marks, D. F. (2013). Health psychology: Overview. In A. Nezu, C. Nezu, & P. Geller (Eds.), Handbook of Psychology, Vol. 9: Health Psychology (2nd ed.). Hoboken, NJ: Wiley. Meier, P., Warde, A., & Holmes, J. (2018). All drinking is not equal: How a social practice theory lens could enhance public health research on alcohol and other health behaviours. Addiction, 113, 206– 213. https://doi.org/10.1111/add.13895 Michie, S., Richardson, M., Johnston, M. et al. (2013). The behavior change technique taxonomy (v1) of 93 hierarchically clustered techniques: Building an international consensus for the reporting of behavior change interventions. Annals of Behavioral Medicine, 46, 81–95. https://doi.org/ 10.1007/s12160-013-9486-6 Michie, S., & Wood, C. (2015). Health behaviour change techniques. In M. Conner & P. Norman (Eds.), Predicting and Changing Health Behaviour: Research and Practice With Social Cognition Models (pp. 358–389). New York: McGraw Hill.

441

Mielewczyk, F., & Willig, C. (2007). Old clothes and an older look: The case for a radical makeover in health behaviour research. Theory and Psychology, 17, 811–837. https://doi.org/10.1177/ 0959354307083496 Murray, M., & Chamberlain, K. (2014). Health psychology. In T. Teo (Ed.), Encyclopaedia of Critical Psychology (pp. 844–850). New York: Springer. https://doi.org/10.1007/978-1-46145583-7_132 Nair, C. (2018). The Sustainable State: The Future of Government, Economy, and Society. Oakland, CA: Berrett-Koehler Publishers. NICE (National Institute for Health and Care Excellence). (2007). Behaviour Change: General Approaches. (Public health guideline [PH6]). London: NICE. NICE (National Institute for Health and Care Excellence). (2014). Behaviour Change: Individual Approaches. (Public health guideline [PH49]). London: NICE. Norris, E., Finnerty, A. N., Hastings, J., Stokes, G., & Michie, S. (2018). Identifying and evaluating ontologies related to human behaviour change interventions: A scoping review. PsyArXiv. https://doi.org/10.31234/osf .io/y7dus Noyes, J., Booth, A., Cargo, M. et al. (2018). Cochrane Qualitative and Implementation Methods Group guidance series – paper 1: Introduction. Journal of Clinical Epidemiology, 97, 35–38. https://doi.org/ 10.1016/j.jclinepi.2017.09.025 Ogden, J. (2016). Celebrating variability and a call to limit systematization: The example of the behaviour change technique taxonomy and the behaviour change wheel. Health Psychology Review, 10, 245–250. https://doi.org/10.1080/ 17437199.2016.1190291 Penn, L., Moffatt, S. M., & White, M. (2008). Participants’ perspective on maintaining behaviour change: A qualitative study within the European Diabetes Prevention Study. BMC Public Health, 8, 235. https://doi.org/10.1186/14712458-8-235 Sheeran, P., Klein, W., & Rothman, A. (2017). Health behavior change: Moving from observation to intervention. Annual Review of Psychology, 68,

442

KERRY CHAMBERLAIN AND ANTONIA LYONS

573–600. https://doi.org/10.1146/annurev-psych010416-044007 Shove, E., Pantzar, M., & Watson, M. (2012). The Dynamics of Social Practice: Everyday Life and How It Changes. London: SAGE. https://doi.org/ 10.4135/9781446250655.n1 Shove, E., & Warde, A. (2002). Inconspicuous consumption: The sociology of consumption, lifestyles and the environment. In R. Dunlap, F. Buttel, P. Dickens, & A. Giswijt (Eds.), Sociological Theory and the Environment: Classical Foundations, Contemporary Insights (pp. 230–250). London: Rowman & Littlefield. Sniehotta, F., Presseau, J., & Araujo-Soares, V. (2014). Time to retire the theory of planned behaviour. Health Psychology Review, 8, 1–7. https://doi.org/ 10.1080/17437199.2013.869710 Stautz, K., Zupan, Z., Field, M., & Marteau, T. (2018). Does self-control modify the impact of

interventions to change alcohol, tobacco, and food consumption? A systematic review. Health Psychology Review, 12, 157–178. https://doi.org/ 10.1080/17437199.2017.1421477 Tombor, I., & Michie, S. (2017). Methods of health behavior change. In O. Braddick (Ed.), Oxford Research Encyclopedia of Psychology. New York: Oxford University Press. https://doi.org/10.1093/ acrefore/9780190236557.013.125 Twine, R. (2018). Materially constituting a sustainable food transition: The case of vegan eating practice. Sociology, 52, 166–181. https://doi.org/10.1177/ 0038038517726647 Van Manen, M. (2016). Phenomenology of Practice: Meaning-Giving Methods in Phenomenological Research and Writing. New York: Routledge. Wodak, R., & Meyer, M. (Eds.). (2016). Methods of Critical Discourse Studies (3rd ed.). London: SAGE.

Part III Behavior Change Interventions: Practical Guides to Behavior Change

31 Attitudes and Persuasive Communication Interventions Kyra Hamilton and Blair T. Johnson

Practical Summary Attitudes help guide people in making decisions in a complex world – what products to buy, what degree to study, what political party to support. They are built up over time and experience and are theorized to assist people in making rapid and efficient behavioral choices and decisions. Despite being relatively enduring, shaping, challenging, or changing people’s attitudes is possible and, given their link to behavior, changing attitudes may be a potential means to change behavior. For example, giving people persuasive information on a particular behavior or course of action may lead to more favorable attitudes toward that behavior and, potentially, behavior change. Attitude interventions show promise for changing behavior and can be delivered by multiple methods including at the individual (e.g., practitionerdelivered) or population (e.g., mass media) level. Several considerations, however, need to be taken into account in the application of these various attitude change techniques and when they may be most or least successful.

31.1 Introduction It is commonly assumed that people act in accordance with their opinions, beliefs, and evaluations of the people, actions, and objects in their environment – put simply, people’s attitudes are assumed to guide their behavior. More than fifty years of research in social psychology has explored the content and structure of attitudes, and their relationship with behavior, based on this common assumption. This body of research has suggested that attitude-behavior relations are complex, and relations between attitudes and behavior are dependent on a number of contextual, interpersonal, and structural factors. In addition, research on attitude-behavior relations has catalyzed interest in how methods to change attitudes, such as persuasive communications,

may lead to behavior change. Such interest is predicated on the assumption that attitudes are modifiable and changeable, which is of paramount interest to interventionists interested in behavior change. This chapter explores the role of attitudes in behavior change and outlines some key theory- and evidence-based strategies that may be leveraged to change behavior through attitude change. The chapter begins by reviewing definitions of attitudes and the typical ways they have been measured. Then follows an overview of the theory underpinning various attitude concepts and the techniques used to change them. Next, select examples of attitude-based behavior change https://doi.org/10.1017/9781108677318.031

446

KYRA HAMILTON AND BLAIR T. JOHNSON

techniques (e.g., information provision and communication-persuasion, cognitive dissonance) are presented. In addition, the evidence base for the use of each technique in changing behavior, along with sample step-by-step guides on how they might be implemented in behavior change interventions, is provided. The chapter concludes with an executive summary of the research on attitude-based behavior change interventions and proposes some challenges and directions for future research using attitudes as a means to change behavior.

31.2 Conceptualization and Measurement of Attitudes 31.2.1 Definitions Attitudes are evaluations of entities, which could be any object, person, behavior, idea, or event in a person’s environment (Eagly & Chaiken, 1993). The evaluations are usually conceptualized in terms of their valence or direction such as bad vs. good, negative vs. positive, dislike vs. like. Attitudes are central to conceptualizations of interpersonal affection (e.g., love, like, hate), intergroup relations (e.g., stereotypes, prejudice, discrimination), objects (e.g., consumer products, neighborhoods, domiciles), and policies (e.g., whether it is desirable to bus children to create mixed racial school classes). Attitudes are typically stored in long-term memory and are often likened to schema, that is, organized sets of stored interconnected information about objects and their evaluations, and are expected to be stored alongside other functional information, such as behavioral scripts – sets of motor patterns that may manifest as behavioral responses. Attitudes are often thought to have cognitive (instrumental), affective (emotional), and conative (behavioral) content and are thought to be central to how people organize and make sense of their social world. Importantly, attitudes are thought to be central to people’s actions and behavioral decisions.

31.2.2 Measurement and Differentiated Components of Attitudes Attitudes are typically measured using self-report scales in which individuals provide graduated ratings of the entity or attitude object. A typical example is the Likert scale, a numbered scale (ranging from, e.g., 1 to 5) with labels to accompany each number (e.g., 1 = very negative, 2 = negative, 3 = neither negative or positive, 4 = positive, 5 = very positive). Also popular is the semantic differential scale, which omits labels on each point and uses only extremes at each end (very negative vs. very positive). Yet scholars debate whether attitudes are truly best measured on bipolar scales. In some instances, it may be more advantageous to measure attitudes as two unipolar dimensions ranging (1) from neutral to very negative and (2) from neutral to very positive (Cacioppo & Berntson, 1994). This scaling strategy makes it possible to gauge ambivalence in a person’s attitudes toward an entity; a person may feel that there are both good and bad aspects associated with it. Some scholars focus on intra-attitudinal consistency, which is the extent to which the tripartite components of attitudes (affective, cognitive, and behavioral) converge (e.g., Breckler, 1984; Fabrigar, MacDonald, & Wegener, 2019). Discrepancies may also be indicative that ambivalence exists. Similarly, other scholars focus on inter-attitudinal consistency, which is the extent to which people possess attitudes toward related entities that agree (Jonas, Broemer, & Diehl, 2000). People can also have relatively strong or relatively weak attitudes, which may reflect the extent to which they endorse their evaluations toward particular objects and may affect the extent to which they are committed to courses of action based on them (Howe & Krosnick, 2017; Johnson & Eagly, 1989). People’s own phenomenological realities typically have no requirement of complete consistency among the sets of attitudes they hold, except under some circumstances. In fact, quite often it is the case that people maintain conflicting attitudes and

Attitudes and Persuasive Communication Interventions

attitudinal components, often inventing theories to organize their beliefs (e.g., Ross, 1989) or having mistaken beliefs and theories about reality (e.g., Nisbett & Wilson, 1977). For example, in contemporary United States, although about half of the population have moral misgivings about slaughterhouses, the vast majority still eat meat. This meat paradox has been the subject of numerous investigations, most of which rest on concepts of interor intra-attitudinal consistency (e.g., Buttlar & Walther, 2018). Finally, one of the limitations of explicit, selfreport attitude measures is that they require individuals to consciously reflect on the attitude object when making their response. However, such measures may lead individuals to provide culturally and socially appropriate responses to attitude scales. For example, individuals from a majority may report positive beliefs toward minority groups (e.g., gender, racial, sexual orientation), which may mask prejudicial beliefs (Dovidio & Gaertner, 2000). This phenomenon has led scholars to use alternative measures (e.g., the implicit association test; Greenwald, McGhee, & Schwartz, 1998) that rely on reaction-time responses pairing up valence stimuli with stimuli from categories of the attitude object (e.g., black faces vs. white faces). The implicit attitude measures aim to capture the “true” underlying attitudes toward stimuli that tend to be suppressed in explicit measures, whether people realize it or not. Unfortunately, evidence on the extent to which implicit measures of attitudes improve behavioral prediction over that explained by explicit attitude measures is inconclusive, although some evidence exists (Gawronski, 2019; Phipps, Hagger, & Hamilton, 2019). In particular, meta-analyses directed by teams that originated implicit measures have found support for their validity in predicting observed behavior (Greenwald et al., 2009). In contrast, critics’ meta-analyses of the same studies have been much more dismissive (Oswald et al., 2013), and it is unclear what factors are responsible for differences across studies (Johnson, Landrum, & McCloskey, 2019). For

447

examples of explicit and implicit measures of attitudes, see Appendixes 31.1 and 31.2 (supplemental materials).

31.3 Theories of Attitude Change Many social psychological theories (e.g., theory of planned behavior, Ajzen, 1991) suggest that attitudes guide behavior; thus, changing attitudes can change behavior. One reasoning for this is because individuals strive for consistency between their thoughts and actions, and the drive for consistency represents a fundamental motivational process (Festinger, 1964; Heider, 1958). The most prominent consistency theory is cognitive dissonance theory (Festinger, 1962, 1964). Dissonance theory concerns itself with the processes by which individuals resolve inconsistencies in their beliefs, attitudes, values, and actions. For example, a climatechange denier might encounter some information that runs counter to their beliefs (e.g., that summers are ever hotter or that weather is ever more variable). The inconsistency between their attitudes toward climate change and the counter-attitudinal information may lead to the experience of cognitive dissonance, a feeling of psychological discomfort caused by the internal inconsistency. Festinger (1962, 1964) posited that a person who experiences dissonance will be motivated to take action to reduce the dissonance. Numerous strategies could be adopted to reduce dissonance. For example, the individual might make adjustments to their beliefs to justify or bolster their existing attitudes (e.g., by adding new beliefs that question and discredit the source of the counter-attitudinal information) or they might avoid further contradictory information (e.g., listening to sources of news that are consistent with their worldview). Of course, they may also alter their behavior and adjust their attitudes toward climate change, although such drastic change may occur only under certain conditions. The mechanisms underpinning attitude change in cognitive dissonance theory is physiological arousal (Croyle & Cooper, 1983) and felt discomfort (Elliot &

448

KYRA HAMILTON AND BLAIR T. JOHNSON

Devine, 1994), although such discomfort is not always necessary (Harmon-Jones et al., 1996). Cognitive dissonance has left an indelible mark on the attitude and attitude change literature, with many experiments and interventions using dissonance as a means to change attitudes (e.g., Freijy & Kothe, 2013; Harmon-Jones, Armstrong, & Olson, 2019) and perhaps also behavior (Eagly & Chaiken, 1993; Harmon-Jones et al., 2019). An alternative to cognitive consistency theory is the Yale model of attitude change, which was based on the dominant reinforcement learning paradigm of the 1940s and 1950s. According to the model, attitude change was a function of the degree to which the arguments and position of a message source rewarded the message recipient to change their attitudes in the direction of the message (McGuire, 1986). Although scholars in this tradition never documented that reinforcements underlie successful attitude change, they proposed a programmatic and practical perspective on attitude change that identified the key elements of a persuasive message that remain popular to this day: the source (who delivers the message), the content (what information the message provides), the audience (to whom the message is addressed), and the effect (changes in the recipient’s attitudes and behavior as a consequence of receiving the message). The Yale group systematically manipulated such features and examined the effects, although these were somewhat “wandering” in focus (McGuire, 1986) and the studies’ effects were more variable than desirable (Eagly & Chaiken, 1993; Petty & Cacioppo, 1986). An alternative approach arising from the psychoanalytic perspective is offered by social judgment theory (Sherif & Cantril, 1947). Attitudes were conceptualized as guiding judgments of contemporaneous materials such as arguments received. This view is consistent with more recent social psychological theories, especially theories of emotion (e.g., Barrett, 2017; Corlett & Marrouch, 2019). In short, individuals’ developmental history – their past interactions with peers,

groups, family, and other social groups – results in attitudes that respond predictably with new stimuli such as counter-attitudinal information. With enough interactions, relatively narrow latitudes of acceptance and relatively wide latitudes of rejection routinely develop, consistent with what Johnson and Eagly (1989) labeled value-relevant involvement. Such attitudes are relatively strong and thus resistant to change in response to counterattitudinal information. Meanwhile, those without such consistent experiences are left with wide latitudes of acceptance (and narrow latitudes of rejection) and are more susceptible to change. Contemporary approaches to attitude change focus more directly on the cognitive processes that underlie attitude change. Leading approaches include the elaboration likelihood model (Petty & Cacioppo, 1986), which has formed the basis of many social cognitive approaches to attitudes and attitude change, the heuristic-systematic model (Chaiken, 1980), the unimodel (Kruglanski & Thompson, 1999), and the cognition in persuasion model (Albarracín, 2002). Each model explicitly addresses the impact of capacity and motivation to process information, the notion of thought-mediated persuasion, and the potentially moderating effects of personality variables. For example, the elaboration likelihood model predicts that message content (e.g., argument strength) will drive attitudes under conditions when people are motivated and able to think about the message (cf. high elaboration, such as high personal relevance, high need for cognition); thus, for example, strong arguments will prove more persuasive than weak arguments for people with high personal relevance (or high need for cognition, high intelligence, etc.). When people are neither able nor motivated to focus on the message (i.e., low elaboration), other factors will drive attitudes, if there is any change at all. Thus, communicator credibility or attractiveness are more likely to move attitudes under these circumstances. The focus on two primary processes, one controlled and deliberate and the other more impulsive, casts

Attitudes and Persuasive Communication Interventions

the elaboration likelihood model as well as other contemporary approaches such as dual-process models of persuasion. These process models have been able to account for the impact of a significant portion of the effects that early persuasion researchers obtained. For example, the elaboration likelihood model and heuristic-systematic model in particular have guided numerous studies outlining specific effects of message-relevant thinking, source expertise, and attractiveness, targeting issues of high personal relevance and implications for behavior. For example, when attitudes change in response to deliberate thinking about the message (e.g., because it contains strong arguments), such attitudes are relatively enduring and better predict subsequent behavior. In essence, therefore, these models effectively explain when direct appeals will succeed and when they will fail (O’Keefe, 2013). In addition to issues of cognitive responses to persuasive messages, process models are able to account for more affectively based processes such as mood and attraction to the source and recognize that stimuli outside conscious awareness can impact attitudes and behaviors (for a

449

detailed comparison of these models’ predictions, see Johnson, Wolf et al., 2019).

31.4 Behavior Change Methods and Evidence Base A number of behavior change methods or techniques identified in the literature have been adopted in interventions aimed at changing people’s behavior by changing attitudes. Arguably, the most common methods include information provision (e.g., information about antecedents, information about health consequences, pros and cons) and communication-persuasion (e.g., credible source, framing/reframing; Johnson, Wolf et al. 2019; Steinmetz et al., 2016). Table 31.1 presents examples of behavior change techniques that are most closely linked to attitude change. Consistent with basic process models that outline mechanisms of action (see Chapters 20 and 46, this volume), these techniques are expected to change behavior by changing attitudes and associated processes (e.g., cognitive dissonance), so attitudes are central to the mechanism of action for the effect of these techniques on behavior change.

Table 31.1 Example techniques for changing attitudes from the behavior change technique taxonomy v1 (Michie et al., 2013) and related descriptions Cluster and Technique

Description

Cluster: 4. Shaping Knowledge Provide information about antecedents that reliably predict performance of Technique: 4.2. Information the behavior. about antecedents Example: Advise to keep a record of snacking and of situations or events occurring prior to snacking. Provide information (e.g., written, verbal, visual) about health Cluster: 5. Neutral consequences of performing the behavior. Consequences Example: Present the likelihood of contracting a sexually transmitted Technique: 5.1. Information infection following unprotected sexual behavior. about health consequences Use methods specifically designed to emphasize the consequences of Cluster: 5. Neutral performing the behavior with the aim of making them more memorable Consequences (goes beyond informing about consequences). Technique: 5.2. Salience of Example: Produce cigarette packets showing pictures of health consequences consequences, e.g., diseased lungs, to highlight the dangers of continuing to smoke. Continued

450

KYRA HAMILTON AND BLAIR T. JOHNSON

Table 31.1 (cont.) Cluster and Technique

Description

Cluster: 5. Neutral Consequences Technique: 5.3. Information about social and environmental consequences Cluster: 5. Neutral Consequences Technique: 5.6. Information about emotional consequences Cluster: 7. Associations Technique: 7.2 Associative learning

Provide information (e.g., written, verbal, visual) about social and environmental consequences of performing the behavior. Example: Tell family physician about financial remuneration for conducting health screening. Provide information (e.g., written, verbal, visual) about emotional consequences of performing the behavior. Example: Explain that quitting smoking increases happiness and life satisfaction.

Present a neutral stimulus jointly with a stimulus that already elicits the behavior repeatedly until the neutral stimulus elicits that behavior (includes “Classical/Pavlovian conditioning”). Example: Repeatedly present fatty foods with a disliked sauce to discourage the consumption of fatty foods. Present verbal or visual communication from a credible source in favor of, Cluster: 9. Comparison of or against, the behavior. Outcomes Technique: 9.1 Credible source Example: Present a speech given by a high-status professional to emphasize the importance of not exposing patients to unnecessary radiation by ordering x-rays for back pain. Advise the person to identify and compare reasons for wanting (pros) and Cluster: 9. Comparison of not wanting to (cons) change the behavior (includes “Decisional Outcomes balance”). Technique: 9.2 Pros and cons Example: Advise the person to list and compare the advantages and disadvantages of prescribing antibiotics for upper respiratory tract infections. Prompt or advise the imagining and comparing of future outcomes of Cluster: 9. Comparison of changed versus unchanged behavior. Outcomes Example: Prompt the person to imagine and compare likely or possible Technique: 9.3 Comparative outcomes following attending versus not attending a screening imagining of future outcomes appointment. Cluster: 13. Identity Suggest the deliberate adoption of a perspective or new perspective Technique: 13.2 Framing/ on behavior (e.g., its purpose) in order to change cognitions or reframing emotions about performing the behavior (includes “Cognitive structuring”). Example: Suggest that the person might think of the tasks as reducing sedentary behavior (rather than increasing activity). Draw attention to discrepancies between current or past behavior and Cluster: 13. Identity self-image, in order to create discomfort (includes “Cognitive Technique: 13.3 Incompatible dissonance”). beliefs Example: Draw attention to a doctor’s liberal use of blood transfusion and their self-identification as a proponent of evidence-based medical practice.

Attitudes and Persuasive Communication Interventions

31.4.1 Changing Attitudes Using Information Provision and Communication-Persuasion Research adopting information provision as a method to change attitudes has typically presented a “case” for the behavior, providing information in favor of the positive outcomes for performing the behavior and that these outweigh any negative outcomes. Social cognitive theories, such as the theory of planned behavior (see Chapter 2, this volume), specify that attitudes toward the behavior are thought to be determined from the individual’s salient beliefs about the likely outcomes of the behavior (behavioral beliefs) and the evaluations, either positive or negative, of these outcomes (outcome evaluation). Although individuals’ salient behavioral beliefs are thought to determine their attitude toward the behavior (Ajzen, 1991), and may yield information of potential value for designing intervention programs (Steadman, Rutter, & Field, 2002), in practice, however, it is more convenient to identify the set of most frequently cited or modal beliefs relevant to the population whose behavior needs changing (Ajzen, 2011; Ajzen & Fishbein, 1980). These modal salient beliefs can be identified through conducting an elicitation study in a representative sample of the target population using a series of open-ended questions (Ajzen, 2011) and conducting a thematic and/or content analysis (Braun & Clarke, 2013) on the data (see Chan et al., 2015; Cowie & Hamilton, 2014; de Leeuw et al., 2015; Hamilton et al., 2012; Spinks & Hamilton, 2015). Interventions to change attitudes can then base message content on these modal beliefs and deliver them using print, online, or verbal communications (Fishbein & Cappella, 2006; French & Cooke, 2012). For samples of intervention material using information provision, see Appendix 31.3 (supplemental materials). Recent meta-analytic research has demonstrated that interventions targeting attitudes lead

451

to changes in intentions and behavior with small and medium effect sizes, respectively (Sheeran et al., 2016; Steinmetz et al., 2016). More specifically, research in a range of behaviors such as physical activity (Chatzisarantis & Hagger, 2005), sun safety (White et al., 2019), drowning prevention (Hamilton, Peden et al., 2018), and environmental behaviors (Buttlar & Walther, 2018) have demonstrated that interventions targeting salient beliefs lead to significant changes in attitudes, intentions, and behavior. More importantly, research has demonstrated that intervention effects occur through attitude change (Chatzisarantis & Hagger, 2005). Specifically, measures of attitudes have been shown to mediate attitude-based intervention effects with indirect effects of the intervention on physical activity through attitude change (Chatzisarantis & Hagger, 2005). Chatzisarantis and Hagger’s design included communication targeting nonsalient beliefs given to a control group, which provided evidence that targeting modal salient beliefs evokes the change rather than the mere presence of the communication (see Appendix 31.3.1, supplemental materials). In delivering information content, the literature offers advice on methods to enhance the likelihood that arguments presented are more persuasive. Message elaboration, for example, is shown to generate stronger attitudes and are more likely to affect behavior. This reasoning is based on Petty and Cacioppo’s (1986) elaboration likelihood model (see Section 31.3 for details), where individuals’ thoughts align with message content through effortful thinking (parallel to systematic processing in Chaiken’s (1980) heuristic-systematic model). However, evidence is less clear on whether such effects exist when attitudes are already strong or polarized (Howe & Krosnick, 2017; Johnson & Eagly, 1989). In addition to message elaboration, experimental research has demonstrated that strong arguments are more persuasive than weak arguments (Johnson & Eagly, 1989; O’Keefe, 2013). A

452

KYRA HAMILTON AND BLAIR T. JOHNSON

suggested reason for this observed effect is that strong messages often present consequences that are more positive for the message recipients (Johnson et al., 2004). However, it should be noted that effects are greater when motivation to process the message is high. For example, Johnson and Eagly’s (1989) meta-analysis showed that (1) with value-relevant involvement (i.e., the psychological state that is created by the activation of attitudes that are linked to important values), high-involvement subjects were less persuaded than low-involvement subjects; (2) with outcome-relevant involvement (i.e., be accurate), high-involvement subjects were more persuaded than low-involvement subjects by strong arguments and (somewhat inconsistently) less persuaded by weak arguments; and (3) with impression-relevant involvement (i.e., accommodate social pressures), high-involvement subjects were slightly less persuaded than low-involvement subjects. For samples of intervention material using strong and weak arguments, see Appendix 31.4 (supplemental materials). Source cue effects are also shown to be influential on message delivery and assessments of credibility and, thus, attitude change. The idea is that communicators are more persuasive when they possess desirable characteristics which, in general, are suggested to fall under two dimensions: epistemic bases, such as competence, knowledge, or expertise, and social bases, such as benevolence, motivation, warmth, helpfulness, or trustworthiness (Johnson, Wolf et al., 2019; Wilson & Sherrell, 1993). Experimental research has shown that social characteristics over epistemic characteristics are considered more credible and persuasive (Van Kleef, van den Berg, & Heerdink, 2015). However, although a source may be considered credible, assessments of high credibility over low credibility can be arbitrarily determined. For example, assessments of expertise and trustworthiness have been shown to be based on cues such as celebrity status (Lee & Thorson, 2008) and attractiveness (Patzer,

1983). Overall, attitude-based intervention research using information provision and communication-persuasion has been shown to be moderately successful in changing individuals’ attitudes and, more importantly, their intentions and actual behavior.

31.4.2 An Example Step-by-Step Guide to Using Information Provision and Communication-Persuasion to Change Attitudes Applications of interventions using information provision and communication-persuasion for attitude change are broad and include using print, online, mass media, group, or individual communications. It is beyond the scope of this chapter to review all of these different approaches. An example based on targeting the salient behavioral beliefs that underpin attitudes will serve as an illustration (see Ajzen, 2019a, 2019b, 2011). This example is consistent with the information provision approach outlined in Section 31.4.1, wherein the intervention comprises an attitudebased persuasive communication that presents a “case” for the target behavior that highlights its benefits and negates its detriments. Consistent with dual-process models, these kinds of communication are likely to be relatively complex so will likely necessitate deep, elaborate processing. As a consequence, the message should be delivered by methods where such processing is more likely, such as through a pamphlet or leaflet, an internet or social media passage, or even by verbal communication. Prior to designing any intervention, including attitude-based interventions, it is fundamental that the target behavior is well-specified (see also Chapters 19 and 21, this volume). According to Ajzen and Fishbein (1980), any behavior can be defined according to four elements: target at which the action is directed, the action involved, the context in which it occurs, and the time frame of occurrence (TACT). No

Attitudes and Persuasive Communication Interventions

matter how the four elements are defined in terms of levels of specificity or generality, when constructing attitude measures, it is important to observe the principle of compatibility, that is, attitudes should correspond to the behavior in all four elements (see Chapter 2, this volume; see also Siegel et al., 2014). For example, to assess parents’ sun-protective behavior for their young children, Hamilton and colleagues’ (2017) measure of attitudes followed the TACT and compatibility principles: “performing sunprotective behaviors (action) for my child (target) every time they go in the sun for more than 10 minutes (context) during the next 2 weeks (time frame)” (see Appendix 31.3.1, supplemental materials). This approach is consistent with a previous meta-analysis (Kraus, 1995), which demonstrated that, when the principle of compatibility was followed, the average attitude-behavior correlation was larger (r = 0.50) compared to when it was not (r = 0.14). It is also important before designing attitudebased messages that preliminary formative work is conducted on a representative sample of the target population to elicit the relevant beliefs to be targeted in the attitude-based communication. This is because behavioral beliefs, and therefore attitudes, may vary among diverse population groups. One way to do this is to conduct an elicitation study, using interview or survey techniques, and using open-ended response questions. For example, in a study on avoiding driving through floodwater, Hamilton, Price et al. (2018) used a purposive sampling method to recruit participants from the target population: Australian adults who held a registered driver’s license and met the experience criteria of having decided to avoid driving through floodwater. The behavioral beliefs for the target behavior were elicited using a series of three open-ended questions: “Please list what you believe are the advantages of avoiding driving through floodwater? Please list what you believe are the disadvantages of avoiding driving through floodwater? Please

453

list anything else you associate with avoiding driving through floodwater?” (see Appendix 31.1.3, supplemental materials). The beliefs elicited are then analyzed using thematic or content analysis (see Braun & Clarke, 2013) to identify a list of key modal salient beliefs that can be used as targets in future behavior change interventions. Once the key beliefs have been identified through the elicitation process, the beliefs that are to be targeted in the persuasive communication in the intervention need to be identified. This decision is usually based on the modal (most frequently cited) beliefs identified in the elicitation procedure. For a discussion of the decision rules that relate to model belief selection, the reader is referred to Ajzen’s (2011) treatment of the matter (see also Chapter 2, this volume). The beliefs could also be based on formative research on the extent to which the belief predict behavior. For example, Hamilton et al. (2016) found the behavioral beliefs of “reach my destination,” “sustain vehicle damage,” and “be swept away” to predict individuals’ willingness to drive through lower risk levels of floodwater. These beliefs were then targeted in persuasive communication delivered in the form of a video infographic (Hamilton, Peden et al., 2018). The message content included presenting statistics associated with driving through floodwater, providing information about the uncertainty of conditions when water is covering the road, and giving specific information about the effect of floodwater on vehicles (see Appendix 31.3.2, supplemental materials). Evaluation of changes in attitudes were assessed pre- and postintervention. Results showed evidence of changes in attitudes toward driving though floodwater after watching the video infographic, and the change was maintained over time, although only for females. Although using information provision and communication-persuasion techniques shows promise in changing behavior through changing attitudes, the modest size of attitude-behavior

454

KYRA HAMILTON AND BLAIR T. JOHNSON

relations suggests that the overall effects of attitude change interventions on behavior may be relatively small (Glasman & Albarracín, 2006). Thus, a wider set of considerations (e.g., audience targeting, message framing, message tailoring, developing evidence-based content, using graphics and text effectively) may need to be considered when implementing such approaches. This suggests that other techniques in addition to those focused on changing attitudes through information provision and communicationpersuasion are needed for effective and longterm behavior change (see Johnson, Landrum, & McCloskey, 2019; Rothman & Salovey, 1997).

31.4.3 Changing Attitudes Through Cognitive Dissonance An alternative approach to changing behavior through attitude change is to induce cognitive dissonance (Festinger, 1962; Freijy & Kothe, 2013; for details, see Section 31.3). Interventions based on cognitive dissonance theory can be derived from several experimental paradigms (e.g., belief-disconfirmation paradigm, free-choice paradigm, effort justification paradigm, induced compliance paradigm, hypocrisy paradigm) and often involve inducing a state of discrepancy between an individual’s beliefs and their current behavior (for an overview, see Freijy & Kothe, 2013). The extant literature provides support for the efficacy of dissonance-based behavior change interventions across a range of health, social, and environmental behaviors. The most compelling evidence, however, comes from systematic review and meta-analytic research in the areas of disordered eating (Stice et al., 2008) and health behaviors more generally (Freijy & Kothe, 2013; Stone & Focella, 2011). For example, in their meta-analysis of twenty dissonancebased interventions, Freijy and Kothe found that the majority of included studies reported positive changes in one or more of participants’ health behavior, attitude, or intention. However, the

authors expressed caution about the findings, given that the majority of studies used self-report measures of behavior, which may have introduced measurement error, study quality was generally poor, and long-term effects were difficult to determine. Thus, it is unknown if basing interventions on dissonance theory alone may lead to lasting behavior change but evidence, in general, supports its use as a key strategy and it features prominently in behavior change technique taxonomies.

31.4.4 An Example Step-by-Step Guide to Using Cognitive Dissonance to Change Attitudes Applications using cognitive dissonance for attitude change, similar to applications using information provision and communicationpersuasion, draw from a broad range of experimental paradigms (see Freijy & Kothe, 2013; Hagger, 2019) and have been delivered at the individual (e.g., client-practitioner contexts) and population (e.g., mass media campaigns) levels. Details of an example, group-based intervention based on cognitive dissonance will serve to illustrate the typical protocol and materials used in this approach to attitude and behavior change (Chatzisarantis, Hagger, & Wang, 2008). Participation in regular physical activity, especially performing more formal forms of exercise like doing a gym workout, is usually perceived as health promoting but may also be perceived as effortful, mundane, and uncomfortable. As such, dissonance may be experienced between holding beliefs about the health benefits of the gym workout and the effort and discomfort involved. To resolve this dissonance, the individual may make justifications that align with their beliefs about the gym workout – “despite feeling sore, this workout is good for my health” – and, thus, continue to go to the gym. Conversely, the individual may change their behavior so that it is more congruent with their beliefs about the gym workout

Attitudes and Persuasive Communication Interventions

and, thus, desist from going to the gym. For these reasons, simply communicating the consequences of performing a behavior, as is often done when using information provision, may fail to change behavior as individuals might start to make justifications for their inaction – “I can’t go to the gym today as getting this assignment finished is more important than doing a workout.” One means to reduce the felt dissonance is to present the activity as one that was freely chosen by the individual – a free-choice paradigm (see Chapter 8, this volume). For example, Chatzisarantis et al. (2008) presented participants with a video of a person performing a benchstepping task, rated boring by participants in a pilot study. They were then informed that the activity was health promoting and asked to practice this task for twenty minutes at a time, three times per week for the next three weeks. Participants assigned to a free-choice, cognitive dissonance group were informed that they could choose whether or not to participate in the task in the future: Now, you are to make a decision about whether or not to practice bench/chair stepping, for 3 days per week, for at least 20 minutes each time, over the next three weeks during your leisure time. The choice is up to you.

Participants in the forced-choice or no dissonance group were informed that they had to perform the task: Now you do not have much choice and you should practice bench/chair stepping, 3 days per week, for at least 20 minutes each time, over the next three weeks, during your leisure time.

A no-choice, control group received information about the task only and did not receive any instructions about choice. Participants were then asked to self-report their attitudes to perform the behavior and their perceived choice and mental frustration toward the task. Results indicated that

455

participants in the free-choice condition were more likely to report having choice and experience dissonance indicated by their levels of frustration relative to participants in the forcedchoice and no-choice conditions. Critically, participants in the free-choice condition expressed more positive attitudes toward the task compared to participants in the other conditions. Consistent with cognitive dissonance theory, participants in the free-choice paradigm altered their attitudes toward the task to be more positive because they justified the fact that they had freely chosen to do the boring task. Those in the forced-choice and no-choice conditions did not because they did not perceive they had such choice and were therefore not freely acting against their attitudes. This study illustrates how the provision of choice can be a powerful means to change attitudes and behavior. Such processes can be used in many other contexts, for example where people are prompted to choose to perform a particular behavior such as donating blood or making more environmentally friendly choices. For samples of intervention material using cognitive dissonance, see Appendix 31.5 (supplemental materials).

31.4.5 Other Attitude-Based Methods of Behavior Change Other common methods in the attitude change research include the use of techniques such as repeated exposure and direct experience. The former is concerned with the observation that repeated exposure to a stimulus generates more positive attitudes toward the stimulus – known as the mere exposure effect (Zajonc, 1965). The latter is based on the observation that beliefs formed via direct exposure (e.g., seeing friends’ behaviors) are more marked than those acquired indirectly and that attitudes based on these beliefs are, in turn, more predictive of intentions and behavior (Wyer & Albarracín, 2005). Meta-analytic research has demonstrated support for both these effects (Glasman & Albarracín, 2006;

456

KYRA HAMILTON AND BLAIR T. JOHNSON

Montoya et al., 2017) but with caveats. For example, mere exposure effects positively influence attitudes up to a point; for example, 25–28 exposures for photographs, and approximately 73 exposures for ideographs, is sufficient to induce attitude change but thereafter no additional effects are observed. In addition, the strength of effects is different for adults and children (Montoya et al., 2017). Further, affective processes that are experienced immediately may play a role in creating more valenced attitudes (see Clore & Schnall, 2019) and being exposed to inconsistent information (e.g., two-sided messages) may result in attitudes being less predictive of behaviors (Glasman & Albarracín, 2006). Other emerging methods in the attitude change research include leveraging implicit cognitions to change behavior (see Gawronski & Brannon, 2019). For example, in a comparative investigation of seventeen interventions to reduce implicit racial preferences, Lai et al. (2014) found that commonly accepted intervention strategies such as perspective-taking, appeals to egalitarian values, or induced positive emotions were largely ineffective. The most potent interventions were those that invoked high self-involvement or involved various ways of linking the relevant target groups with positivity or negativity, such as evaluative conditioning or mental simulation of counter-intentional exemplars. More recent research has shown that implicit evaluations can change in response to a single piece of novel, counter-intentional information (Cone & Ferguson, 2015) and information that suggests a reinterpretation of earlier information (Mann & Ferguson, 2017) and that the observed changes in implicit evaluations can generalize across contexts. Taken together, the current literature suggests that implicit preferences are potentially malleable, at least in the short term (Gawronski, 2019). However, there is a paucity of applied evidence to guide best-practice principles in changing implicit attitudes, and short-term malleability in implicit attitudes may not lead to long-

term change (Lai et al., 2014). Thus, there are questions about the stability of implicit attitudes and future research should continue to investigate the durability of intervention effects aimed at changing implicit attitudes as well as any lasting effects on behavior change.

31.5 Conclusions Interest in behavior change interventions based on attitude change is predicated on the widely held belief that attitudes should guide behavior. However, evidence on the link between attitudes and behavior has suggested that the association is complex and dependent on a number of factors and processes. While, in general, meta-analytic research supports a positive impact of behavior change interventions on attitudes, what is less clear is the impact of individual methods on attitude change. Moreover, interventionists attempting to change attitudes often draw from theories of social cognition and behavior change rather than more formal persuasion models. A challenge for the future for attitude scholars is not only to show that their specific models add to the understanding of persuasion over and above more general frameworks of behavior but also to provide stronger, more robust evidence for the main and interactive effects of isolated behavior change methods aimed at changing behavior by changing attitudes and the proposed mechanisms for the effects.

References Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50, 179–211. https://doi.org/10.1016/ 0749-5978(91)90020-T Ajzen, I. (2011). Behavioral interventions: Design and evaluation guided by the theory of planned behavior. In M. M. Mark, S. I. Donaldson, & B. C. Campbell (Eds.), Social Psychology for Program and Policy Evaluation. New York: Guilford Press.

Attitudes and Persuasive Communication Interventions

Ajzen, I. (2019a). Constructing a Theory of Planned Behavior Questionnaire. https://people.umass .edu/~aizen/pdf/tpb.measurement.pdf Ajzen, I. (2019b). Behavioral Interventions Based on the Theory of Planned Behavior. https://people .umass.edu/~aizen/pdf/tpb.intervention.pdf Ajzen, I., & Fishbein, M. (1980). Understanding Attitudes and Predicting Social Behavior. Upper Saddle River, NJ: Prentice Hall. Albarracín, D. (2002). Cognition in persuasion: An analysis of information processing in response to persuasive communications. Advances in Experimental Social Psychology, 34, 61–130. https://doi.org/10.1016/S0065-2601(02)80004-1 Barrett, L. F. (2017). How Emotions Are Made: The Secret Life of the Brain. New York: Houghton Mifflin Harcourt. Braun, V., & Clarke, V. (2013). Successful Qualitative Research: A Practical Guide for Beginners. London: SAGE. Breckler, S. J. (1984). Empirical validation of affect, behavior, and cognition as distinct components of attitude. Journal of Personality and Social Psychology, 47, 1191–1205. https://doi.org/ 10.1037/0022-3514.47.6.1191 Buttlar, B., & Walther, E. (2018). Measuring the meat paradox: How ambivalence towards meat influences moral disengagement. Appetite, 128, 152–158. https://doi.org/10.1016/j.appet.2018.06.011 Cacioppo, J. T., & Berntson, G. G. (1994). Relationship between attitudes and evaluative space: A critical review, with emphasis on the separability of positive and negative substrates. Psychological Bulletin, 115, 401–423. https://doi.org/10.1037/ 0033-2909.115.3.401 Chaiken, S. (1980). Heuristic versus systematic information processing and the use of source versus message cues in persuasion. Journal of Personality and Social Psychology, 39, 752–766. https://doi.org/10.1037/0022-3514.39.5.752 Chan, D. K. C., Hardcastle, S., Dimmock, J. A. et al. (2015). Modal salient belief and social cognitive variables of anti-doping behaviors in sport: Examining an extended model of the theory of planned behavior. Psychology of Sport and Exercise, 16, 164–174. https://doi.org/10.1016/ j.psychsport.2014.03.002

457

Chatzisarantis, N. L. D., & Hagger, M. S. (2005). Effects of a brief intervention based on the theory of planned behavior on leisure time physical activity participation. Journal of Sport and Exercise Psychology, 27, 470–487. https://doi.org/ 10.1123/jsep.27.4.470 Chatzisarantis, N. L. D., Hagger, M. S., & Wang, C. K. J. (2008). An experimental test of cognitive dissonance theory in the domain of physical exercise. Journal of Applied Sport Psychology, 20, 97–115. https://doi.org/10.1080/ 10413200701601482 Clore, G. L., & Schnall, S. (2019). The influence of affect on attitude. In D. Albarracín & B. T. Johnson (Eds.), The Handbook of Attitudes, Vol. 1 (2nd ed., pp. 259–290). New York: Routledge. Cone, J., & Ferguson, M. J. (2015). He did what? The role of diagnosticity in revising implicit evaluations. Journal of Personality and Social Psychology, 108, 37–57. https://doi.org/10.1037/ pspa0000014 Corlett, P. R., & Marrouch, N. (2019). Social cognitive neuroscience of attitudes and beliefs. In D. Albarracín & B. T. Johnson (Eds.), The Handbook of Attitudes, Vol. 1 (2nd ed., pp. 480–519). New York: Routledge. Cowie, E., & Hamilton, K. (2014). Key beliefs related to decisions for physical activity engagement among first-in-family students transitioning to university. Journal of Community Health, 39, 719–726. https://doi.org/10.1007/s10900-0139817-3 Croyle, R. T., & Cooper, J. (1983). Dissonance arousal: Physiological evidence. Journal of Personality and Social Psychology, 45, 782–791. https://doi .org/10.1037/0022-3514.45.4.782 de Leeuw, A., Valois, P., Ajzen, I., & Schmidt, P. (2015). Using the theory of planned behavior to identify key beliefs underlying pro-environmental behavior in high-school students: Implications for educational interventions. Journal of Environmental Psychology, 42, 128–138. https:// doi.org/10.1016/j.jenvp.2015.03.005 Dovidio, J. F., & Gaertner, S. L. (2000). Aversive racism and selection decisions: 1989 and 1999. Psychological Science, 11, 315–319. https://doi .org/10.1111/1467-9280.00262

458

KYRA HAMILTON AND BLAIR T. JOHNSON

Eagly, A. H., & Chaiken, S. (1993). The Psychology of Attitudes. New York: Harcourt Brace Jovanovich. Elliot, A. J., & Devine, P. G. (1994). On the motivational nature of cognitive dissonance: Dissonance as psychological discomfort. Journal of Personality and Social Psychology, 67, 382– 394. https://doi.org/10.1037/0022-3514.67.3.382 Fabrigar, L. R., MacDonald, T. K., & Wegener, D. T. (2019). The origins and structure of attitudes. In D. Albarracín & B. T. Johnson (Eds.), The Handbook of Attitudes, Vol. 1 (2nd ed., pp. 109– 157). New York: Routledge. Festinger, L. (1962). A Theory of Cognitive Dissonance. Stanford: Stanford University Press. Festinger, L. (1964). Conflict, Decision, and Dissonance. Stanford: Stanford University Press. Fishbein, M., & Cappella, J. N. (2006). The role of theory in developing effective health communications. Journal of Communication, 56, S1–S17. https://doi.org/10.1111/j.14602466.2006.00280.x Freijy, T., & Kothe, E. J. (2013). Dissonance-based interventions for health behaviour change: A systematic review. British Journal of Health Psychology, 18, 310–337. https://doi.org/10.1111/ bjhp.12035 French, D. P., & Cooke, R. (2012). Using the theory of planned behaviour to understand binge drinking: The importance of beliefs for developing interventions. British Journal of Health Psychology, 17, 1–17. https://doi.org/10.1111/ j.2044-8287.2010.02010.x. Gawronski, B. (2019). Six lessons for a cogent science of implicit bias and its criticism. Perspectives on Psychological Science, 14, 574–595. https://doi .org/10.1177/1745691619826015 Gawronski, B., & Brannon, S. M. (2019). Attitudes and the implicit-explicit dualism. In D. Albarracín & B. T. Johnson (Eds.), The Handbook of Attitudes, Vol. 1 (2nd ed., pp. 158–196). New York: Routledge. Glasman, L. R., & Albarracín, D. (2006). Forming attitudes that predict future behavior: A metaanalysis of the attitude-behavior relation. Psychological Bulletin, 132, 778–822. https://doi .org/10.1037/0033-2909.132.5.778

Greenwald, A. G., McGhee, D. E., & Schwartz, J. L. (1998). Measuring individual differences in implicit cognition: The implicit association test. Journal of Personality and Social Psychology, 74, 1464–1480. https://doi.org/10.1037/00223514.74.6.1464 Greenwald, A. G., Poehlman, T. A., Uhlmann, E. L., & Banaji, M. R. (2009). Understanding and using the Implicit Association Test: III. Meta-analysis of predictive validity. Journal of Personality and Social Psychology, 97, 17–41. https://doi.org/ 10.1037/a0015575 Hagger, M. S. (2019). The role of attitudes in physical activity. In D. Albarracín & B. T. Johnson (Eds.), The Handbook of Attitudes, Vol. 2 (2nd ed., pp. 92–120). New York: Routledge. Hamilton, K., Kirkpatrick, A., Rebar, A., & Hagger, M. S. (2017). Child sun safety: Application of an integrated behavior change model. Health Psychology, 36, 916–926. https://doi.org/10.1037/ hea0000533. Hamilton, K., Peden, A., Keech, J. J., & Hagger, M. S. (2018). Changing people’s attitudes and beliefs toward driving through floodwaters: Evaluation of a video infographic. Transportation Research Part F: Traffic Psychology and Behaviour, 53, 50–60. https://doi.org/10.1016/j.trf.2017.12.012. Hamilton, K., Peden, A. E., Pearson, M., & Hagger, M. S. (2016). Stop there’s water on the road! Identifying key beliefs guiding people’s willingness to drive through flooded waterways. Safety Science, 86, 308–314. https://doi.org/ 10.1016/j.ssci.2016.07.004 Hamilton, K., Price, S., Keech, J., Peden, A., & Hagger, M. S. (2018). Drivers’ experiences during floods: Investigating the psychological influences underpinning decisions to avoid driving through floodwater. International Journal of Disaster Risk Reduction, 28, 507–518. https://doi.org/10.1016/ j.ijdrr.2017.12.013 Hamilton, K., White, K. M., Young, R., Hawkes, A., Starfelt, L. C., & Leske, S. (2012). Identifying critical sun-protective beliefs among Australian adults. Health Education Research, 27, 834–843. https://doi.org/10.1093/her/cys093 Harmon-Jones, E., Armstrong, J., & Olson, J. M. (2019). The influence of behavior on attitudes. In

Attitudes and Persuasive Communication Interventions

D. Albarracín & B. T. Johnson (Eds.), The handbook of Attitudes, Vol. 1 (2nd ed., pp. 404– 449). New York: Routledge. Harmon-Jones, E., Brehm, J. W., Greenberg, J., Simon, L., & Nelson, D. E. (1996). Evidence that the production of aversive consequences is not necessary to create cognitive dissonance. Journal of Personality and Social Psychology, 70, 5–16. https://doi .org/10.1037/0022-3514.70.1.5 Heider, F. (1958). The Psychology of Interpersonal Relations. New York: John Wiley & Sons. Howe, L. C., & Krosnick, J. A. (2017). Attitude strength. Annual Review of Psychology, 68, 327–351. https://doi.org/10.1146/annurevpsych-122414-033600 Johnson, B. T., & Eagly, A. H. (1989). Effects of involvement on persuasion: A meta-analysis. Psychological Bulletin, 106, 290–314. https://doi .org/10.1037/0033-2909.106.2.290 Johnson, B. T., Landrum, A., & McCloskey, K. (2019). Attitude scholarship in the 21st century: Accomplishments, challenges, and gaps. In D. Albarracín & B. T. Johnson (Eds.), The Handbook of Attitudes, Vol. 1 (2nd ed., pp. 625–649). New York: Routledge. Johnson, B. T., Smith-McLallen, A., Killeya, L.A., & Levin, K. D. (2004). Truth or consequences: Overcoming resistance to persuasion with positive thinking. In E. S. Knowles & J. Linn (Eds.), Resistance and Persuasion (pp. 215–233). Mahwah, NJ: Lawrence Erlbaum Associates. Johnson, B. T., Wolf, L., Maio, G., & Smith-McLallen, A. (2019). Communication-induced persuasion or resistance: Processes and effects of who says what to whom. In B. T. Johnson & D. Albarracín (Eds.), The Handbook of Attitudes, Vol. 1 (2nd ed., pp. 557–601). New York: Routledge. Jonas, K., Broemer, P., & Diehl, M. (2000). Attitudinal ambivalence. European Review of Social Psychology, 11, 35–74. https://doi.org/10.1080/ 14792779943000125 Kraus, S. J. (1995). Attitudes and the prediction of behavior: A meta-analysis of the empirical literature. Personality and Social Psychology Bulletin, 21, 58–75. https://doi.org/10.1177/ 0146167295211007

459

Kruglanski, A., & Thompson, E. (1999). Persuasion by a single route: A view from the unimodel. Psychological Inquiry, 10, 83–109. https://doi .org/10.1207/S15327965PL100201 Lai, C. K., Marini, M., Lehr, S. A. et al. (2014). A comparative investigation of 17 interventions to reduce implicit racial preferences. Journal of Experimental Psychology: General, 143, 1765– 1785. https://doi.org/10.1037/a0036260 Lee, J., & Thorson, E. (2008). The impact of celebrityproduct incongruence on the effectiveness of the product endorsement. Journal of Advertising Research, 48, 433–449. https://doi.org/ 449.10.2501/S0021849 Mann, T., & Ferguson, M. J. (2017). Reversing implicit first impressions through reinterpretation after a two-day delay. Journal of Experimental Social Psychology, 68, 122–127. https://doi.org/10.1016/ j.jesp.2016.06.004 McGuire, W. J. (1986). The vicissitudes of attitudes and similar representational constructs in twentieth century psychology. European Journal of Social Psychology, 16, 89–130. https://doi.org/10.1002/ ejsp.2420160202 Michie, S., Richardson, M., Johnston, M. et al. (2013). The Behavior Change Technique Taxonomy (v1) of 93 hierarchically clustered techniques: Building an international consensus for the reporting of behavior change interventions. Annals of Behavioral Medicine, 46, 81–95. https:// doi.org/10.1007/s12160-013-9486-6 Montoya, R. M., Horton, R. S., Vevea, J. L., Citkowicz, M., & Lauber, E. A. (2017). A re-examination of the mere exposure effect: The influence of repeated exposure on recognition, familiarity, and liking. Psychological Bulletin, 143, 459–498. https://doi.org/10.1037/bul0000085 Nisbett, R. E., & Wilson, T. D. (1977). Telling more than we can know: Verbal reports on mental processes. Psychological Review, 84, 231–259. https://doi.org/10.1037//0033-295x.84.3.231 O’Keefe, D. J. (2013). The elaboration likelihood model. In J. P. Dillard & L. Shen (Eds.), The SAGE Handbook of Persuasion: Developments in Theory and Practice (pp. 137–149). London: SAGE. https://doi.org/10.1002/9781405186407 .wbieceOl 1.pub2

460

KYRA HAMILTON AND BLAIR T. JOHNSON

Oswald, F. L., Mitchell, G., Blanton, H., Jaccard, J., & Tetlock, P. E. (2013). Predicting ethnic and racial discrimination: A meta-analysis of IAT criterion studies. Journal of Personality and Social Psychology, 105, 171–192. https://doi.org/ 10.1037/a0032734 Patzer, G. L. (1983). Source credibility as a function of communicator physical attractiveness. Journal of Business Research, 11, 229–241. https://doi.org/ 10.1016/0148-2963(83)90030-9 Petty, R. E., & Cacioppo, J. T. (1986). The elaboration likelihood model of persuasion. In L. Berkowitz (Ed.), Advances in Experimental Social Psychology, Vol. 19 (pp. 123–205). Orlando, FL: Academic Press. Phipps, D. J., Hagger, M. S., & Hamilton, K. (2019). A meta-analysis of implicit and explicit attitudes in children and adolescents. PsyArXiv. https://doi .org/10.31234/osf.io/52jrs Ross, M. (1989). Relation of implicit theories to the construction of personal histories. Psychological Review, 96, 341–357. https://doi.org/10.1037/ 0033-295X.96.2.341 Rothman, A. J., & Salovey, P. (1997). Shaping perceptions to motivate healthy behavior: The role of message framing. Psychological Bulletin, 121, 3–19. Sheeran, P., Maki, A., Montanaro, E. et al. (2016). The impact of changing attitudes, norms, and self-efficacy on health-related intentions and behavior: A meta-analysis. Health Psychology, 35, 1178–1188. https://doi.org/10.1037/ hea0000387 Sherif, M., & Cantril, H. (1947). The Psychology of Ego-Involvements: Social Attitudes and Identifications. New York: Wiley. Siegel, J. Y., Navarro, M. A., Tan, C. N., & Hyde, M. K. (2014). Attitude–behavior consistency, the principle of compatibility, and organ donation: A classic innovation. Health Psychology, 33, 1084–1091. https://doi.org/ 10.1037/hea0000062 Spinks, T., & Hamilton, K. (2015). Investigating key beliefs guiding mothers’ dietary decisions for their 2–3 year old. Appetite, 89, 167–174. https://doi .org/10.1016/j.appet.2015.02.004.

Steadman, L., Rutter, D. R., & Field, S. (2002). Individually elicited versus modal normative beliefs in predicting attendance at breast screening: Examining the role of belief salience in the theory of planned behaviour. British Journal of Health Psychology, 7, 317–330. https://doi.org/ 10.1348/135910702760213706 Steinmetz, H., Knappstein, M., Ajzen, I., Schmidt, P., & Kabst, R. (2016). How effective are behavior change interventions based on the Theory of Planned Behavior? A three-level meta-analysis. Zeitschrift Fur Psychologie/Journal of Psychology, 224, 216–233. Stice, E., Shaw, H., Becker, C. B., & Rohde, P. (2008). Dissonance-based interventions for the prevention of eating disorders: Using persuasion principles to promote health. Prevention Science, 9, 114–128. https://doi.org/10.1007/s11121-008-0093-x. Stone, J., & Focella, E. (2011). Hypocrisy, dissonance and the self-regulation processes that improve health. Self and Identity, 10, 295–303. https://doi .org/10.1080/15298868.2010.538550 Van Kleef, G. A., van den Berg, H., & Heerdink, M. W. (2015). The persuasive power of emotions: Effects of emotional expressions on attitude formation and change. Journal of Applied Psychology, 100, 1124–1142. https://doi.org/10.1037/apl0000003 White, K. M., Zhao, X., Starfelt, L. C. et al. (2019). Effectiveness of a theory-based sun-safe behavioural change intervention among adolescents. Psychology-Oncology, 28, 505–510. https://doi.org/10.1002/pon.4967. Wilson, E. J., & Sherrell, D. L. (1993). Source effects in communication and persuasion research: A metaanalysis of effect size. Journal of the Academy of Marketing Science, 21, 101–111. https://doi.org/ 10.1007/BF02894421 Wyer, R. S., Jr., & Albarracín, D. (2005). Belief formation, organization, and change: Cognitive and motivational influences. In D. Albarracín, B. T. Johnson, & M. P. Zanna (Eds.), Handbook of Attitudes (pp. 273–322). New York: Psychology Press. Zajonc, R. B. (1965). Social facilitation. Science, 149, 269–274. https://doi.org/10.1126/ science.149.3681.269

32 Self-Efficacy Interventions Lisa M. Warner and David P. French

Practical Summary People rarely form an intention to try new behaviors or maintain them in the face of barriers unless they believe they can achieve something by their own actions. This belief in one’s own capabilities to successfully carry out actions is called self-efficacy. It can be prompted by giving people the chance to try new behaviors (mastery experience), by learning from others (vicarious experience), by being persuaded that they can do it (verbal persuasion), and by teaching them to control or reinterpret their nervousness before trying (affective and somatic states). This chapter defines the concept of self-efficacy, reviews research findings on its usefulness for behavior change, and elaborates on how to promote this belief for successful behavior change through concrete examples.

32.1 Introduction Self-efficacy can be broadly defined as individuals’ belief in their capability to implement a behavior needed to reach a goal or perform a task successfully. A large and growing amount of empirical research shows that self-efficacy is a key factor in predicting and explaining the successful initiation and maintenance of behavior change. This chapter focuses on how self-efficacy can be utilized to develop efficacious and effective professional-led and self-guided behavior change programs.

to accomplish. They do not necessarily reflect actual abilities, which often only moderately relate to people’s perceived skill level (for an overview on how self-efficacy differs from similar constructs, see Warner & Schwarzer, 2017). Outcome expectations reflect individuals’ evaluations that a particular action or behavior will result in desirable outcomes. Together, both sets of beliefs are important when it comes to determining future participation in a given behavior.

32.2.1 Measuring Self-Efficacy

32.2 Definition The construct of self-efficacy stems from Bandura’s (1986) social cognitive theory (see Chapter 3, this volume). This theory posits that two core sets of cognitions guide human behavior – self-efficacy expectations and outcome expectations. Self-efficacy beliefs reflect what people think they are able

Self-efficacy is best measured with previously validated self-report measures and psychometric scaling procedures, and measures are often domain specific. For example, in the health domain, self-efficacy is usually assessed for specific health behaviors such as safer sex selfhttps://doi.org/10.1017/9781108677318.032

462

LISA M. WARNER AND DAVID P. FRENCH

efficacy (e.g., “I feel confident in my ability to use a condom correctly”; Forsyth, Carey, & Fuqua, 1997). In the work context, measures of general work self-efficacy (e.g., “In my job, I am always able to perform well even when I’m particularly pressured for time”) or more specific forms (e.g., leadership self-efficacy, “I know what it takes to make a group accomplish its task”) are available (Chemers, Watson, & May, 2000; Magklara, Burton, & Morrison, 2014). In the academic context, self-efficacy can be assessed for various subjects (e.g., chemistry self-efficacy, “How well can you choose an appropriate formula to solve a chemistry problem?”; Capa-Aydin, Uzuntiryaki-Kondakci, & Ceylandag, 2018). Although Bandura originally conceived self-efficacy as a construct relating to individuals’ beliefs in their own capabilities, he later extended it to the level of collective beliefs (Bandura, 1993). Collective self-efficacy is seen as the perception of coordination of individual resources to a common potential (“When we set goals, I’m sure we will achieve them”; Bandura, 1986). Detailed guidelines on the construction of self-efficacy measures have been provided by Bandura (2006). In several studies, Williams and Rhodes (2016) showed that self-efficacy items might overlap with the assessment of motivation and outcome expectancies, especially if the studied behaviors are easy to enact (e.g., walking). To solve this issue, they recommended equipping self-efficacy items with the addition of “… if I wanted to” to distinguish them from motivational constructs.

32.3 Theory and Mechanisms of Change 32.3.1 Self-Efficacy Within Behavior Change Theories The self-efficacy construct originates from Bandura’s (1977a) theorizing, in the context of social learning theory (Bandura, 1977b) and then more fully in social cognitive theory (Bandura,

1986). The predictive and explanatory power in various domains of life prompted many authors to include self-efficacy in their behavior change theories as well – for example, the protection motivation theory (Maddux & Rogers, 1983; see Chapter 4, this volume), the revised health belief model (see Chapter 4, this volume), and the transtheoretical model (see Chapter 10, this volume). Perceived behavioral control – a related construct – was added to the theory of reasoned action (Fishbein & Ajzen, 1975), which was then renamed the theory of planned behavior (see Chapter 2, this volume). One model particularly emphasizes the role of selfefficacy in health behavior change by incorporating phase-specific self-efficacy beliefs: The health action process approach posits that individuals need action, maintenance, and recovery selfefficacy, depending on the phases of behavior change (see Chapter 7, this volume). Common among these theories is that self-efficacy is a key predictor of motivation or intention, the precursor of behavior, suggesting that the stronger individuals’ self-efficacy beliefs, the stronger their intentions and the more likely they were to perform the behavior. This also points to the centrality of the construct to multiple theories and its potential importance as a modifiable target for behavior change interventions.

32.3.2 Sources of Self-Efficacy Bandura claims four theoretical sources of selfefficacy: successfully having performed the task in the past (mastery experience), vicariously learning from observing others successfully performing a task (vicarious experience), being persuaded or convinced that one can perform an action (verbal persuasion), and reducing negative physiological and affective states that might be associated with hesitation to try the new behavior (affective and somatic states). This chapter reviews the sources of self-efficacy and links these with the behavior change techniques from current taxonomies (e.g., Kok et al., 2016; Michie et al., 2013).

Self-Efficacy Interventions

32.3.2.1 Mastery Experiences According to Bandura (1997), mastery experiences are “the most effective source of efficacy information because they provide the most authentic evidence of whether one can master whatever it takes to succeed” (p. 80). The most straightforward approach to building mastery experience is to provide opportunities to experience a new behavior, such as offering classes or workshops that directly train the necessary skills, such as exercise courses, girls’ days at universities to experience science, technology, engineering, and mathematics (STEM) subjects, or job interview trainings for unemployed persons. Wood and Bandura (1989) state that individuals’ beliefs and individuals’ actions are mutually reinforcing, suggesting an upward spiral between mastery experiences and self-efficacy. Meta-analyses reviewing the associations of past behavior and self-efficacy over time, however, found stronger evidence for past behavior forming self-efficacy than self-efficacy predicting future behavior (Sitzmann & Yeo, 2013; Talsma et al., 2018). This line of research clearly speaks for the integration of behavior change techniques to prompt mastery experience in selfefficacy interventions, as nothing can convince an individual more of their abilities than direct experience. Table 32.1 summarizes further techniques that may promote mastery experiences.

32.3.2.2 Vicarious Experiences Vicarious experiences, such as observing similar others perform the behavior or enjoy or succeed at a task can activate beliefs of being able to master it in the observer – especially among observers with low beliefs in their own capabilities. Seeing others succeed at a behavior not only increases confidence but also teaches the observer useful strategies and skills needed to overcome certain barriers (Bandura, 1997). Bandura assumes self-disclosing coping models – role models that also endured difficulties on their way to behavior change – to be particularly

463

effective (Bandura, 1997). Model similarity, foremost age, and gender were found to increase social learning (Schunk, 1987). Two further factors that should enhance the social learning process are a similar level of capabilities between observer and model and a positive outcome of the observed behavior (Bandura, 1977a). It is also possible that individuals can be “self-models” and it is hypothesized that imagery and mental simulation interventions may serve to change behavior by bolstering self-efficacy through “vicarious” experience (see Chapter 33, this volume). See Sidebar 32.1 and Appendix 32.1 (supplemental materials) on the “This Girl Can” campaign for some concrete examples of role modeling, as well as Table 32.2 for more behavior change techniques to prompt vicarious experiences.

32.3.2.3 Verbal Persuasion Verbal or social persuasion encompasses the assurance of other persons that they believe in a client’s capabilities to succeed or encouraging feedback on progress. Bandura (1997) proposed that this is generally the weakest source of selfefficacy: If someone already possesses a strong belief in not being able to succeed, potentially emphasized by negative experiences, verbal persuasion needs to be addressed with care and experience to avoid reactance. Verbal self-persuasion through instructional self-talk or “just-do-it” mantras was, however, found to be effective in job search behavior (Yanar, Budworth, & Latham, 2009) and among lay and elite-level athletes (Tod, Hardy, & Oliver, 2011). Table 32.3 summarizes behavior change techniques to prompt verbal (self-)persuasion.

32.3.2.4 Affective and Somatic States Closely before starting a challenging task or when thinking of performing something completely new, most people experience at least a feeling of apprehension. This is usually accompanied by perplexing symptoms of physiological and

464

LISA M. WARNER AND DAVID P. FRENCH

Table 32.1 Techniques to prompt mastery experiences matched with techniques from behavior change taxonomies General Description of Technique Examples Create direct opportunities for performance

First-hand experience with behavior, e.g. girls’ days for STEM; skill building exercises, e.g. job interview training; exposure to feared stimuli

Technique Label from Behavior Change Taxonomies •

• •

• • • •

Graded mastery experiences Mental imagery

Preparation for setbacks

Self-monitoring of behavior and behavioral outcome

Setting goals with increasing difficulty toward final behavior, e.g. anxiety hierarchy, scheduling more exercise each week Imagining success or progress, mental simulation, and visualization of task, e.g. in sensu exposure in phobia treatment, withstanding cues to smoke Explaining abstinence violation effect and self-efficacy enhancing attribution styles (attributing setbacks to external and achievements to internal causes) Monitoring behavior as well as goal progress, e.g. food and weight-loss diary

• •

•

•

• •

•

•

•

Instruction on how to perform a behavior [4.1]a Exposure [7.7]a Behavioral practice/rehearsal [8.1]a Active learning [table 1]b Direct experience [table 5]b Repeated exposure [table 5]b Guided practice & enactive mastery experiences [table 7]b Graded tasks [8.7]a Set graded tasks [table 7]b Mental rehearsal of successful performance [15.2]a Using imagery [table 2]b Reattribution [4.3]a Reattribution training [table 4]b

Self-monitoring of behavior [2.3]a Self-monitoring of outcome of behavior [2.4]a Self-monitoring of behavior [table 7]b Focus on past success [15.3]a Reattribution training [table 7]b

• Writing down or retelling mastery • situations, e.g. in a biographical manner, prompting transfer of previously successful strategies to new behavior None codable Learning orientation Promoting the adaptation of a learning orientation (motivation to develop new skills, e.g. acquiring skills to pass an exam, training to present oneself at job interviews) as opposed to avoidance orientation (avoiding failure) and performance orientation (passing an exam, getting a job)

Reflection on past successes

Note. a Behavior change technique taxonomy version 1 (Michie et al., 2013); bIntervention mapping taxonomy (Kok et al., 2016).

Self-Efficacy Interventions

465

Sidebar 32.1 The “This Girl Can” campaign

The “This Girl Can” campaign was developed by Sport England and launched in 2015.1 It was developed to address the long-standing issue of women being less physically active than men and experiencing a variety of barriers to participating in physical activity, including lack of confidence, lack of engagement in physical activity since school, and body consciousness. The campaign used a variety of social and traditional media, including posters and posts on social media platforms. A key concern of the campaign was to increase self-efficacy, by convincing women of all ages that “this girl can” (engage in physical activity) by addressing common barriers to participation that undermine confidence in the capability to participate. The campaign addressed women of a variety of ethnicities, ages, and body shapes and promoted a variety of forms of physical activity from walking and running to less traditionally “female” forms of physical activity such as football and martial arts. Table 32.2 Techniques to prompt vicarious experiences matched with techniques from behavior change taxonomies General Description of Technique

Technique Label from Behavior Change Taxonomies

Examples

Life/symbolic modeling

Observing similar role models perform the behavior successfully, e.g. life role models in self-help groups, symbolic models in videos or success stories

Self-modeling

Pictures or videos of self-performing the behavior successfully, e.g. videotaping performance during training session for job interview training or exercise technique

Demonstration of the behavior [6.1]a • Vicarious consequences [16.3]a • Modeling [tables 1, 9]b • Mass media role modeling [table 10]b • Use of lay health workers; peer education [table 11]b None codable •

Note. aBehavior change technique taxonomy version 1 (Michie et al., 2013); bIntervention mapping taxonomy (Kok et al., 2016).

affective arousal (e.g., increased heart rate, sweat attacks). If such somatic and affective states are subjectively interpreted as unpreparedness, anticipation of poor performance, or inability, they can decrease individuals’ self-efficacy, add to the distress perceived, and impair subsequent performance (Bandura, 1997). Bandura posits

that guided mastery experience will enhance the level of coping skills and future situations will be perceived as less threatening while anxious thoughts may still occur but not overwhelm the 1

See www.sportengland.org/our-work/women/this-girlcan; see also screenshots in Appendix 32.1 (supplemental materials).

466

LISA M. WARNER AND DAVID P. FRENCH

Table 32.3 Techniques to prompt verbal persuasion matched with techniques from behavior change taxonomies General Description of Techniques Encouragement or verbal placebo from professional or significant other

Technique Label from Behavior Change Taxonomies

Examples Credible source stating that successful performance is very likely

•

• •

• •

Instructional or motivational self-talk

• Motivational self-talk, verbal self-guidance or “just-do-it” mantras, e.g. reiterating behavioral steps to oneself; identification of dysfunctional self-statements and transformation into positive alternatives

Verbal persuasion about capability [15.1]a Credible source [9.1]a Social support (emotional) [3.3]a Verbal persuasion [table 7]b Mobilizing social networks [table 10]b Self-talk [15.4]a

Note. aBehavior change technique taxonomy version 1 (Michie et al., 2013); bIntervention mapping taxonomy (Kok et al., 2016).

self-efficacious individual (Bandura, 1993). More behavior change techniques for this source can be found in Table 32.4.

32.4 Evidence Base: How Well Do Self-Efficacy Interventions Work and What Are the Active Ingredients and Possible Moderators of Intervention Effects? Bandura and Locke (2003) summarized the results of nine meta-analyses on self-efficacy and found relations between self-efficacy and individual performance outcomes with small-tomedium-size effects (range d = 0.27 to 0.40) and a small-to-medium effect size for the relation between collective self-efficacy and group outcomes (d = 0.37). These meta-analyses, however, drew mostly on correlational research. The extent to which self-efficacy can effectively be

facilitated via interventions, and which behavior change techniques or sources of self-efficacy are most effective in which behavioral domains, can be better estimated in systematic reviews and meta-analyses based on randomized controlled trials. Sections 32.4.1 and 32.4.2 give a very brief overview of meta-analytic findings, while Appendix 32.2 (supplemental materials) describes the effects of self-efficacy interventions in more detail.

32.4.1 Self-Efficacy Interventions: Do They Change Self-Efficacy? Systematic reviews of self-efficacy interventions have usually found small-to-moderate effects of the intervention on self-efficacy beliefs for multiple behaviors in the health domain, including physical activity (d = 0.16–0.37; French et al., 2014; Olander et al., 2013; S. L. Williams & French, 2011), dietary behaviors (g = 0.24; Prestwich

Self-Efficacy Interventions

467

Table 32.4 Techniques to prompt somatic and affective states matched with techniques from behavior change taxonomies General Description of Techniques Psychoeducation

Biofeedback Identification/acquisition of coping skills

Opportunities to test coping skills

Examples

Technique Label from Behavior Change Taxonomies

• Explanation of how mental processes influence biological functions • (mind-body association), e.g., explanation of somatic and affective consequences of catastrophizing bodily • symptoms, learning to interpret nervousness as sign of bodily preparation, avoiding misinterpretations of bodily symptoms as dangerous Demonstration of mind-body association • • Training of coping, relaxation or stress • management strategies to increase preparedness and reduce or control negative somatic and affective states prior to and during performance • Collection of corrective information on • somatic and affective symptoms through behavioral experiments, e.g., • mastery of feared situation together with therapist in interoceptive exposure • session, aiming at experiencing symp• toms as harmless and controlling • somatic and affective reactions with acquired stress management skills

Information about antecedents [4.2]a Information about emotional consequences [5.6]a Improving physical and emotional states [table 7]b

Biofeedback [2.6]a Problem solving [1.2]a Reduce negative emotions [11.2]a

Exposure [7.7]a Behavioral practice/ rehearsal [8.1]a Improving physical and emotional states [table 7]b Dramatic relief [table 3]b Direct experience [table 5]b Guided practice & enactive mastery experiences [table 7]b

Note. aBehavior change technique taxonomy version 1 (Michie et al., 2013); bIntervention mapping taxonomy (Kok et al., 2016).

et al., 2014), and condom use (d = 0.19; Noar, Pierce, & Black, 2010) and large effects on selfefficacy to reduce panic (d = 1.46; Fentz et al., 2014). Furthermore, interventions were shown to promote reading self-efficacy in pupils (g = 0.33; Unrau et al., 2018), as well as self-efficacy for parenting behaviors (d = 0.42–1.25; Liyana Amin, Tam, & Shorey, 2018; Wittkowski, Dowling, & Smith, 2016), and breastfeeding (d = 0.40–0.86; Brockway, Benzies, & Hayden, 2017; Galipeau et al., 2018). Taken together, these results

point to a pervasive effect of interventions targeting self-efficacy on actual self-efficacy change.

32.4.2 Self-Efficacy Interventions: Do They Change Behavior and Are Intervention Effects Mediated by Self-Efficacy? Some of the reviews mentioned in Sections 32.4.1 and 32.4.2 also investigated whether these interventions had an effect on subsequent

468

LISA M. WARNER AND DAVID P. FRENCH

behavior. Small-to-medium-size effects of selfefficacy–based intervention on behavior were found for physical activity behavior in healthy (d = 0.26; S. L. Williams & French, 2011), obese (d = 0.50; Olander et al., 2013), and older adults (d = 0.14; French et al., 2014). Similarly, interventions produced changes in behavioral outcomes for HIV prevention (Mize et al., 2002; Robinson et al., 2017), addiction (Hyde et al., 2008), and breastfeeding (OR = 1.56; RR = 0.97; Brockway et al., 2017; Galipeau et al., 2018) behaviors with small-to-medium-size effects. There is also good evidence that selfefficacy serves to mediate the effects of selfefficacy interventions on physical activity participation in children and adolescents (Lubans, Foster, & Biddle, 2008). Similarly, self-efficacy to reduce panic was found to mediate the effects of self-efficacy interventions on panic severity (Fentz et al., 2014). These studies provide some evidence of the role of self-efficacy beliefs as a mechanism of action by which interventions “work” in promoting behavior change.

32.4.3 Self-efficacy Interventions: What Makes Them Effective? Although it would be desirable to be able to give universal guidance on which behavior change techniques are most effective to increase selfefficacy in general, finding the best overall method across all behaviors and populations may result in recommendations that are less useful for each specific context. This is consistent with research that suggests that different sources of self-efficacy are related to self-efficacy across different behaviors. Mastery experiences usually emerge as the strongest predictor of self-efficacy beliefs (Byars-Winston et al., 2017; Morris, Usher, & Chen, 2017; Schunk, 2003; Warner et al., 2014). However, it depends on which behavioral domain and which theoretical and methodological framework is applied to identify the most effective techniques of self-efficacy

interventions. For a more detailed review of effective behavior change techniques in selfefficacy interventions, see Appendix 32.2 (supplemental materials).

32.4.4 Moderators of Self-Efficacy Interventions 32.4.4.1 Methodological Aspects Some meta-analyses find that intervention effects depend on the measure used to operationalize self-efficacy (Noar et al., 2010; Wittkowski et al., 2016). It is therefore recommended to use specific self-efficacy measures so that they correspond well with the behaviors in question, similar to the principle of correspondence outlined in social cognitive theories (see Chapter 2, this volume). It is therefore important that measures of self-efficacy are adopted for each behavior targeted in complex behavior change interventions.

32.4.4.2 Subgroups A number of demographic factors may affect the efficacy of self-efficacy interventions and interact with the specific behavioral context in determining behavior. In some contexts, gender has been identified as an influential moderator. For example, in the academic context, women are more likely than men to indicate vicarious experiences as a significant influence on their math selfefficacy (Lent et al., 1996). In HIV prevention, self-efficacy beliefs in condom use skills were important for males, whereas self-efficacy beliefs in interpersonal skills were important for females (Albarracín et al., 2005). Similarly, age has been identified as a moderator. For example, among young and middle-aged individuals, self-regulatory techniques such as self-monitoring are among the strongest behavior change techniques to increase dietary selfefficacy as well as physical activity and healthy eating (Michie et al., 2009; Prestwich et al., 2014). For older adults, self-monitoring of

Self-Efficacy Interventions

physical activity behavior was, however, found to decrease the likelihood of interventions to be effective in promoting self-efficacy (French et al., 2014). Prompting to self-monitor the behavioral outcomes instead of the behavior itself was, however, associated with increased physical activity self-efficacy in older adults. Age or experience may also play a role in academic self-efficacy, as interventions in elementary schools yielded stronger effects than those in higher grade levels (Unrau et al., 2018). Further, culture and ethnicity may need to be considered as moderators of how well different sources of self-efficacy relate to self-efficacy. For example, Gainor and Lent (1998) found that vicarious experiences were the strongest correlate of math efficacy beliefs for black college students. In a meta-regression of the sources of self-efficacy, Byars-Winston et al. (2017) also point out that there might be differences in the strengths of sources of self-efficacy depending on whether STEM or non-STEM subjects are investigated. A meta-analysis summarizing five parental selfefficacy interventions showed that self-efficacy could best be increased in mothers as compared to fathers and in Latino compared to African American parents (Wittkowski et al., 2016). There is also some evidence that breastfeeding self-efficacy interventions might work better in countries with higher socioeconomic status (Brockway et al., 2017). Taken together, these moderator findings highlight that it should not be assumed that interventions that are effective for one demographic group are effective for another (Tang et al., 2019). This flags up the importance of pilot work in each new setting. The findings are consistent with the mechanisms that social cognitive theory propose, for example vicarious experience appearing to be more influential for women than men. However, current theories in this area do not make strong predictions about the circumstances under which this could happen, as they do not capture wider contextual factors about, for example, how men

469

and women may be socialized differently within education.

32.4.4.3 Overconfidence People overestimating their abilities can decrease their chances of successful behavior change. For example, high academic self-efficacy in adolescents with learning disabilities might signal inappropriate awareness of the task at hand, with the result that motivation and performance can be impaired (Klassen, 2006). Several studies on motivation and various tasks confirm that very high levels of self-efficacy sometimes lead to complacency, resulting in people preparing less for the task ahead or spending less time on it, which affects performance and outcomes (Schönfeld, Preusser, & Margraf, 2017). Some trials even find self-efficacy levels decline after interventions, such as interventions promoting quitting smoking (Staring & Breteler, 2004) or implementing a new teaching technique in teachers (Tschannen-Moran & McMaster, 2009). Individuals might overestimate their own capabilities and only realize the difficulty in initiating lasting behavior change when they first try, also known as the “implementation dip” in self-efficacy (see also Bandura’s reply: Bandura & Locke, 2003). A solution to this dilemma can be to assess self-efficacy levels prior to an intervention and only support the accrual of self-efficacy up to moderately high levels (Schönfeld et al., 2017).

32.4.4.4 Open Research Questions A large number of studies underpinning the effect of self-efficacy on successful performance in various domains of life have been published since Bandura first introduced the construct, but there has only been little empirical research on the origins of self-efficacy beliefs in most domains, with the notable exception of academic selfefficacy, and very little novel theorizing. As long as the sources of self-efficacy are not measured as mediating variables in interventions, and factorial randomized controlled trials are

470

LISA M. WARNER AND DAVID P. FRENCH

scarce, there is a substantive research gap between the theoretical sources of self-efficacy and identification of the most effective behavior change techniques that should be targeted in interventions. Overall, the available evidence suggests that mastery experiences are the best source of self-efficacy beliefs. There is no recommendation yet on which sources of self-efficacy best combine with mastery or which source of self-efficacy should be prompted first if mastery experiences are not yet existent and performing the behavior is not yet possible, for example in the case of phobias. Future behavior change programs based on social cognitive theory, and self-efficacy in particular, should therefore test different sources of selfefficacy and different combinations of sources against one another. Not only should they calculate the overall effect of a complex intervention on self-efficacy and behavior but they should also try to assess changes in the sources to identify the active ingredients within interventions (see Section 32.5.7 on evaluation). This may provide valuable insights for researchers and practitioners on how to develop the most effective yet parsimonious interventions.

32.5 Step-by-Step Guide to Self-Efficacy Interventions Although the multitude of strategies to increase selfefficacy may appear overwhelming, there is a silver lining: Not all sources of self-efficacy need to be addressed for all individuals in all behavioral domains. Identifying needs and possibilities to increase self-efficacy for a specific client at a given point in time may set off a positive spiral of successful first attempts, which likely lead to increased selfefficacy subsequently and the setting of higher goals and subsequent progression with the behavior.

32.5.1 Delivery and Format Self-efficacy interventions can be either implemented as a self-help technique (see Knittle

et al., 2019) or prompted by professionals. Interventions aimed at promoting mastery experiences could be self-administered by the individual through print or text interventions or through selfhelp guides such as brochures or online text interventions, prompting “try-and-see” experiences. Alternatively, mastery experience interventions can be delivered as face-to-face programs either through one-on-one sessions or as group-level interventions led by a practitioner or peer model, involving graduated introduction to the behavior. If mastery experiences have already been acquired for similar behaviors, text interventions may encourage participants to look back on past successes and reflect on the factors that determined the change (for an example see Sidebar 32.2; Aro et al., 2018). For vicarious experience, direct role modeling of a live model or online tutorial videos can be useful in demonstrating how a certain behavior can be executed. Direct and symbolic modeling is used for behaviors that require acquisitions of new and rather difficult behavioral sequences, such as learning to brush one’s teeth as a child. For behaviors that pose a challenge to individuals’ self-regulation (e.g., physically active lifestyle, managing diabetes, quitting smoking), the typical means of delivery for vicarious experiences is through vignettes or success stories illustrating strategies that help change behavior and develop new habits. Interventions targeting verbal persuasion are typically delivered via sentences such as “you can do it” or “if I was able to make a change, you will as well,” delivered via face-to-face or written material. Verbal persuasion should be combined with other techniques to increase self-efficacy, as it might by perceived as unauthentic if the encouragement comes from someone who does not know the client’s abilities or from a text the client reads in an online intervention or leaflet (Wise & Trunnell, 2001). This form of anonymous verbal persuasion might cause reactance in individuals with low

Self-Efficacy Interventions

471

Sidebar 32.2 Can reading fluency and self-efficacy of reading fluency be enhanced with an intervention targeting the sources of self-efficacy? (Aro et al., 2018)

This quasi-experimental study developed a twelve-week education program targeting all four sources of self-efficacy for reading fluency in third-to-fifth-grade students against a standard reading fluency program (see example materials in Appendix 32.3, supplemental materials). Both intervention groups trained their fluency during group sessions and with a computer program. The self-efficacy group was also prompted to write down their mastery experiences during the session and received visualized feedback from the program on learning progress. Teachers also gave verbal feedback on learning progress and made sure the development was noted by every child and interpreted in a self-efficacy–enhancing way. To enhance vicarious experience, students were encouraged to observe improvements by peers. Affective states were addressed using an emotional checklist in which students marked how they felt after the session. Students were encouraged to compare their performance over time rather than compare themselves with other students. Both training groups improved in reading fluency but only within the self-efficacy intervention group; self-efficacy for reading fluency was associated with increases in word-level reading fluency.

Sidebar 32.3 Intervention on sources of self-efficacy for breastfeeding (Nichols et al., 2009)

Women in their third trimester of pregnancy were recruited in Australia. The control group received a five-page self-exploratory interactive workbook on parenting issues with no reference to breastfeeding. The intervention group received a nine-page selfexploratory interactive workbook aiming at enhancing breastfeeding self-efficacy. The mastery experience part invited participants to note thoughts and feelings related to previous accomplishments and to write them down to generalize identified breastfeeding skills. Vicarious experience was prompted by testimonials from breastfeeding mothers, for example “I felt useless as things weren’t going well … I am so glad I persevered.” In the verbal persuasion part, women were asked to write down encouraging statements they would find useful: “Write down what sort of things your mother/partner/friend could say to you to encourage you in breastfeeding.” Physiological states and attribution styles are addressed by explaining “We might not be able to change the situation but we can change how we think and feel toward the situation.” It was also explained that having negative thoughts and doubts is okay but that an effort should be undertaken not to procrastinate and lose oneself in self-debilitating thoughts. To escape a negative attribution style, the workbook leaves space to write down individual encouraging sentences like “I’m not Superwoman, but I am a good mother.” self-efficacy, with them thinking that no one can possibly know how hard it is for them to change. For example, it might be useful to use verbal persuasion in conjunction with vicarious

experience, or mastery experience, so that individuals gain a sense of competence with the behavior or activity prior to being told that they can be successful.

472

LISA M. WARNER AND DAVID P. FRENCH

Somatic and affective states are typically addressed in self-efficacy interventions for tasks or behaviors that are associated with negative physiological or emotional sensations – foremost in the treatment of phobias or panic attacks. These sessions usually take place in a face-to-face manner following an individualized therapeutic approach. An excerpt from a self-exploratory workbook developed to enhance all four sources of selfefficacy successfully implemented in Japan and Australia can be found in Sidebar 32.3 (for the full workbook based on this approach, see Appendix 32.4, supplemental materials; Nichols et al., 2009; Otsuka et al., 2014). Meta-analyses have found that self-efficacy interventions adopting a combination of different formats or modes of delivery are more effective (Galipeau et al., 2018).

32.5.2 Target Audience and Behaviors As mentioned in Section 32.4.4.3, it is only sensible to boost self-efficacy in individuals who lack the belief in their ability to change; otherwise, there is potential for an overconfidence effect. This lack of self-efficacy will be more apparent for behaviors that are perceived as very challenging, such as giving up on substance abuse or overcoming anxieties. For easier tasks, such as regularly flossing one’s teeth, behavior change techniques that target self-regulatory efforts and the formation of habits are likely to be more effective.

32.5.3 Combinations of Behavior Change Techniques Combining other behavior change techniques with techniques targeting the sources of self-efficacy may maximize the effects of self-efficacy interventions. Observational studies in the context of coping with stress show that social support had an “enabling” effect by serving as a

determinant of self-efficacy but also demonstrated that social support had a “cultivating effect” by serving as a mediator of effects of self-efficacy on stress-related outcomes such as depressed mood (Schwarzer & Knoll, 2007). Woodgate and Brawley (2008), for example, showed how supportive and uplifting messages vividly describing strategies used by similar others in a rehabilitation context led to greater self-efficacy. There is also a promising body of research demonstrating that means to promote self-efficacy alongside planning interventions (see Chapter 6, this volume) lead to greater behavioral outcomes than techniques targeting selfefficacy and planning alone (Luszczynska et al., 2011; Wieber, Odenthal, & Gollwitzer, 2010).

32.5.4 Training and Skills Required for Practitioners Interventions using written material such as leaflets, workbooks, websites, apps, and video material require no specialist skills for delivery once these are set up and have been evaluated and piloted for their content. Unless the behavior change problem is a clinically diagnosed psychiatric problem, formal training as a counselor is usually not necessary to develop and conduct self-efficacy interventions. However, if the intervention is delivered face-to-face, the deliverer should have the appropriate interpersonal skills to interact with the recipient of the intervention and should receive adequate training by the intervention team to be able to respond to questions and interact appropriately in order to ensure fidelity of delivery (see Section 32.5.6).

32.5.5 Intensiveness and Timing No recommendation on the optimal number or length of sessions can be given in general, as the dosage highly depends on the behavior and population targeted. For example, interventions have been shown to be effective in very small doses

Self-Efficacy Interventions

(e.g., one or two face-to-face sessions lasting less than ninety minutes) in supporting the selfefficacy of parents (Liyana Amin et al., 2018; Wittkowski et al., 2016), in reading self-efficacy (Unrau et al., 2018), and in breastfeeding selfefficacy (Brockway et al., 2017).

32.5.6 Evaluation of Fidelity The US National Institutes of Health behavior change consortium (Bellg et al., 2004) proposes a range of aspects of intervention fidelity, such as fidelity in design, training, delivery, receipt, and enactment of behavioral interventions (see Chapters 21 and 22, this volume). With regard to design, intervention designers should establish whether a formal approach such as intervention mapping (Bartholomew, Parcel, & Kok, 1998) could be used to identify self-efficacy as a causal determinant of the behavior and provide a justification for why self-efficacy should be targeted in the intervention. Furthermore, if self-efficacy is identified as a target, one could ensure that planned intervention materials are assessed based on the extent to which they adequately contain methods to increase self-efficacy. Assessment of adequacy and uptake of the training of those delivering the intervention is also important and would necessitate an analysis of the training material content. In terms of delivery, interventionists could examine what was actually delivered to the recipients, for example by coding video or audio recordings of face-to-face intervention sessions, or coding the extent to which individuals receiving the intervention delivered via paper or electronic paper engaged with the material and responded to questions relating to the material. Such questions might involve assessing the extent to which participants understood that the content of the intervention was designed to increase self-efficacy. This may be particularly important in some populations. For example, there is evidence that older adults do not understand the reasoning behind some methods to

473

increase self-efficacy (French et al., 2014). Finally, enactment would consider whether participants actually use the behavior change techniques that were designed to increase self-efficacy. For example, this might involve assessing whether members of the target population engage sufficiently with those providing modeling to increase vicarious experience (see Chapter 22, this volume).

32.5.7 Evaluation of Effectiveness: How to Establish Whether SelfEfficacy Has Changed? Regardless of how well an intervention changed the behavior, robust tests of theory-based interventions need to assess whether this change was brought about by the hypothesized working mechanism – self-efficacy. Self-efficacy interventions should, therefore, compare at least two groups – one with and one without behavior change techniques that target one or more sources of self-efficacy. It would be even better to compare different techniques in factorial designs with groups defined by techniques targeting change in different sources of self-efficacy, again against comparisons. The first step to an evaluation of a self-efficacy intervention is to select appropriate measures of self-efficacy, the construct targeted, and the target behavior, relevant to the behavioral domain under study. The study design should incorporate at least one baseline assessment point of these measures and one measurement point after the intervention and, ideally, more follow-up measures, for both intervention and control, or comparison, group or groups. As enactment of behavior can result in increases in self-efficacy, it is particularly important that studies assess changes in selfefficacy immediately after the intervention but before the behavior can be enacted outside the intervention context, to allow unambiguous attribution of behavior change effects to self-efficacy rather than self-efficacy being a consequence of

474

LISA M. WARNER AND DAVID P. FRENCH

behavior change (French, 2013). If different sources of self-efficacy are targeted in one intervention, not only the change in self-efficacy but also changes in the manipulated sources should be assessed as additional mediating mechanisms. To investigate which of the sources or which combinations of sources work best in different behavioral domains, specific measures assessing the sources of self-efficacy should be constructed, like those existing for the assessment of the source of teacher self-efficacy (Morris et al., 2017), self-efficacy for reading and writing (Schunk, 2003), and self-efficacy for physical activity (Warner et al., 2014).

32.6 Conclusion A vast literature has tested interventions targeting change in self-efficacy in different populations, behaviors, and contexts, and research has shown them to be effective, although the strength of effects vary considerably. Consistent with social cognitive theory, self-efficacy can be promoted through sets of techniques targeting change in four sources of self-efficacy: mastery experience, vicarious experience, verbal persuasion, and affective states. Effects of interventions targeting these sources may differ depending on the behavioral domain, target population, and delivery format. As a consequence, it is difficult to provide definitive advice on the content of “typical” self-efficacy interventions. Instead, the present chapter has provided examples of effective self-efficacy interventions, and the reader is directed to the appendixes for complete materials of these examples. Practitioners are referred to research that provides in-depth and comprehensive treatment of the development, implementation, and tests of effectiveness of self-efficacy interventions, such as the nurse-led intervention to reduce drinking (Holloway, Watson, & Starr, 2006), a practical summary describing interventions to increase selfefficacy for chronic disease management (Marks, Allegrante, & Lorig, 2005), or the intervention to promote schoolchildren’s college and career self-

efficacy (Glessner, Rockinson-Szapkiw, & Lopez, 2017). More research is needed, however, to elucidate the unique effects of specific techniques targeting the different sources of self-efficacy on behavior change, as the literature is surprisingly sparse in this area. Bandura (2004) provides an apt summary of behavior change interventions using self-efficacy: “As you venture forth to promote your own health and that of others, may the efficacy force be with you” (p. 162).

References Albarracín, D., Gillette, J. C., Earl, A. N., Glasman, L. R., Durantini, M. R., & Ho, M.-H. (2005). A test of major assumptions about behavior change: A comprehensive look at the effects of passive and active HIV-prevention interventions since the beginning of the epidemic. Psychological Bulletin, 131, 856–897. https://doi.org/10.1037/ 0033-2909.131.6.856 Aro, T., Viholainen, H., Koponen, T. et al. (2018). Can reading fluency and self-efficacy of reading fluency be enhanced with an intervention targeting the sources of self-efficacy? Learning and Individual Differences, 67, 53–66. https://doi.org/ 10.1016/j.lindif.2018.06.009 Bandura, A. (1977a). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84, 191–215. https://doi.org/10.1037/ 0033-295x.84.2.191 Bandura, A. (1977b). Social Learning Theory. Englewood Cliffs, NJ: Prentice Hall. Bandura, A. (1986). Social Foundations of Thought and Action. Englewood Cliffs, NJ: Prentice Hall. Bandura, A. (1993). Perceived self-efficacy in cognitive development and functioning. Educational Psychologist, 28, 117–148. https:// doi.org/10.1207/s15326985ep2802_3 Bandura, A. (1997). Self-Efficacy: The Exercise of Control. New York: Freeman. Bandura, A. (2004). Health promotion by social cognitive means. Health Education and Behavior, 31, 143–164. https://doi.org/10.1177/ 1090198104263660

Self-Efficacy Interventions

Bandura, A. (2006). Guide for constructing selfefficacy scales. In F. Pajares & T. Urdan (Eds.), Self-Efficacy Beliefs of Adolescents, Vol. 5 (pp. 307–337). Greenwich, CT: Information Age Publishing. Bandura, A., & Locke, E. A. (2003). Negative selfefficacy and goal effects revisited. Journal of Applied Psychology, 88, 87–99. https://doi.org/ 10.1037/0021-9010.88.1.87 Bartholomew, L. K., Parcel, G. S., & Kok, G. (1998). Intervention mapping: A process for developing theory and evidence-based health education programs. Health Education and Behavior, 25, 545–563. https://doi.org/10.1177/ 109019819802500502 Bellg, A. J., Borrelli, B., Resnick, B. et al. (2004). Enhancing treatment fidelity in health behavior change studies: Best practices and recommendations from the NIH Behavior Change Consortium. Health Psychology, 23, 443. https://doi.org/10.1037/02786133.23.5.443 Brockway, M., Benzies, K., & Hayden, K. A. (2017). Interventions to improve breastfeeding selfefficacy and resultant breastfeeding rates: A systematic review and meta-analysis. Journal of Human Lactation, 33, 486–499. https://doi.org/ 10.1177/0890334417707957 Byars-Winston, A., Diestelmann, J., Savoy, J. N., & Hoyt, W. T. (2017). Unique effects and moderators of effects of sources on self-efficacy: A modelbased meta-analysis. Journal of Counseling Psychology, 64, 645. https://doi.org/10.1037/ cou0000219 Capa-Aydin, Y., Uzuntiryaki-Kondakci, E., & Ceylandag, R. (2018). The relationship between vicarious experience, social persuasion, physiological state, and chemistry self-efficacy: The role of mastery experience as a mediator. Psychology in the Schools, 55, 1224–1238. https:// doi.org/10.1002/pits.22201 Chemers, M. M., Watson, C. B., & May, S. T. (2000). Dispositional affect and leadership effectiveness: A comparison of self-esteem, optimism, and efficacy. Personality and Social Psychology Bulletin, 26, 267–277. https://doi.org/10.1177/ 0146167200265001

475

Fentz, H. N., Arendt, M., O’Toole, M. S., Hoffart, A., & Hougaard, E. (2014). The mediational role of panic self-efficacy in cognitive behavioral therapy for panic disorder: A systematic review and metaanalysis. Behaviour Research and Therapy, 60, 23–33. https://doi.org/10.1016/j.brat.2014.06.003 Fishbein, M., & Ajzen, I. (1975). Belief, Attitude, Intention, and Behavior: An Introduction to Theory and Research. Reading, MA: AddisonWesley. Forsyth, A. D., Carey, M. P., & Fuqua, R. W. (1997). Evaluation of the validity of the Condom Use SelfEfficacy Scale (CUSES) in young men using two behavioral simulations. Health Psychology, 16, 175–178. https://doi.org/10.1037/02786133.16.2.175 French, D. P. (2013). The role of self-efficacy in changing health-related behaviour: Cause, effect or spurious association? British Journal of Health Psychology, 18, 237–243. https://doi.org/10.1111/ bjhp.12038 French, D. P., Olander, E., Chisholm, A., & Mc Sharry, J. (2014). Which behaviour change techniques are most effective at increasing older adults’ selfefficacy and physical activity behaviour? A systematic review. Annals of Behavioral Medicine, 48, 225–234. https://doi.org/10.1007/ s12160-014-9593-z Gainor, K. A., & Lent, R. W. (1998). Social cognitive expectations and racial identity attitudes in predicting the math choice intentions of Black college students. Journal of Counseling Psychology, 45, 403. https://doi.org/10.1037/ 0022-0167.45.4.403 Galipeau, R., Baillot, A., Trottier, A., & Lemire, L. (2018). Effectiveness of interventions on breastfeeding self-efficacy and perceived insufficient milk supply: A systematic review and meta-analysis. Maternal and Child Nutrition, 14, e12607. https://doi.org/10.1111/ mcn.12607 Glessner, K., Rockinson-Szapkiw, A. J., & Lopez, M. L. (2017). “Yes, I Can”: Testing an intervention to increase middle school students’ college and career self-efficacy. The Career Development Quarterly, 65, 315–325. https://doi.org/https:// doi.org/10.1002/cdq.12110

476

LISA M. WARNER AND DAVID P. FRENCH

Holloway, A. S., Watson, H. E., & Starr, G. (2006). How do we increase problem drinkers’ selfefficacy? A nurse-led brief intervention putting theory into practice. Journal of Substance Use, 11, 375–386. https://doi.org/10.1080/ 14659890601037376 Hyde, J., Hankins, M., Deale, A., & Marteau, T. M. (2008). Interventions to increase self-efficacy in the context of addiction behaviours: A systematic literature review. Journal of Health Psychology, 13, 607–623. https://doi.org/10.1177/ 1359105308090933 Klassen, R. (2006). Too much confidence? The selfefficacy beliefs of adolescents with learning disabilities. In F. Pajares & T. Urdan (Eds.), SelfEfficacy Beliefs of Adolescents (pp. 181–200). Greenwich, CT: Information Age Publishing. Knittle, K., Heino, M. T. J., Marques, M. et al. (2020). The compendium of self-enactable techniques to change and self-manage motivation and behaviour v1. 0. Nature Human Behavior. https://doi.org/ 10.1038/s41562-019-0798-9 Kok, G., Gottlieb, N. H., Peters, G.-J. Y. et al. (2016). A taxonomy of behavior change methods: An intervention mapping approach. Health Psychology Review, 10, 297–312. https://doi.org/ 10.1080/17437199.2015.1077155 Lent, R. W., Brown, S. D., Gover, M. R., & Nijjer, S. K. (1996). Cognitive assessment of the sources of mathematics self-efficacy: A thought-listing analysis. Journal of Career Assessment, 4, 33–46. https://doi.org/10.1177/106907279600400102 Liyana Amin, N. A., Tam, W. W. S., & Shorey, S. (2018). Enhancing first-time parents’ selfefficacy: A systematic review and meta-analysis of universal parent education interventions’ efficacy. International Journal of Nursing Studies, 82, 149–162. https://doi.org/10.1016/ j.ijnurstu.2018.03.021 Lubans, D. R., Foster, C., & Biddle, S. J. (2008). A review of mediators of behavior in interventions to promote physical activity among children and adolescents. Preventive Medicine, 47, 463–470. https://doi.org/10.1016/j.ypmed.2008.07.011 Luszczynska, A., Schwarzer, R., Lippke, S., & Mazurkiewicz, M. (2011). Self-efficacy as a moderator of the planning-behaviour relationship

in interventions designed to promote physical activity. Psychology and Health, 26, 151–166. https://doi.org/10.1080/08870446.2011.531571 Maddux, J. E., & Rogers, R. W. (1983). Protection motivation and self-efficacy: A revised theory of fear appeals and attitude change. Journal of Experimental Social Psychology, 19, 469–479. https://doi.org/10.1016/0022-1031(83)90023-9 Magklara, E., Burton, C. R., & Morrison, V. (2014). Does self-efficacy influence recovery and wellbeing in osteoarthritis patients undergoing joint replacement? A systematic review. Clinical Rehabilitation, 28, 835–846. https://doi.org/ 10.1177/0269215514527843 Marks, R., Allegrante, J. P., & Lorig, K. (2005). A review and synthesis of research evidence for selfefficacy-enhancing interventions for reducing chronic disability: Implications for health education practice (part I). Health Promotion Practice, 6, 37–43. https://doi.org/10.1177/ 1524839904266790 Michie, S., Abraham, C., Whittington, C., McAteer, J., & Gupta, S. (2009). Effective techniques in healthy eating and physical activity interventions: A meta-regression. Health Psychology, 28, 690– 701. https://doi.org/10.1037/a0016136 Michie, S., Richardson, M., Johnston, M. et al. (2013). The behavior change technique taxonomy (v1) of 93 hierarchically clustered techniques: Building an international consensus for the reporting of behavior change interventions. Annals of Behavioral Medicine, 46, 81–95. https://doi.org/ 10.1007/s12160-013-9486-6 Mize, S. J., Robinson, B., Bockting, W., & Scheltema, K. (2002). Meta-analysis of the effectiveness of HIV prevention interventions for women. AIDS care, 14, 163–180. https://doi.org/10.1080/ 09540120220104686 Morris, D. B., Usher, E. L., & Chen, J. A. (2017). Reconceptualizing the sources of teaching selfefficacy: A critical review of emerging literature. Educational Psychology Review, 29, 795–833. https://doi.org/10.1007/s10648-016-9378-y Nichols, J., Schutte, N. S., Brown, R. F., Dennis, C.-L., & Price, I. (2009). The impact of a self-efficacy intervention on short-term breast-feeding outcomes. Health Education and Behavior, 36,

Self-Efficacy Interventions

250–258. https://doi.org/10.1177/ 1090198107303362 Noar, S. M., Pierce, L. B., & Black, H. G. (2010). Can computer-mediated interventions change theoretical mediators of safer sex? A meta-analysis. Human Communication Research, 36, 261–297. Olander, E. K., Fletcher, H., Williams, S., Atkinson, L., Turner, A., & French, D. P. (2013). What are the most effective techniques in changing obese individuals’ physical activity self-efficacy and behaviour: A systematic review and metaanalysis. International Journal of Behavioral Nutrition and Physical Activity, 10, 29. https://doi .org/10.1186/1479-5868-10-29 Otsuka, K., Taguri, M., Dennis, C.-L. et al. (2014). Effectiveness of a breastfeeding self-efficacy intervention: Do hospital practices make a difference? Maternal and Child Health Journal, 18, 296–306. https://doi.org/10.1007/s10995-0131265-2 Prestwich, A., Kellar, I., Parker, R. et al. (2014). How can self-efficacy be increased? Meta-analysis of dietary interventions. Health Psychology Review, 8, 270–285. https://doi.org/10.1080/ 17437199.2013.813729 Robinson, J. L., Narasimhan, M., Amin, A. et al. (2017). Interventions to address unequal gender and power relations and improve self-efficacy and empowerment for sexual and reproductive health decision-making for women living with HIV: A systematic review. PLoS ONE, 12, e0180699. https://doi.org/10.1371/journal.pone.0180699 Schönfeld, P., Preusser, F., & Margraf, J. (2017). Costs and benefits of self-efficacy: Differences of the stress response and clinical implications. Neuroscience and Biobehavioral Reviews, 75, 40–52. https://doi.org/10.1016/ j.neubiorev.2017.01.031 Schunk, D. H. (1987). Peer models and children’s behavioral change. Review of Educational Research, 57, 149–174. https://doi.org/10.2307/ 1170234 Schunk, D. H. (2003). Self-efficacy for reading and writing: Influence of modeling, goal setting, and self-evaluation. Reading and Writing Quarterly, 19, 159–172. https://doi.org/10.1080/ 10573560308219

477

Schwarzer, R., & Knoll, N. (2007). Functional roles of social support within the stress and coping process: A theoretical and empirical overview. International Journal of Psychology, 42, 243–252. https://doi.org/10.1080/ 00207590701396641 Sitzmann, T., & Yeo, G. (2013). A meta-analytic investigation of the within-person self-efficacy domain: Is self-efficacy a product of past performance or a driver of future performance? Personnel Psychology, 66, 531–568. https://doi .org/10.1111/peps.12035 Staring, A. B. P., & Breteler, M. H. M. (2004). Decline in smoking cessation rate associated with high self-efficacy scores. Preventive Medicine, 39, 863–868. https://doi.org/10.1016/ j.ypmed.2004.03.025 Talsma, K., Schüz, B., Schwarzer, R., & Norris, K. (2018). I believe, therefore I achieve (and vice versa): A meta-analytic cross-lagged panel analysis of self-efficacy and academic performance. Learning and Individual Differences, 61, 136–150. https://doi.org/10.1016/ j.lindif.2017.11.015 Tang, M. Y., Smith, D. M., Mc Sharry, J., Hann, M. & French, D. P. (2019). The role of self-efficacy in the initiation and maintenance of physical activity: Systematic review with meta-analysis. Annals of Behavioral Medicine, 53, 801–815. https://doi .org/10.1093/abm/kay090 Tod, D., Hardy, J., & Oliver, E. (2011). Effects of selftalk: A systematic review. Journal of Sport and Exercise Psychology, 33, 666–687. https://doi.org/ 10.1123/jsep.33.5.666 Tschannen-Moran, M., & McMaster, P. (2009). Sources of self-efficacy: Four professional development formats and their relationship to self-efficacy and implementation of a new teaching strategy. The Elementary School Journal, 110, 228–245. https:// doi.org/10.1086/605771 Unrau, N. J., Rueda, R., Son, E., Polanin, J. R., Lundeen, R. J., & Muraszewski, A. K. (2018). Can reading self-efficacy be modified? A metaanalysis of the impact of interventions on reading self-efficacy. Review of Educational Research, 88, 167–204. https://doi.org/10.3102/ 0034654317743199

478

LISA M. WARNER AND DAVID P. FRENCH

Warner, L. M., Schüz, B., Wolff, J. K., Parschau, L., Wurm, S., & Schwarzer, R. (2014). Sources of self-efficacy for physical activity. Health Psychology, 33, 1298–1308. https://doi.org/ 10.1037/hea0000085 Warner, L. M., & Schwarzer, R. (2017). Selfefficacy. In A. Wenzel (Ed.), The SAGE Encyclopedia of Abnormal and Clinical Psychology, Vol. 1 (pp. 3036–3038). Thousand Oaks, CA: SAGE. Wieber, F., Odenthal, G., & Gollwitzer, P. (2010). Self-efficacy feelings moderate implementation intention effects. Self and Identity, 9, 177–194. https://doi.org/10.1080/15298860902860333 Williams, D. M., & Rhodes, R. E. (2016). The confounded self-efficacy construct: Conceptual analysis and recommendations for future research. Health Psychology Review, 10, 113–128. https:// doi.org/10.1080/17437199.2014.941998 Williams, S. L., & French, D. P. (2011). What are the most effective intervention techniques for changing physical activity self-efficacy and physical activity behaviour – and are they the same? Health Education Research, 26, 308–322. https://doi.org/10.1093/her/cyr005

Wise, J. B., & Trunnell, E. P. (2001). The influence of sources of self-efficacy upon efficacy strength. Journal of Sport and Exercise Psychology, 23, 268–280. https://doi.org/10.1123/jsep.23.4.268 Wittkowski, A., Dowling, H., & Smith, D. M. (2016). Does engaging in a group-based intervention increase parental self-efficacy in parents of preschool children? A systematic review of the current literature. Journal of Child and Family Studies, 25, 3173–3191. Wood, R., & Bandura, A. (1989). Social cognitive theory of organizational management. Academy of Management Review, 14, 361–384. https://doi .org/10.5465/amr.1989.4279067 Woodgate, J., & Brawley, L. R. (2008). Self-efficacy for exercise in cardiac rehabilitation: Review and recommendations. Journal of Health Psychology, 13, 366–387. https://doi.org/10.1177/ 1359105307088141 Yanar, B., Budworth, M. H., & Latham, G. P. (2009). The effect of verbal self-guidance training for overcoming employment barriers: A study of Turkish women. Applied Psychology, 58, 586–601. https://doi.org/10.1111/j.14640597.2008.00366.x

33 Imagery, Visualization, and Mental Simulation Interventions Martin S. Hagger and Dominic Conroy

Practical Summary Imagery or “visualization” involves a person imagining or rehearsing future events, actions, or tasks, usually with the person performing the imagery imagining themselves actually performing an action or task. It has been used as a strategy to change behavior in many contexts and groups, and its effects on behavior change have been supported by an expanding body of research. Imagery interventions change behavior by changing beliefs or internal states of individuals, such as promoting greater confidence, or self-efficacy, for doing future tasks or assisting them to more effectively manage their emotional responses. Practitioners considering using mental imagery to change behavior should adopt imagery techniques that are suited to the population and context of interest (e.g., whether the imagery is guided by a practitioner or administered using print communication), give those doing the “imaging” or “visualizing” clear instructions, provide a higher “dose” of imagery exercises (frequency and duration), consider training the imagery ability of those doing the imagery and those delivering the imagery (if “face-toface”), and make sure those doing the imagery adhere to the imagery exercises.

33.1 Introduction Use of mental imagery or “visualization” as a technique to change behavior has a relatively long history in many behavioral domains. Domains such as sport and business have embraced mental imagery as an important means to enhance the motivation and confidence of individuals in performance contexts, such as athletes or executives (Murphy, Nordin, & Cumming, 2008; Neck & Manz, 1996). Imagery is considered an effective and established strategy in the “toolbox” of mental skills of people in performance contexts. The use of imagery to change behavior in nonperformance–related contexts is less well established and only relatively recently has there been efforts to capitalize on imagery as a

means to change behavior in these other contexts (e.g., Andrade et al., 2016; Chan & Cameron, 2012; Conroy & Hagger, 2018; Loft & Cameron, 2013; Rennie, Harris, & Webb, 2014). Imagery or “visualization” is an umbrella term that encompasses a broad category of techniques that involve individuals engaging in mental exercises in which they imagine or mentally rehearse performing a future desired behavior or action. Imagery exercises often require individuals to draw on their own experiences with the action and reproduce it mentally. Some types of imagery prompt individuals to imagine the feelings linked to performing the action or how they https://doi.org/10.1017/9781108677318.033

480

MARTIN S. HAGGER AND DOMINIC CONROY

would feel after doing it rather than steps required to participate in the action itself (Pham & Taylor, 1999b). Imagery exercises aim to change behavior by promoting increased confidence, attention and concentration, and motivation toward the behavior (Conroy & Hagger, 2018; Pham & Taylor, 1999b). The purpose of the present chapter is to provide an overview of mental imagery techniques employed to change behavior, and to provide a practical guide on how practitioners can use mental imagery to change behavior. Definitions and types of mental imagery available will be presented. Theoretical perspectives on how mental imagery works in changing behavior, including the proposed mechanisms and psychological factors that are proposed to mediate mental imagery effects will also be reviewed. In addition, the evidence base for the efficacy of imagery interventions in changing behavior will be reviewed and evaluated, including evidence for the proposed mechanisms and effects. The chapter will also outline how imagery techniques are represented in taxonomies of behavior change techniques. Finally, a set of guidelines on how imagery techniques can be used in interventions and implemented in practice is provided, including detail on typical means of delivery, target audience, factors that might enhance or diminish the efficacy of the imagery, skills required to deliver imagery interventions, how intensive imagery interventions need to be, the typical materials needed, and how to evaluate the efficacy and fidelity of imagery interventions. Typical examples are provided on how to implement imagery interventions. Throughout the chapter practical examples that demonstrate the use of imagery interventions for behavior change will be referenced, alongside an accompanying set of online materials that provide resources that can be used to implement and evaluate imagery interventions (Appendix 33, supplemental materials).

33.2 Definitions Although there are many variations of imagery interventions, there are common components.

Most approaches define imagery as a mental representation of a future event, action, or task, usually with the specification that the person performing the imagery visualizes themselves performing the action, synonymous with a firstperson narrative (Andrade et al., 2016; Conroy & Hagger, 2018; Hamilton et al., 2013; Pham & Taylor, 1999a; Taylor & Schneider, 1989). There is also a general recognition that the imagery has the purpose of increasing the individual’s preparedness or motivation toward the future action. In addition, imagery techniques often require detailed visualization of the action itself and the contextual features alongside internal components such as how imagined action feels and the associated imagined physical, cognitive, and emotional responses. Imagery is often self-directed, initiated by a written text or prompt, but can also be guided by a facilitator. Imagery techniques can also be conducted in group contexts, although each individual within the group will perform the imagery exercise independently. Most interventions that apply imagery techniques to change behavior do not refer to a particular form of imagery (e.g., Chan & Cameron, 2012; Hattar, Hagger, & Pal, 2015; Knäuper et al., 2009; Knäuper et al., 2011; Loft & Cameron, 2013). However, a number of specific forms have gained prominence through their frequent adoption and application such as guided imagery, mental simulations, and functional imagery training. Guided imagery refers to a set of therapeutic techniques with purpose of enhancing individuals’ motivation toward future events by imagining a “positive future” and to practice and rehearse skills useful to future behavioral engagement (Hamilton et al., 2013; Utay & Miller, 2006). Guided imagery has also been used to help individuals explore different outcomes and possibilities of future action and to predict potential problems and work through alternatives. Typically, guided imagery is administered by therapists, counselors, and psychologists in individual one-on-one client-practitioner sessions.

Imagery, Visualization, and Mental Simulation Interventions

The technique usually involves the practitioner talking an individual through imagined scenes relating to the desired behavior or action, with intermittent prompts to guide the images interspersed with periods in which the individual performs the imagery. The imagery will often begin with some exercises to familiarize the client with imagery as a technique, before moving into the exercises in which the behavior is progressively introduced to the client before progressing on to specific actions and scenes. The practitioner provides key phrases and words to focus the client on specific aspects of the imagined scenario, providing means to assist the individual in maintaining a vivid image such as prompting thoughts about anticipated emotional responses and sensory experiences. Guided imagery has also been known to encompass relaxation techniques with the purpose of assisting people to cope with future events that might be stressful or traumatic. The focus of the current chapter is on forms of guided imagery aimed at facilitating behavior change through mental rehearsal of a particular action of behavior because this form of guided imagery is considered most relevant to behavior change. Guided imagery for other purposes such as relaxation does not have the goal of changing behavior and is not directly relevant to the current chapter. Mental simulations are another frequently used form of mental imagery (Pham & Taylor, 1999a, 1999b). Mental simulations are defined as mental rehearsals of future events. Two types of mental simulation have been proposed: outcome and process mental simulations. Outcome mental simulations require individuals to imagine the feelings and positive emotions associated with successfully achieving a behavioral goal such as reducing dietary sugar and fat intake, abstaining from smoking, or being more physically active. Outcome mental simulations prompt an individual to consider the meaning of the behavioral goal and what it would feel like to attain the goal. Process mental simulations require a person to

481

identify the specific sets of actions needed to successfully participate in a desired action or fulfill a desired goal. For example, a process mental simulation for increasing physical activity would require the individual to identify the type of physical activity, when it was going to be performed, for how long and how often it was going to be done, the equipment and location required, whether it will be alone or require another person or team, the financial implications, potential barriers to engaging in the behavior, and how to fit it in with everyday life (Koka & Hagger, 2017). Imagining these processes will provide a personal example (known as a “self-model”) for doing the behavior. Visualizing actually doing the behavior and enacting the relevant steps will increase confidence in doing the behavior successfully. Further approaches use imagery along with other strategies to change behavior. Functional imagery training is a client-centered counseling strategy with imagery as a central strategy as a means to train “functional” behavior change (Andrade et al., 2016). The technique is administered in a one-on-one client-practitioner environment, with the practitioner guiding clients through a series of mental imagery exercises and providing training in performing imagery that the client can then practice at home. Similar to guided imagery, functional imagery is used to explore behavioral goals, identify challenges and problems, and develop potential solutions rather than through discussion. The client is trained to identify discrete, proximal goals that are easily and more vividly visualized, and highly relevant to everyday contexts, than larger, long-term goals. The imagery provides individuals with a behavioral alternative that will compete with the default, well-learned response in critical decision moments, such as choosing between drinking an alcoholic beverage and a nonalcoholic alternative (May et al., 2012). In order to make the imagined behavioral alternatives clearer, more concrete, and more vivid, the client is prompted to identify the advantages of working toward goals, with a

482

MARTIN S. HAGGER AND DOMINIC CONROY

focus on immediate benefits (e.g., positive emotions and a sense of accomplishment for attaining their goals), and visualize them. This aspect of functional imagery training has strong parallels with outcome mental simulations. Clients are also encouraged to imagine past successes with the behavior and to develop plans or courses of action to achieve their goal and visualize them, similar to process mental simulations. Mental contrasting is a recent approach that incorporates imagery as a component (Oettingen, 2012). In mental contrasting, individuals are prompted to imagine a desired outcome related to behavior change and the concomitant positive feelings of achievement and accomplishment and to reflect on the current situation, particularly the problematic and undesirable aspects (e.g., barriers, difficulties, challenges), and to contrast them. The contrasting leads to the activation of expectations to attain the desired (“fantasized”) future outcome and assists in identifying direction in attaining those future outcomes. Although mental contrasting incorporates imagery as a component, the imagery itself is not the component that directly evokes behavior change; rather, it is the contrasting between an idealized imagined future state, similar to outcome mental simulations, with the current status of the behavior that evokes the change. As a consequence, mental contrasting is not specifically an imagery-focused intervention and is not considered further in the current chapter; for a more detailed treatment, readers are directed to theoretical and empirical articles that outline the theory and basis for mental contrasting (Cross & Sheffield, 2019; Oettingen, 2012).

33.3 Theory and Mechanisms of Change Numerous theories have been proposed outlining how mental imagery techniques “work” to enact behavior change (see Sidebar 33.1 and Table 33.1 for details on imagery-related behavior change

techniques). In this section, the potential mechanisms by which imagery interventions work in changing behavior through multiple theorybased mediators are outlined (see Figure 33.1; for more detailed discussion of mechanisms, see Chapters 19, 20, and 46, this volume). Many theories on imagery are derived from the social cognition tradition, which assumes that mental imagery affects behavior through change in attitudes and beliefs. A prominent explanation is provided by Bandura’s (1986) social cognitive theory (see Chapter 3, this volume). According to the theory, observation of others successfully performing a given behavior enhances individuals’ confidence, or self-efficacy, with respect to performing the behavior in future. Coupled with expectations that the behavior will lead to desirable outcomes, outcome expectancies, enhancing self-efficacy is the proposed mechanism by which modeling of the behavior leads to successful enactment of the behavior in future. Imagery taps into this modeling process by offering a “self-model”; rather than observing another perform the behavior successfully, the individual serves as a model for their own behavior using imagery. The modeling process using imagery is therefore expected to enhance self-efficacy and outcome expectancies, leading to increased motivation to perform the behavior in the future. Self-efficacy and outcome expectancies are, therefore, proposed mediators of the effects of interventions using imagery techniques on behavior change (see social cognitive theory pathway, Figure 33.1). A further perspective is offered by the role that imagery plays in assisting in the regulation of emotional reactions that may enhance or derail such efforts (Pham & Taylor, 1999b). Positive emotional states may enhance goal-directed behavior due to the links between positive affect and persistence with behavior and reduce negative emotional states that may impede the behavior such as anxiety and worry (see mental simulation pathway, Figure 33.1).

Imagery, Visualization, and Mental Simulation Interventions

Emotional Regulation

483

Mental simulations (Pham & Taylor, 1999)

Social cognitive theory (Bandura, 1986) Self-Efficacy

Outcome Expectancies Imagery intervention

Behavior Cue Accessibility Cognitive explanation (Kosslyn et al., 2001) Desires

Intrusive thoughts

Elaborated intrusion theory (Kavanagh, Andrade, & May, 2005)

Figure 33.1 Diagram illustrating the potential mechanisms of imagery interventions in changing

behavior

Other theories based on the cognitive tradition focus on the role imagery plays in increasing the accessibility of cues linked to the behavior (Kosslyn, Ganis, & Thompson, 2001). Research has suggested that mental imagery may provide stronger links between environmental and internal cues that are likely to lead to behavioral enactment, perhaps through previous associations and learned patterns of action, and make them more accessible. Cue accessibility, and the associated patterns of action, is an important determinant of behavior. Cue accessibility is therefore another candidate mediator of the impact of imagery interventions on behavior (see cognitive explanation pathway in Figure 33.1). A further cognitive approach to explaining imagery effects comes from elaborated intrusion

theory (Kavanagh, Andrade, & May, 2005). The theory makes the distinction between basic associative processes, which result from learned associations between stimuli in the environment related to an “appetitive target” (e.g., food, cigarettes, alcohol) and behavioral responses, leading to states of desire, and higher cognitive processes that arise from elaboration based on available information regarding the appetitive target stored in memory. Intrusive thoughts toward “appetitive targets” are activated by environmental or physiological cues that become more elaborated thoughts through additional processing, leading to subjective states of desires. According to the theory, imagery is integral to the activation of the state of desire because it is implicated in the cognitive process that elaborates and amplifies

484

MARTIN S. HAGGER AND DOMINIC CONROY

effects of associative inputs by making them richer and more vivid. Functional imagery interventions are proposed to work by providing people with vivid, personally relevant, sensory-based images of the desired goal-directed action that will be activated when desires inconsistent with their goal arise. The imagery can enhance the strength and vividness of the goal-directed avoidance or quitting behavior if linked to the tempting situation. The imagery therefore serves to remind the individual of the intended goal each time the tempting cue arises (see elaborated intrusion theory pathway in Figure 33.1).

33.4 Evidence Base Studies have shown that imagery interventions, broadly defined, have been effective in changing motivation and behavior. Mental simulations have been found to be successful in increasing physical activity (Koka & Hagger, 2017), improving study behavior (Pham & Taylor, 1999a), controlling anxiety and promoting healthy stress management (Armitage & Reidy, 2012; Keech, Hagger, & Hamilton, 2019), reducing alcohol (Conroy, Sparks, & de Visser, 2015; Hagger, Lonsdale, & Chatzisarantis, 2011), healthy eating (Knäuper et al., 2011), achieving health goals (Greitemeyer & Würz, 2005), improving sleep quality (Loft & Cameron, 2013), and promoting academic motivation (Vasquez & Buehler, 2007). In addition, research has also shown effects of imagery interventions on the measure of key constructs implicated in the theoretical mechanisms for their effects. Consistent with explanations derived from social cognitive theory, research has demonstrated effects of imagery interventions on self-efficacy and attitudes (Armitage & Reidy, 2008; Markland et al., 2015; Rennie et al., 2014). In addition, research has demonstrated effects of imagery interventions in reducing anxiety, consistent with Taylor and Schneider’s (1989) hypothesis that imagery works by reducing disruptive

negative affective responses like worry and concerns with respect to the upcoming behavior (Armitage & Reidy, 2012; Pham & Taylor, 1999b). Some studies have also tested the mediation of imagery intervention effects on behavioral outcomes using mediation analysis. Results have, however, been inconclusive, with some studies demonstrating mediation of imagery effects by anxiety reduction (Pham & Taylor, 1999b), selfefficacy (Rennie et al., 2014), and construals of importance (Vasquez & Buehler, 2007), while others have failed to find mediation effects (Conroy et al., 2015; Hagger, Lonsdale, Koka, et al., 2012; Hagger, Lonsdale, & Chatzisarantis, 2012). Overall, there are relatively few tests of mediation of imagery intervention effects, and identifying potential mediation effects is a clear avenue for future research given the observed variability in mediation effects in current tests. Hagger and Conroy (2018) conducted a metaanalytic review of twenty-six studies that applied imagery interventions in health behavior. The majority of the studies used brief, nonguided imagery types. The analysis revealed an overall small-sized average effect on behavior. In addition, the analysis also found small effects of imagery interventions on intention, perceptions of control and self-efficacy, and attitude. Further, a small-to-medium-size effect of imagery interventions on physiological measures of health status or improvement (e.g., weight loss, resting heart rate, body mass index) was found. Overall, the findings suggest that imagery interventions are generally effective in changing behavior, indexes of health, and theory-related mediators in health contexts. However, the review revealed substantial variability in the effect sizes across studies. The investigators examined effects of a number of potential moderator variables expected to magnify or diminish intervention effects: inclusion of additional imagery components at follow-up, whether health-related information was presented prior to the intervention, the type of engagement involved in the imagery intervention (imagery

Imagery, Visualization, and Mental Simulation Interventions

485

Sidebar 33.1 Behavior change techniques relating to mental imagery from existing taxonomies

Mental imagery techniques feature prominently in existing taxonomies of behavior change. In three prominent taxonomies, mental imagery is isolated as a specific irreducible technique. For example, in the widely used behavior change technique taxonomy version 1 (BCTTv1; Michie et al., 2013), mental imagery is captured by the “mental rehearsal of successful performance” entry and is also identified in other taxonomies as the “prompt use of mental imagery” (Michie et al., 2011) and “use imagery” (Kok et al., 2016) entries. A summary of the techniques and associated descriptions from the taxonomies is provided in Table 33.1. Techniques with overlapping or closely related content have also been identified; these include techniques relating to modeling and demonstration of the target behavior, consistent with imagery as a means to “self-model” the behavior.

Table 33.1 Mental imagery techniques matched with techniques from behavior change taxonomies with related techniques and descriptions

Taxonomy

Primary/Closely Technique and Cluster (if any) Related

BCTTv1 (Michie Cluster: 15. Self-belief et al., 2013) Technique: 15.2. Mental rehearsal of successful performance

Primary

Closely Related Cluster: 6. Comparison of behavior Technique: 6.1. Demonstration of the behavior (related to Observational Learning; Bandura, 1986) Closely Related Cluster: 9. Comparison of outcomes Technique: 9.3. Comparative imagining of future outcomes (related to Mental Contrasting; Oettingen, 2012)

CALO-RE (Michie et al., 2011)

Technique: 34. Prompt use of imagery

Primary

Description Advise to practice imagining performing the behavior successfully in relevant contexts (e.g., advise to imagine eating and enjoying a salad in a work canteen). Provide an observable sample of the performance of the behavior, directly in person or indirectly, e.g. via film or pictures, for the person to aspire to or imitate (includes “Modeling”). Prompt or advise the imagining and comparing of future outcomes of changed versus unchanged behavior (e.g., prompt the person to imagine and compare likely or possible outcomes following attending versus not attending a screening appointment). Teach the person to imagine successfully performing the behavior or to imagine finding it easy to perform the behavior, including component or easy versions of the behavior. Continued

486

MARTIN S. HAGGER AND DOMINIC CONROY

Table 33.1 (cont.)

Taxonomy

Intervention Mapping Taxonomy (Kok et al., 2016)

Primary/Closely Technique and Cluster (if any) Related Technique: 22. Model/ Demonstrate the behavior (related to Observational Learning; Bandura, 1986)

Closely Related

Technique: Using imagery

Primary

Technique: Modeling (related to Observational Learning; Bandura, 1986)

Closely Related

Description Involves showing the person how to perform a behavior, e.g. through physical or visual demonstrations of behavioral performance, in person or remotely. Using artifacts that have a similar appearance to some subject.

Providing an appropriate model being reinforced for the desired action.

Note. BCTTv1 = Michie et al.’s (2013) behavior change technique taxonomy version 1.

only or imagery and writing), and health behavior domain (studies on physical activity vs. studies not on physical activity). Results revealed that imagery interventions were more effective in older, nonstudent samples, when detailed instructions on imagery were included, if they were longer in duration, and if the studies were rated of higher quality. These findings illustrate the importance of methodological quality and design in enhancing imagery intervention effects.

and fidelity of the intervention can be evaluated. In addition, a step-by-step guide is provided on how to develop an imagery intervention. The section will be augmented with examples, references to previous interventions, and links to supplemental materials that provide further examples and templates of imagery intervention materials (for additional examples of intervention materials see Hamilton et al., 2019).

33.5.1 Delivery and Format

33.5 Step-by-Step Guide This section addresses important considerations that need to be taken into account when designing interventions using imagery techniques: the type of imagery to be used; the typical means by which imagery interventions are delivered; which populations are likely to benefit from imagery interventions; factors that may facilitate or inhibit the efficacy of the intervention; what sort of training or skills are required to administer an imagery intervention; the appropriate dose or intensiveness of the intervention for them to be effective; the typical materials needed to implement the intervention; and how the efficacy

Imagery interventions have typically been administered as print or text-only interventions (e.g., pamphlet, leaflet, brochure) or as an inperson intervention either through one-on-one client-practitioner sessions or as a group-level intervention led by a practitioner. The appropriate means of delivery will depend on the type of imagery intervention adopted, budget, target audience, type of behavior that requires change, and the expected intensiveness of the intervention. The major component of imagery interventions in both delivery formats is an “exercise,” or series of exercises, in which the participant is

Imagery, Visualization, and Mental Simulation Interventions

prompted or guided to complete a task in which they imagine a scenario related to the behavior or outcome of interest. The individual is often instructed to find a place where they can be comfortable, away from distractions, and, preferably, seated or prone. Participants in imagery exercises are often prompted to close their eyes to minimize visual distractions and enhance the imagery. Guided imagery, by definition, tends to be delivered in person, with the client providing ongoing prompts and instructions. The practitioner will usually work from a script, with a brief introduction to the exercises and some instructions on how to get the most out of the imagery exercises, such as being comfortable, and providing tips such as closing one’s eyes. The practitioner will then proceed with the exercise, often starting with a general imagery exercise (e.g., a “tangy lemon” exercise in which the individual is prompted to imagine cutting into a lemon and visualize the accompanying sensations; Andrade et al., 2016), followed by a series of guided imagery exercises focusing on changing the behavior, interspersed with brief periods allowing for reflection or discussion. Mental simulations are typically administered by print communication, with participants given brief written instructions providing context and rationale for the mental imagery, followed by instructions on how to complete the imagery exercises (Pham & Taylor, 1999b). Participants are sometimes prompted to provide a written narrative of their intervention afterwards to help reinforce the imagery and allow for better recall of the imagery for subsequent occasions (e.g., Hagger, Lonsdale, Koka, et al., 2012; Pham & Taylor, 1999b). Such interventions negate the need for the presence of a facilitator and could, therefore, be delivered in a pamphlet or on a web page. However, in other studies, the passage was read aloud to groups of people by a practitioner while they read along with the paragraph in front of them, which allows the practitioner to emphasize certain points and determine the time spent on the imagery exercise (e.g., Meslot et al., 2016).

487

33.5.2 Target Audience Imagery interventions are appropriate for individuals and groups of people who already have a degree of motivation to change. This is important because the exercises used in imagery interventions generally assume that changing the behavior is something that is of interest to, and has meaning for, the target population. However, they need not have formed clear goals or intentions or have dedicated in-depth thought concerning behavior change; some initial thought or consideration is sufficient. Although most imagery interventions have been used to promote change in community populations, they have also been applied in clinical populations that have serious conditions, such as cardiovascular disease (Meslot et al., 2016) or obesity (Hattar et al., 2015). Imagery interventions are less appropriate for individuals who have not contemplated change (see Chapter 10, this volume) or have no intention to change (see Chapter 2, this volume). In such cases, imagery exercises will lack meaning and be unlikely to foster motivation or self-efficacy to change. Strategies to promote motivation and selfefficacy to assist with intention formation are likely to be appropriate for these populations (see also Chapter 32, this volume). Guided imagery may be used to assist unmotivated individuals or those with no intention to change by exploring possible goals and motives for change (see Chapter 38, this volume). However, this use of guided imagery deviates from the standard use of imagery to change behavior, as it does not focus on visualizing a particular behavior or outcome but likely focuses on reasons and rationale for change.

33.5.3 Enabling or Inhibiting Factors Two prominent enabling factors of imagery interventions are suggested in the research literature: imagery ability and the provision of clear instructions on how to conduct the imagery exercises.

488

MARTIN S. HAGGER AND DOMINIC CONROY

Sidebar 33.2 Defining and measuring imagery “ability”

Imagery ability has been defined, conceptualized, and measured in several ways, with each approach placing distinctive emphasis on how individual differences in imagery ability might be conceived in relation to differences in dispositional skills and experience or practice with using imagery. For example, some researchers have measured imagery ability in terms of ease or familiarity with imagery (Adams et al., 2015; Knäuper et al., 2011; Stanley & Cumming, 2010). Researchers in the physical activity domain have developed the exercise imagery questionnaire (EIQ), drawing on work that has suggested that individuals who exercise regularly use appearance, energy, and technique in their imagery to maximize performance (Hausenblas et al., 1999). In contrast, the exercise imagery inventory (EII; Giacobbi, Hausenblas, & Penfield, 2009) conceptualizes imagery ability in broader terms and comprises four distinct subscales: appearance–health imagery, exercise technique imagery, exercise feelings imagery, and exercise self-efficacy imagery. Other authors have favored measures of imagery that emphasize the quality or vividness of the images produced (Andrade et al., 2016; Babin & Burns, 1998; Chan & Cameron, 2012). Given evidence indicating that ability to visualize is a moderator of imagery effects (Adams et al., 2015; Andrade et al., 2016; Chan & Cameron, 2012), imagery ability offers an important measure for mental imagery practitioners to explore and account for in imagery interventions. Intervention design should seek to put in place means to maximize participants’ capacity to clearly and vividly visualize the subject matter of the exercises of imagery interventions and to sustain those images consistently.

While there is some suggestion that there are individual differences in imagery ability, it is generally acknowledged that imagery is a skill, and individuals who have had experience with imagery exercises express better capability to generate mental images and perform mental imagery exercises effectively that those with no or little experience (see Sidebar 33.2). It is therefore important that practitioners designing imagery interventions provide participants with sufficient practice on imagery tasks in order to develop skills to perform the exercises effectively and increase the likelihood they will gain the concomitant benefits. Guided imagery strategies frequently provide progressive exercises that focus entirely on honing participants’ skills in generating vivid images prior to engaging in exercises that focus on behavior change (see

Appendix 33.1, Example 1, supplemental materials). Imagery interventions using print communication as a typical means of delivery, such as mental simulations, do not usually include exercises to practice or even prime imagery skills. This is usually because the exercises are designed to be time-efficient and conducted in a discrete period. However, providing brief imagery practice exercises alongside mental simulations exercises is still possible and may improve the quality of subsequent images. Although this will extend the time on the exercises, it may pay off in the long run as higher-quality imagery may be more effective in changing behavior. The provision of clear instructions to accompany the imagery exercises has been shown to improve their efficacy (Conroy & Hagger, 2018). Instructions on how to conduct imagery exercises

Imagery, Visualization, and Mental Simulation Interventions

are integral to the guided imagery approach, with the facilitator providing ongoing instruction on how to conduct the exercise and reminders on the important components (e.g., using the senses) (see Appendix 33.1, Example 1, supplemental materials). However, other imagery techniques, such as mental simulations, tend to be “minimalist” by comparison in the detail provided in the instructions for the imagery exercise – they generally ask participants to merely “visualize” the scene related to behavior change (see Appendix 33.1, Example 2, supplemental materials). Mental simulation “scripts” can be augmented to include clearer instructions on how to most effectively conduct the process of visualization itself, such as finding a suitable location away from distractions, being comfortable and able to relax, and keeping eyes closed. The instructions should also extend to describing the kind of things that an imagined scene should contain, including use of all senses rather than vision alone (Knäuper et al., 2011). As with the inclusion of practice exercises, providing instructions on how to visualize will extend the time taken, but the benefits of more effective imagery on outcomes are likely to outweigh the relatively modest time cost. Inclusion of other behavior change techniques alongside the imagery interventions may maximize their benefits. Augmenting imagery exercises with a goal setting activity (see Chapter 38, this volume) may be useful to evoke sufficient rationale for change and make the subsequent imagery exercises relevant. In addition, there is also evidence indicating that including planning exercises alongside imagery interventions may maximize behavior change (Hagger, Lonsdale, Koka, et al., 2012; Knäuper et al., 2011). As imagery exercises generally focus on promoting greater motivation and self-efficacy, including planning exercises such as implementation intentions or action plans (Hagger et al., 2016) alongside them may provide means to implement the visualized behavior (see Chapter 6, this volume).

489

33.5.4 Training and Skills Required Levels of specific practitioner skills required to deliver imagery interventions depend on the type of imagery intervention and its intensiveness. Guided imagery should be conducted with a practitioner who has experience in interpersonal communication in order to administer the imagery “script” to the client or group and to provide appropriate, informed responses to questions regarding the exercises and their content. Formal training as a counselor, however, is not usually necessary. Training to deliver guided imagery interventions should include familiarization with the basic content, theory, and mechanisms behind their use for behavior change as well as a period of supervised practice in the delivery of imagery exercises with a trained facilitator. Imagery interventions delivered by print communication are advantageous because they obviate the need for a practitioner with basic knowledge and skills. In some cases, mental simulation exercises are read aloud by a practitioner, which may facilitate participants’ attention to the exercise and prevent “passive reading,” as well as encourage participants to spend the requisite amount of time on the exercises.

33.5.5 Intensiveness Even though imagery interventions have been shown to be effective in changing behavior in relatively low doses, such as mental simulations practiced for five minutes per day (Pham & Taylor, 1999b) or even just once (e.g., Hagger, Lonsdale, Koka, et al., 2012), meta-analysis of imagery interventions in health contexts indicated that including “follow-up” imagery exercises results in better behavior change (Conroy & Hagger, 2018). The follow-up exercises are often of shorter duration than the initial exercise and can either be practitioner led or be self-administered. If the follow-up exercise is to be self-administered, it is good practice to provide some sort

490

MARTIN S. HAGGER AND DOMINIC CONROY

of reminder such as a text message on a mobile device or an email. The use of diaries to log follow-up sessions has also been shown to be useful (Pham & Taylor, 1999b). There is, however, little evidence indicating the minimum number of sessions required to provide a meaningful effect of imagery interventions. Overall, evidence seems to point to the value of repeating mental simulation exercises regularly for successful behavior change.

33.5.6 Evaluation of Fidelity Maximal efficacy of imagery interventions is dependent on sufficient engagement of participants with the imagery intervention components. In the case of one-on-one client-practitioner administered interventions, like guided imagery, individuals participating in the intervention need to pay attention and respond to the instructions of the practitioner and carry out the instructions as specified. In the case of print communication imagery interventions, such as mental simulation interventions, individuals need to read the instructions and carry out the mental simulation exercises as directed. A number of methods can be used to ensure participants’ compliance with the imagery intervention components in practice. In the case of guided imagery, the practitioner can prompt the individual for acknowledgment and understanding of the exercises and observe whether they comply with the instructions. They may also prompt individuals to provide verbal or written feedback on the content and experience of their imagery experience and analyze the content of their feedback in order to ascertain whether the imagined scenarios are appropriate (for an example, see Koka & Hagger, 2017). For instance, participants in imagery interventions administered by print communication can be prompted to provide detailed written feedback (e.g., diaries) on the content and experience of the exercises (Pham & Taylor, 1999b). Alternatively, participants can be asked to write down their imagined

outcomes and actions immediately after they participated in the exercise (Hagger et al., 2011). The diaries and written scripts can be content analyzed to check that participants have followed the exercises as instructed (e.g., Koka & Hagger, 2017).

33.5.7 Evaluation of Efficacy As with many interventions, tests of efficacy should adopt reliable and valid measures of behavior, preferably more “objective” measures that do not rely on self-report. Effective evaluation also needs to adopt appropriate designs, particularly randomized controlled designs, to evaluate the efficacy of the intervention alongside reasonable control or comparison groups (see Chapters 21 and 22, this volume). It is also important to include measures of psychological factors that are expected to change as a result of participating in imagery interventions. Previous research has included measures of motivation, intentions, self-efficacy, and attitudes as potential psychological factors that are likely to change as a result of participation in imagery exercises (e.g., Chan & Cameron, 2012; Conroy et al., 2015; Hagger et al., 2011; Knäuper et al., 2011). For example, imagery exercises aimed at changing behavior by targeting change in self-efficacy, as specified by theory and an intervention “logic model,” should use measures of self-efficacy in mediation analyses to explain the effects of the mental simulation intervention on behavior change (see Chapters 19 and 20, this volume). Example measures of theory-based mediators of imagery interventions (e.g., intentions, motivation, attitudes, self-efficacy) are provided in Appendix 33.2 (supplemental materials).

33.5.8 Typical Materials Needed The main materials needed to implement imagery interventions are the scripts developed to deliver the intervention. The content of the

Imagery, Visualization, and Mental Simulation Interventions

script will be determined by the type of imagery intervention adopted. Guided imagery interventions usually require a printed script that details the instructions, imagery exercise, and closing statements to be provided by the practitioner to the client (see Appendix 33.1, Example 1, supplemental materials). For print imagery interventions delivered by print communication, printed materials (e.g., forms, sheets, pamphlets, leaflets, brochures) providing information about, and description of, the exercise may be all that is needed (see Appendix 33.1, Examples 2–6, supplemental materials). The scripts may also be accompanied by forms or preprinted diary pages or logbooks that the participant may need to complete either as part of the session or during the course of engaging in the technique or strategy on their own (see Appendix 33.1, Examples 2, 4, and 5, supplemental materials). In cases where intervention efficacy is to be evaluated, self-report measures of behavior and candidate mediators will also be required (see Appendix 33.2, supplemental materials). If the intervention is to be delivered online, or via a device like a mobile phone, means to communicate with the participant and software to collect responses to typewritten scripts and self-report measures are required.

33.6 Summary and Conclusion Interventions adopting imagery and visualization strategies to change behavior are gaining popularity and attention in the scientific literature and among practitioners, given their relatively simplicity and demonstrable effectiveness. The current chapter has provided an overview of the theory behind imagery and visualization interventions (e.g., social cognitive theory, management of emotional reactions, cue accessibility, elaborated intrusion theory), focusing on the different types available (e.g., guided imagery, mental simulations,

491

functional imagery training) and their content as well as the evidence for their efficacy. It has also provided some indication of the main behavior change techniques that have been used in imagery interventions (e.g., mental rehearsal of successful performance, prompt use of imagery), the kinds of factors that may enhance their efficacy (e.g., provision of detailed instructions, application in older samples, higher-quality designs, longer duration), and the kinds of behaviors, contexts, and populations in which they are more likely to be effective, based on current evidence. The chapter also provides a step-by-step guide that outlines the important issues that interventionists considering using imagery interventions should take into account with developing, implementing, and evaluating imagery interventions. Key considerations include the format of the imagery intervention; selection of the appropriate target audience (e.g., individuals with some level of motivation to change); incorporating means to promote factors likely to enable imagery interventions (e.g., enhancement of imagery ability, provision of clear instructions, inclusion of other behavior change techniques alongside imagery); identification of training and skills required to promote effective imagery; providing the appropriate dose (frequency and duration) of imagery; including measures to evaluate fidelity of the imagery strategies (e.g., diaries, prompt description of image content); and including measures to evaluate the efficacy of the intervention (e.g., valid and reliable measures of behavior or outcomes) and its theory-based processes (e.g., measures of self-efficacy). Research evaluating the efficacy of imagery and visualization interventions has supported their use with small-to-medium effect sizes. However, more systematic, high-quality research is needed, particularly examining the key moderators of effects and in a broader set of behaviors, contexts, and populations.

492

MARTIN S. HAGGER AND DOMINIC CONROY

References Adams, C., Rennie, L., Uskul, A. K., & Appleton, K. M. (2015). Visualising future behaviour: Effects for snacking on biscuit bars, but no effects for snacking on fruit. Journal of Health Psychology, 20, 1037–1048. https://doi.org/10.1177/ 1359105313506760 Andrade, J., Khalil, M., Dickson, J., May, J., & Kavanagh, D. J. (2016). Functional imagery training to reduce snacking: Testing a novel motivational intervention based on elaborated intrusion theory. Appetite, 100, 256–262. https:// doi.org/10.1016/j.appet.2016.02.015 Armitage, C. J., & Reidy, J. G. (2008). Use of mental simulations to change theory of planned behaviour variables. British Journal of Health Psychology, 13, 513–524. https://doi.org/10.1348/ 135910707x227088 Armitage, C. J., & Reidy, J. G. (2012). Evidence that process simulations reduce anxiety in patients receiving dental treatment: Randomized exploratory trial. Anxiety Stress and Coping, 25, 155–165. https://doi.org/10.1080/ 10615806.2011.604727 Babin, L. A., & Burns, A. C. (1998). A modified scale for the measurement of communicationevoked mental imagery. Psychology and Marketing, 15, 261–278. https://doi.org/ 10.1002/(SICI)1520-6793(199805)15:33.0.CO;2-8 Bandura, A. (1986). Social Foundations of Thought and Action: A Social-Cognitive Theory. Englewood Cliffs, NJ: Prentice Hall. Chan, C. K. Y., & Cameron, L. D. (2012). Promoting physical activity with goal-oriented mental imagery: A randomized controlled trial. Journal of Behavioral Medicine, 35, 347–363. https://doi.org/10.1007/s10865-0119360-6 Conroy, D., & Hagger, M. S. (2018). Imagery interventions in health behavior: A meta-analysis. Health Psychology, 37, 668–679. https://doi.org/ 10.1037/hea0000625 Conroy, D., Sparks, P., & de Visser, R. (2015). Efficacy of a non-drinking mental simulation intervention for reducing student alcohol consumption. British

Journal of Health Psychology, 20, 688–707. https://doi.org/10.1111/bjhp.12133 Cross, A., & Sheffield, D. (2019). Mental contrasting for health behaviour change: A systematic review and meta-analysis of effects and moderator variables. Health Psychology Review, 13, 209–225. https:// doi.org/10.1080/17437199.2019.1594332 Giacobbi, P. R., Jr., Hausenblas, H. A., & Penfield, R. D. (2009). Further refinements in the measurement of exercise imagery: The exercise imagery inventory. Measurement in Physical Education and Exercise Science, 9, 251–266. https://doi.org/10.1207/s15327841mpee0904_4 Greitemeyer, T., & Würz, D. (2005). Mental simulation and the achievement of health outcomes: The role of goal difficulty. Imagination, Cognition and Personality, 25, 239–251. https://doi.org/10.2190/ D4UA-RQFQ-0H5T-W9YY Hagger, M. S., Lonsdale, A., Koka, A. et al. (2012). An intervention to reduce alcohol consumption in undergraduate students using implementation intentions and mental simulations: A crossnational study. International Journal of Behavioral Medicine, 19, 82–96. https://doi.org/ 10.1007/s12529-011-9163-8 Hagger, M. S., Lonsdale, A. J., & Chatzisarantis, N. L. D. (2011). Effectiveness of a brief intervention using mental simulations in reducing alcohol consumption in corporate employees. Psychology, Health and Medicine, 16, 375–392. https://doi .org/10.1080/13548506.2011.554568 Hagger, M. S., Lonsdale, A. J., & Chatzisarantis, N. L. D. (2012). A theory-based intervention to reduce alcohol drinking in excess of guideline limits among undergraduate students. British Journal of Health Psychology, 17, 18–43. https://doi.org/ 10.1111/j.2044-8287.2010.02011.x Hagger, M. S., Luszczynska, A., de Wit, J. et al. (2016). Implementation intention and planning interventions in health psychology: Recommendations from the Synergy expert group for research and practice. Psychology and Health, 31, 814–839. https://doi.org/10.1080/ 08870446.2016.1146719 Hamilton, J., Fawson, S., May, J., Andrade, J., & Kavanagh, D. J. (2013). Brief guided imagery and body scanning interventions reduce food cravings.

Imagery, Visualization, and Mental Simulation Interventions

Appetite, 71, 158–162. https://doi.org/10.1016/ j.appet.2013.08.005 Hamilton, K., Keech, J. J., Peden, A., & Hagger, M. S. (2019). A protocol for developing a mental imagery intervention: A randomised controlled trial testing a novel implementation imagery e-health intervention to change driver behaviour during floods. BMJ Open, 9, e025565. https:// doi.org/10.1136/bmjopen-2018-025565 Hattar, A., Hagger, M. S., & Pal, S. (2015). Weight-loss intervention using implementation intentions and mental imagery: A randomised control trial study protocol. BMC Public Health, 15, 196. https://doi .org/10.1186/s12889-015-1578-8 Hausenblas, H. A., Hall, C. R., Rodgers, W. M., & Munroe, K. J. (1999). Exercise imagery: Its nature and measurement. Journal of Applied Sport Psychology, 11, 171–180. https://doi.org/10.1080/ 10413209908404198 Kavanagh, D. J., Andrade, J., & May, J. (2005). Imaginary relish and exquisite torture: The elaborated intrusion theory of desire. Psychological Review, 112, 446–467. https://doi .org/10.1037/0033-295X.112.2.446 Keech, J. J., Hagger, M. S., & Hamilton, K. (2019). Changing stress mindsets with a novel imagery intervention: A randomized controlled trial. Emotion. https://doi.org/10.1037/emo0000678 Knäuper, B., McCollam, A., Rosen-Brown, A., Lacaille, J., Kelso, E., & Roseman, M. (2011). Fruitful plans: Adding targeted mental imagery to implementation intentions increases fruit consumption. Psychology and Health, 26, 601– 617. https://doi.org/10.1080/08870441003703218 Knäuper, B., Roseman, M., Johnson, P. J., & Krantz, L. H. (2009). Using mental imagery to enhance the effectiveness of implementation intentions. Current Psychology, 28, 181–186. https://doi.org/ 10.1007/s12144-009-9055-0 Kok, G., Gottlieb, N. H., Peters, G.-J. Y. et al. (2016). A taxonomy of behavior change methods: An intervention mapping approach. Health Psychology Review, 10, 297–312. https://doi.org/ 10.1080/17437199.2015.1077155 Koka, A., & Hagger, M. S. (2017). A brief intervention to increase physical activity behavior among adolescents using mental simulations and action

493

planning. Psychology, Health and Medicine, 22, 701–710. https://doi.org/10.1080/ 13548506.2016.1211298 Kosslyn, S. M., Ganis, G., & Thompson, W. L. (2001). Neural foundations of imagery. Nature Reviews Neuroscience, 2, 635. https://doi.org/10.1038/ 35090055 Loft, M. H., & Cameron, L. D. (2013). Using mental imagery to deliver self-regulation techniques to improve sleep behaviors. Annals of Behavioral Medicine, 46, 260–272. https://doi.org/10.1007/ s12160-013-9503-9 Markland, D., Hall, C. R., Duncan, L. R., & Simatovic, J. (2015). The effects of an imagery intervention on implicit and explicit exercise attitudes. Psychology of Sport and Exercise, 17, 24–31. https://doi.org/10.1016/j.psychsport.2014.11.007 May, J., Andrade, J., Kavanagh, D. J., & Hetherington, M. (2012). Elaborated intrusion theory: A cognitive-emotional theory of food craving. Current Obesity Reports, 1, 114–121. https://doi .org/10.1007/s13679-012-0010-2 Meslot, C., Gauchet, A., Allenet, B., François, O., & Hagger, M. S. (2016). Theory-based interventions combining mental simulation and planning techniques to improve physical activity: Null results from two randomized controlled trials. Frontiers in Psychology, 7, 1789. https://doi.org/ 10.3389/fpsyg.2016.01789 Michie, S., Ashford, S., Sniehotta, F. F., Dombrowski, S. U., Bishop, A., & French, D. P. (2011). A refined taxonomy of behaviour change techniques to help people change their physical activity and healthy eating behaviours: The CALO-RE taxonomy. Psychology and Health, 26, 1479–1498. https:// doi.org/10.1080/08870446.2010.540664 Michie, S., Richardson, M., Johnston, M. et al. (2013). The behavior change technique taxonomy (v1) of 93 hierarchically clustered techniques: Building an international consensus for the reporting of behavior change interventions. Annals of Behavioral Medicine, 46, 81–95. https://doi.org/ 10.1007/s12160-013-9486-6 Murphy, S. M., Nordin, S., & Cumming, J. (2008). Imagery in sport, exercise, and dance. In T. S. Horn (Ed.), Advances in Sport Psychology (3rd ed., pp. 297–324). Champaign, IL: Human Kinetics.

494

MARTIN S. HAGGER AND DOMINIC CONROY

Neck, C. P., & Manz, C. C. (1996). Thought selfleadership: The impact of mental strategies training on employee cognition, behavior, and affect. Journal of Organizational Behavior, 17, 445–467. https://doi.org/10.1002/(SICI)10991379(199609)17:53.0.CO;2-N Oettingen, G. (2012). Future thought and behaviour change. European Review of Social Psychology, 23, 1–63. https://doi.org/10.1080/ 10463283.2011.643698 Pham, L. B., & Taylor, S. E. (1999a). The effect of mental simulation on goal-directed performance. Imagination, Cognition and Personality, 18, 253– 268. https://doi.org/10.2190/VG7 L-T6HK-264 H-7XJY Pham, L. B., & Taylor, S. E. (1999b). From thought to action: Effects of process- versus outcomebased mental simulations on performance. Personality and Social Psychology Bulletin, 26, 250–260. https://doi.org/10.1177/01461672990 25002010 Rennie, L. J., Harris, P. R., & Webb, T. L. (2014). The impact of perspective in visualizing health-related

behaviors: First-person perspective increases motivation to adopt health-related behaviors. Journal of Applied Social Psychology, 44, 806– 812. https://doi.org/10.1111/jasp.12266 Stanley, D. M., & Cumming, J. (2010). Are we having fun yet? Testing the effects of imagery use on the affective and enjoyment responses to acute moderate exercise. Psychology of Sport and Exercise, 11, 582–590. https://doi.org/10.1016/ j.psychsport.2010.06.010 Taylor, S. E., & Schneider, S. K. (1989). Coping and the simulation of events. Social Cognition, 7, 174–194. https://doi.org/10.1521/ soco.1989.7.2.174 Utay, J., & Miller, M. (2006). Guided imagery as an effective therapeutic technique: A brief review of its history and efficacy research. Journal of Instructional Psychology, 33, 40–43. Vasquez, N. A., & Buehler, R. (2007). Seeing future success: Does imagery perspective influence achievement motivation? Personality and Social Psychology Bulletin, 33, 1392–1405. https://doi .org/10.1177/0146167207304541

34 Affect-Based Interventions Mark Conner, David M. Williams, and Ryan E. Rhodes

Practical Summary This chapter looks at ways to change behavior that focus on changing feelings or affect. Different types of affect or feelings are defined and the evidence showing how to change affect is reviewed. Step-by-step guides are provided in relation to changing how people feel about performing a behavior (affective attitudes) and how people might feel after performing a behavior (anticipated affect). Research studies using ifthen plans (implementation intentions), messages to emphasize the negative outcomes of a behavior (fear appeals), presenting the behavior paired with a positive or negative evaluation (evaluative conditioning), and targeting affect in exercise games are also presented. Finally, research on minimizing the impact of feelings or affect on behavior (focusing attention away from feelings/affect, reducing the intensity of feelings/affect) is considered.

34.1 Introduction Affect has been defined as an evaluative neurobiological state that manifests in (1) coordinated patterns of physiological (e.g., release of hormones, increased heart rate) and involuntary behavioral (e.g., facial expression, vocalization) changes and (2) subjective experiential feelings (e.g., the phenomenal experience of pleasure, anger, embarrassment, etc.) (Williams et al., 2018). Such a definition of affect is consistent with its use as an umbrella term and encompasses a range of interrelated concepts. These importantly include core affect (e.g., hedonic responses such as pleasure/displeasure and arousal), emotions (e.g., anger, fear, sorrow, joy), and moods (e.g., happy, contented, depressed, irritable) (Davidson, Scherer, & Goldsmith, 2009; Lewis, Haviland-Jones, & Barrett, 2008), and multiple other divisions of these affect components (for a

taxonomy of affect constructs, see Rhodes, Williams, & Conner, 2018). Multiple theorists converge on suggesting all affect includes a core affect component (Larsen, 2000; Russell, 1980; Thayer, 1978; Watson & Tellegen, 1985). In the circumplex model, core affect is characterized by two orthogonal dimensions: a “valence” dimension ranging from positive to negative and an “activation” dimension ranging from high to low (Russell, 1980). Core affect is held to be ever-present when conscious and awake, although not always the focus of one’s attention (Russell, 1980). Although shifts in direction or magnitude of core affect do not require cognitive appraisals (e.g., stubbing one’s toe), changes in core affect may underlie more complex appraisal-based emotions and moods (Russell & Barrett, 1999). https://doi.org/10.1017/9781108677318.034

496

MARK CONNER, DAVID M. WILLIAMS, AND RYAN E. RHODES

Emotions (e.g., anger, fear, sorrow, joy) involve cognitive appraisals of a specific stimulus, which lead to a combination of coordinated and distinctive physiological and/or behavioral responses and experiential feelings with an underlying core affect component (Frijda, 2008). For example, the emotion of anger involves the appraisal that one has been wronged, accompanied by increased heart rate, flushed skin, a scowling facial expression, and an increase in negative activated affect, culminating in the distinctive phenomenal experience (i.e., feeling) of anger. Moods (e.g., happy, contented, anxious, depressed/sad, irritable) involve the same components as emotions (i.e., cognitive appraisal, change in core affect, physiological, behavioral, and experiential manifestations) but, relative to emotions, are more diffuse, less time limited, and less focused on a specific stimulus (Morris, 1999). Thus, relative to the emotion of anger, an irritable mood may not be attributable to any specific stimulus and can last for days or weeks, with no distinct beginning or end. Given that core affect is experienced “in the moment,” researchers and practitioners are often faced with understanding thoughts about affect pertaining to a particular behavior. Affect processing, affective judgments, and cognitively mediated affect are umbrella terms that encompass cognitive processing of previous or anticipated affective responses to the target behavior (Conner et al., 2015; Rhodes et al., 2009; Williams & Evans, 2014). Key constructs in this category are anticipated affect and affective attitudes, although several other interrelated constructs also fit under this category (for a full taxonomy, see Rhodes, Williams, & Conner, 2018). Conceptually, affect processing is posited to partially or completely mediate the effects of core affect on future behavior (Williams et al., 2018). Anticipated affect is an expectation of one’s explicit affective response to the target behavior, consistent with the broader notion of outcome expectancy in social cognitive theories of behavior (Conner & Norman, 2015; see also Chapters 2 and

3, this volume). Affective attitudes are explicit evaluations of the target behavior based on aggregation of anticipated affective responses.

34.2 Theory and Mechanisms of Change and Evidence Base As noted in Section 34.1, a variety of theories include affect components, although it is a central focus in relatively few (for exceptions, see Cabanac, 1992; Johnston, 2003). The most common approach to changing behavior via affectrelated constructs is through direct modification. This can focus on both core affect or affect processing. For example, McCarthy et al. (2018) focus on direct effects of core affect on behavior. Other research looks at direct versus mediated effects of affect (e.g., Conner, 2018; Rhodes, Williams, & Conner, 2018; Sheeran et al., 2018). A common approach in this area has been on various forms of messages targeting different types of affect (Conner, 2018; Day & Coups, 2018; Rhodes & Gray, 2018). This includes messages targeting affective attitudes (Conner, 2018), anticipated affect (Conner, 2018; Ellis et al., 2018), and fear appeals (White & Albarracín, 2018). Other work has used a range of other means to change affect constructs in order to change behavior, including affect-based games (Rhodes & Gray, 2018), evaluative conditioning (Hollands et al., 2011), and implementation intentions (Sheeran et al., 2018). Other experimental manipulations of affective determinants have also been considered in a limited number of studies (e.g., Day & Coups, 2018; Kiviniemi & Klasko-Foster, 2018; Rhodes & Gray, 2018; Sheeran et al., 2018; Weirs et al., 2018). The state of this research suggests that some effects on behavior change may be possible through directly targeting the affect construct as a putative mediator, although the effect sizes may be only small-to-moderate in magnitude (Ellis et al., 2018). This area of research is in its infancy,

Affect-Based Interventions

with only a handful of studies in any particular behavioral domain. This route of affect-behavior intervention is an important area for sustained future research as it represents the most straightforward approach to intervention (Rhodes, Williams, & Conner, 2018).

34.3 Step-by-Step Guide Additional studies in this relatively new area will hopefully clarify which behavior change methods or techniques (see Chapter 20, this volume) are most effective and how they are best implemented to produce behavior change. Sidebar 34.1 outlines techniques that have been used to change behavior that relate to affect, derived from existing taxonomies of behavior change techniques. Nevertheless, it is possible to provide some stepby-step advice in relation to the application of some of the more well-used techniques to change affective determinants of behavior. The present section focuses on two affect-focused interventions that have received more attention: messages that target change in affective attitudes and anticipated affect. It is notable that both focus on cognitively mediated affect rather than affect proper. In a set of sidebars and appendixes, more limited attention is given to four other affect-based interventions: evaluative conditioning (Sidebar 34.2), exercise games (Sidebar 34.3), implementation intentions (Appendix 34.1), and fear appeals (Appendix 34.2).

34.3.1 Messages Targeting Affective Attitudes There have been a number of experimental studies showing the value of messages targeting affective attitude (sometimes termed experiential attitudes) on behavior change. For example, Sirriyeh et al. (2010) showed that receiving a daily affectively oriented text (SMS) message (i.e., “physical activity is enjoyable”) over a two-week period compared to a cognitive (i.e.,

497

“physical activity is beneficial”) or a combined message (i.e., “physical activity is enjoyable” and “physical activity is beneficial”) was sufficient to significantly increase self-reported physical activity. This technique maps onto 5.6. Information about emotional consequences in Michie et al.’s (2013) BCTTv1 (see Chapters 19 and 20, this volume).

34.3.1.1 Typical Means of Delivery Messages targeting affective attitudes have been delivered in a number of ways, mainly in relation to health behaviors. For example, Sirriyeh et al. (2010) and Carfora et al. (2016) used daily text messages, while Parrott et al. (2008) used a positively framed email message partly targeting affective attitude. Conner et al. (2011) and Morris et al. (2015) used more detailed written messages presented along with images. To date, there has been no comparison of the effectiveness of different means of delivery or tests of video messages targeting affective attitudes.

34.3.1.2 Target Audience and Behaviors The audiences and behaviors targeted in messages targeting affective attitudes have been limited. The majority of studies have focused on students (Carfora et al., 2016; Conner et al., 2011; Morris et al., 2015) or adolescents (Sirriyeh et al., 2010). Studies have focused on changing physical activity (e.g., Sirriyeh et al., 2010) or dietary (Carfora et al., 2016) behaviors. Studies using these messages outside the health domain could not be found. This is an important limitation, although there is no reason to believe the technique could not be used to change environmental, safety, political, or educational behaviors.

34.3.1.3 Enabling or Inhibiting Factors Only one study has examined factors that might moderate the effectiveness of messages targeting affective attitudes. Conner et al. (2011, Study 2) showed an affective message designed to increase affective attitudes toward physical activity to be

498

MARK CONNER, DAVID M. WILLIAMS, AND RYAN E. RHODES

Sidebar 34.1 Behavior change technique(s) from existing taxonomies

Affective intervention techniques feature prominently in existing taxonomies of behavior change. For example, in the widely used behavior change technique taxonomy version 1 (BCTTv1; Michie et al., 2013), cluster 5 Natural consequences, includes 5.2 Salience of consequences, 5.4 Monitoring of emotional consequences, 5.5 Anticipated regret, and 5.6 Information about emotional consequences. While cluster 11 Regulation, includes 11.2 Reduce negative emotions. A summary of the techniques and associated descriptions from the taxonomies is provided in Table 34.1. Table 34.1 Affective intervention techniques matched with techniques from behavior change taxonomies with related techniques and descriptions Taxonomy BCTTv1 (Michie et al., 2013)

Intervention Mapping Taxonomy (Kok et al., 2016)

Technique and Cluster (if any)

Primary/ Closely Related Description

Primary Cluster: 5. Natural consequences Technique: 5.2. Salience of consequences Technique: 5.4. Monitoring of emotional consequences Technique: 5.5. Anticipated regret Technique: 5.6. Information about emotional consequences

Messages focused on anticipated affect related to techniques 5.2, 5.4, 5.5, and 5.6. Messages on affective attitudes related to technique 5.6. Related to direct modification of other sources of behavioral influence in order to overcompensate for the impact of other affect influences; direct modification of the affect constructs; or intervention upon moderators of the affect-behavior link. Cluster: 11. Regulation Closely Related Related to direct modification of other sources of Technique: 11.2. Reduce behavioral influence and negative emotions direct modification of the affect constructs. Technique: Persuasive Primary Messages to tackle affective communication attitudes or anticipated affect. Technique: Implementation intentions Technique: Anticipated regret

Primary

Primary

Used to change affect or moderate impact of affect on behavior. A focus of many anticipated affect studies changed through persuasive messages or salience change.

Note. BCTTv1 = Michie et al.’s (2013) Behavior Change Technique Taxonomy Version 1.

Affect-Based Interventions

499

Sidebar 34.2 Evaluative conditioning

Evaluative conditioning is a process that involves the repeated pairing of an attitude object (the conditioned stimulus) with a positively or negatively valenced stimulus (unconditioned stimulus) (De Houwer et al., 2001). For example, Hollands et al. (2011) repeatedly paired images of snack foods with images of potentially adverse health consequences (e.g., pairing images of chocolate with images of an obese individual). Such studies generally change implicit attitudes more than explicit attitudes toward the object but have been shown to also result in changes to behavior. In the Hollands et al. (2011) study, the pairing of snack food with negative images compared to presenting the snack food images alone resulted in implicit affective attitudes toward snacks that were significantly more negative (a small-sized effect). It also resulted in significantly more choice of healthy over unhealthy snacks (a medium-sized effect). Importantly, there were no changes in explicit attitudes and the changes in implicit attitudes partially mediated the impact of the intervention (evaluative conditioning) on snack food choices. Evaluative conditioning falls under technique 17.8 Associative learning in Michie et al.’s (2013) BCTTv1. A meta-analysis of evaluative conditioning studies (Hofmann et al., 2010) reported a medium-sized effect on behavior (d+ = 0.52, k = 214). Although many of these tests were small-scale laboratory studies with students, they did cover a range of behaviors, including health, racial discrimination, and consumer products. Further tests of evaluative conditioning in relation to behavior change and the mediating role of changes in implicit affective attitudes are required. These studies could usefully examine a broader range of behaviors, be conducted outside the laboratory, and use nonstudent populations. Nevertheless, evaluative conditioning represents a promising technique for changing behavior via changing implicit affective attitudes and may be one of the few effective techniques for changing hedonic motivation.

more effective in increasing physical activity for those high in need for affect (i.e., a tendency to focus on affect over cognition). The suggestion was that there may be individual differences in the way individuals respond to such information. Need for affect might also be expected to influence the impact of affect on behavior, a moderating effect, and could therefore influence interventions aiming to moderate the impact of affect on behavior.

34.3.1.4 Training and Skills Required As interventions based on developing messages targeting affect are relatively simple, little in the way of training or skills is required to implement

this intervention technique. Pilot-testing of messages to ensure impact on changing affective attitudes might be expected to be useful, although most studies do not report details of such work.

34.3.1.5 Intensiveness The messages targeting affective attitudes tested to date are generally low in intensiveness. For example, a single exposure to longer messages (e.g., Conner et al., 2011) or repeated exposure to shorter (text) messages (e.g., Carfora et al., 2016) is common. Text message studies have typically used a single message per day and sent messages over a limited number of days (7–21 days is typical). Further research might

500

MARK CONNER, DAVID M. WILLIAMS, AND RYAN E. RHODES

usefully assess dose-response effects of messages targeting affective attitudes.

34.3.1.6 Evaluation of Fidelity Studies in this area have generally not assessed fidelity beyond ensuring the messages were received. Several studies have examined the extent to which the messages were effective in changing the proposed mechanism of action (i.e., changes in affective attitudes) and tested whether such changes mediate the impact of the intervention on behavior (Carfora et al., 2016; Conner et al., 2011). For example, Carfora et al. (2016) showed that the effects of daily text messages targeting affective attitudes on self-reported fruit and vegetable consumption were explained by changes in affective attitude. Further studies focusing on fidelity, including message engagement, would be useful (see Chapter 22, this volume).

effects on behavior (d+ = 0.43, k = 32). These are comparable effect sizes to those reported for a broader range of health behaviors where interventions have changed overall (instrumental and affective) attitudes. For example, Sheeran et al. (2016) reported that studies changing overall attitudes about health behaviors were associated with small-to-medium effects on intentions (d+ = 0.48, k = 59) and behavior (d+ = 0.38, k = 67). There is, as yet, no published, broad meta-analysis of studies that change affective attitudes and observe effects on behavior. It would be useful if future studies in this area could demonstrate these effects on objectively measured behavior.

34.3.1.8 Typical Materials Needed Messages targeting affective attitudes need to be developed and pilot-tested for each behavior and setting.

34.3.1.9 Typical Examples of Implementation

34.3.1.7 Evaluation of Effectiveness Meta-analyses of experimental studies appear to support the idea that changing affective attitude can be a useful way to change behavior, although the reported effect sizes tend to be small (Rhodes et al., 2009; Sheeran et al., 2014). In relation to physical activity, a review by Rhodes, Gray, and Husband (2019) showed interventions targeting affective judgments (similar to affective attitudes) were associated with small-to-medium-sized

Table 34.2 provides examples of messages used to change affective constructs in several studies. For example, Carfora et al. (2016) randomly allocated adolescents to receive no messages, messages about the instrumental benefits, or messages about the affective benefits of fruit and vegetable intake (for a full list of messages used, see Appendix 34.3, supplemental materials). Those in the message conditions received daily text messages for fourteen days. Findings

Table 34.2 Examples of simple messages used to change affective attitudes Study

Behavior

Example Message

Carfora et al. (2016)

Fruit and vegetable consumption

Carfora et al. (2017) Sirriyeh et al. (2010)

Processed meat consumption Physical activity

“A study on 80,000 British people found that higher consumption of fruit and vegetables coincides with a higher sense of satisfaction and well-being. A diet rich in vegetables may have a beneficial effect on mood not only in the long term, but also dayto-day, influencing the positive emotions of individuals daily.” “Think about regret that you could experience if this week you exceed the recommended portions of processed meat.” “Physical activity can make you feel cheerful. What activity will you do today?”

Affect-Based Interventions

showed that messages significantly increased fruit and vegetable intake, particularly in the affective message condition, and this effect was partially mediated by changes in affective attitude and intentions toward fruit and vegetable intake.

34.3.2 Anticipated Affect There have been a number of studies targeting anticipated affect as a means to change behavior, sometimes termed affective forecasting (Ellis et al., 2018). A particular focus has been on anticipated regret.

34.3.2.1 Typical Means of Delivery Two types of study are common in this area. In the first type, the salience of anticipated affect is manipulated by asking questions tapping this form of affect or not (e.g., Sandberg & Conner, 2009; “If I did not attend for a cervical smear in the next few weeks I would feel regret, definitely no–definitely yes”). In the second type, messages are used to change degree of anticipated affect (e.g., Dillard et al., 2010). Both types map onto technique 5.5 Anticipated regret in Michie et al.’s (2013) taxonomy of behavior change techniques, while the former also maps onto technique 5.2 Salience of consequences, and the latter also maps onto technique 5.6 Information about emotional consequences (Table 34.1).

34.3.2.2 Target Audience and Behaviors Studies have examined effects of behavioral interventions targeting change in anticipated affects in young adult (Abraham & Sheeran, 2004), student (Sandberg & Conner, 2011), and general adult population (Godin et al., 2010) samples. These studies have exclusively focused on health behaviors. Behaviors examined include exercise (Sandberg & Conner, 2011), unsafe sex (Abraham et al., 2004), screening (Dillard et al., 2010), genetic testing (Fisher et al., 2012), blood donation (Godin et al., 2010), alcohol consumption (Murgraff et al., 1999), and organ donation

501

(O’Carroll et al., 2016). Future studies need to explore impacts in non-health behaviors. For example, environmental (e.g., “If I do not recycle that plastic bottle, I might regret it later”), voting (e.g., “If I do not use my right to vote, I might regret it later”), or studying (e.g., “If I do not study hard now, I will probably regret it when I get my results”) behaviors all seem useful foci for interventions targeting anticipated regret.

34.3.2.3 Enabling or Inhibiting Factors There have been few direct tests of the enabling or inhibiting factors for change techniques targeting anticipated affect. Ellis et al.’s (2018) metaanalysis revealed that significantly larger effects were associated with studies containing a larger proportion of women. Tests of whether this moderating effect also emerges in primary studies is required.

34.3.2.4 Training and Skills Required Interventions targeting salience of anticipated affect by, for example, asking questions require little in the way of training and skills. Writing messages or narratives to change anticipated affect may require greater training and skills to identify persuasive messages/narratives for particular combinations of behaviors and populations.

34.3.2.5 Intensiveness Interventions targeting salience and change in anticipated affect are low in intensiveness. Future studies might usefully address doseresponse relationships.

34.3.2.6 Evaluation of Fidelity Studies in this area have generally not assessed fidelity of anticipated affect interventions. Several studies have examined the extent to which messages were effective in changing anticipated affect and tested whether such changes mediated the impact of the intervention on behavior, the proposed mechanism of action. For example, Ellis et al.’s (2018) meta-analysis revealed that studies were effective in changing

502

MARK CONNER, DAVID M. WILLIAMS, AND RYAN E. RHODES

anticipated affect (d+ = 0.24, k = 20). Carfora et al. (2017) showed that daily text messages reminding people of recommendations on processed meat consumption produced changes in behavior that were explained by changes in anticipated regret and intentions. Further research on the fidelity of interventions targeting anticipated affect is required.

34.3.2.7 Evaluation of Effectiveness Sheeran et al. (2014) reported that studies that had changed anticipatory emotions such as fear and worry (d+ = 0.72, k = 107) were associated with small changes in intention (d+ = 0.31, k = 97) and behavior (d+ = 0.21, k = 46). See Sidebar 34.3 for a brief overview of work in the related area of fear appeals. In a more focused review, Ellis et al. (2018) meta-analyzed thirty-seven studies targeting change in anticipated affect in relation to health behaviors. These studies mainly focused on anticipated regret and tried to change either anticipated affect using messages and narrative text (k = 14; e.g., Dillard et al., 2010) or the salience of anticipated affect using questions (k = 23; e.g., Sandberg & Conner, 2009). Ellis et al. (2018) reported a small but significant effect size across these studies on behavior (d+ = 0.29, k = 18). There appeared to be no difference in effect size for behavior in studies attempting to change anticipated affect directly or change the salience of anticipated affect, although all but two of these studies (Ferrer et al., 2011; Smerecnik & Ruiter, 2010) were salience studies. There is an urgent need for additional studies that target changing anticipated affect, for studies that target change and salience of anticipated affect beyond the health domain, and for studies that examine other forms of anticipated affect beyond regret (e.g., guilt).

34.3.2.8 Typical Materials Needed Typical means to change anticipated affect involve developing, and testing the effectiveness of, messages that highlight the likely affective

consequences of engaging in, or failing to do, a particular behavior or follow a course of action. For example, Dillard et al. (2010) used a persuasive narrative to change anticipated affect (see Appendix 34.4). Their narrative manipulation described one individual’s experience with the decision to attend for colorectal cancer screening. There were four segments of the narrative. In the first segment, the character discussed feeling uncertain about screening and having little knowledge about the screening tests and included a photograph matched in gender, age, and race characteristics to the individuals who need to change. In the next two segments of the narrative, the character discussed various barriers to screening. In one of these segments the character described their most important barrier (tailored to the individuals who need to change baseline reports). For example, if an individual who needs to change reported during baseline that their most important barrier to screening would be embarrassment during a test, the character in the narrative said that embarrassment was the most important barrier for them. The final segment of the narrative focused on reducing the impact of bias. Based on the work of Wilson and Gilbert (2005), this segment attempted to tackle focalism (i.e., a bias whereby an individual underestimates how much other events will influence thoughts and feelings at the time of a future or anticipated event) and adaptation or immune neglect (i.e., a bias in which an individual underestimates their ability to make sense of an experience, particularly a negative one). For example, to address focalism, the character described other events that occurred during the screening experience (“I got to catch up on my reading”; “My daughter gave me a ride to the appointment”). To address adaptation neglect, the character described adapting to the screening result (“Whatever happens, I’ll deal with it just like I deal with everything else”). In studies manipulating salience of anticipated affect, it is simply the presence of questions

Affect-Based Interventions

503

Sidebar 34.3 Gamification and exergames

One of the most dynamic and evolving forms of changing behavior is through the use of gamification. Gamification is the implementation of the most common and enjoyable mechanics of video games, applied to a non–video game context (Deterding, Dixon, & Khaled, 2011). Thus, the central purported mediators between gamification mechanics and behavior are constructs aligned in the affect domain (e.g., affect proper and affective processing). Video games are a multibillion-dollar industry (Riley, 2010) and optimizing game mechanics to maximize the skill versus challenge experiences among players is considered essential to superior game design (Madigan, 2016). Leaderboards, points and levels, badges, challenges and quests, strong creative narratives, as well as social engagement, are among the most commonly implemented mechanics of gamification (Baranowski et al., 2008; Miller, Cafazzo, & Seto, 2016). While the use of gamification to modify behavior is vast, covering domains of education, business, and health, a brief example of this approach is highlighted by how gamification may modify physical activity. “Exergames” are games where players engage physically (using leg, arm, or whole-body movement) in response to some on-screen virtual activity. Household examples of this technology include games played on the Nintendo Wii™, Microsoft Kinect™, cycle ergometers, or virtual reality headsets. Evidence suggests that these games can significantly increase energy expenditure, although this is highly dependent on the type of game and console used (Kaushal & Rhodes, 2014; LeBlanc et al., 2013). Virtual boxing, for example, can elicit a response concordant with a dose of physical activity at public health guidelines, while virtual bowling is a lower-intensity activity (Barnett, Cerin, & Baranowski, 2011). The intended effect of exergames on affect-related mediators has also been supported. For example, exergames are reliably established as more enjoyable than traditional forms of physical activity among young people (Rhodes, Kaos et al., 2018; Vernadakis et al., 2014), although the effect does appear to wane over time (Rhodes, Beauchamp et al., 2019). Second-generation exergames research has shown that continual releases of new games to play over time and gamification such as leveling/ points, accessories to earn to customize one’s avatar, and social interaction may be important to sustain play (Kaos et al., 2018; Kaos et al., 2019). Still, formal tests of mediation between exergames, affect-related constructs, and behavior are scarce and results are inconclusive (Rhodes, Beauchamp et al., 2019; Rhodes, Kaos et al., 2018). Research on exergames, and the role of gamification in behavior change more generally, is exciting, with considerable untapped potential, but it also has an immense challenge. The main limitation in this line of research is the slow process of research when paired with the speed of change in the gaming and technology marketplace. Most research knowledge of gamification is on platforms and games that are now gone from the marketplace. The research evidence base is constantly trailing behind the overwhelming speed of technology change. Still, gamification and its role in modifying affect for behavior change is clearly an important area that will continue to grow and be refined over the next decade.

504

MARK CONNER, DAVID M. WILLIAMS, AND RYAN E. RHODES

tapping anticipated affect that is varied. For example, Sandberg and Conner (2009) manipulated the presence/absence of questions tapping anticipated regret (“If I did not attend for a cervical smear [pap test] in the next few weeks, I would feel regret,” with responses provided on scales with definitely no and definitely yes as scale endpoints; “If I did not attend for a cervical smear [pap test] in the next few weeks, I would later wish I had,” with responses provided on scale with strongly disagree and strongly agree as endpoints).

34.3.2.9 Typical Examples of Implementation In relation to interventions that manipulate the salience of anticipated affect, Sandberg and Conner (2009) looked at women invited for cervical screening by post. Women were sent one of two questionnaires by mail that either did or did not include anticipated regret questions. Among those completing and returning the mailed questionnaires, objectively recorded attendance for screening was considerably higher among those who received and responded to anticipated regret questions (65 percent) compared to no anticipated regret questions (44 percent). Further analyses showed this was mainly due to changes in attendance among those with a strong intention to attend for screening. In relation to manipulating the degree of anticipated affect, Carfora et al. (2017) looked at reducing processed meat consumption in students. Individuals who might want to change were randomly allocated to receive no message or a daily text message for seven days. The text messages focused on anticipated regret (e.g., “Think about the regret that you could experience this week if you exceed the recommended portions of processed meat”). Self-reported processed meat consumption was significantly reduced in the message compared to the no-message condition. Mediation analyses indicated partial serial mediation via changes in anticipated regret and intentions, that is, the intervention was shown to change both anticipated

regret and intentions and these changes partly explained the effect of the intervention on behavior.

34.4 Future Directions in Changing Behavior through Affect While direct intervention on the affect construct has received the most research attention, two other routes possible routes can be identified: direct modification of one influence (affective or otherwise) in order to overcompensate for another affective influence or intervention targeting change in the moderators of the affectbehavior link (Rhodes, Williams, & Conner, 2018). For example, implementation intentions (simple if-then plans; see Appendix 34.1 [supplemental materials] and Chapter 6, this volume) could be used to focus attention on cognitive influences (and detract from affective influences) to change affect or to change the impact of affect on behavior (Sheeran et al., 2018; e.g., “As soon as I feel anxious about attending my medical appointment, then I ignore that feeling and tell myself it is perfectly understandable”). A focus on anticipated affect in order to lessen the effects of affective attitude on a given behavior could be an effective means of intervention when the affective experience of the behavior is less amenable to change. Future research could fruitfully use multiple mediation tests to explore the effect of affective interventions that change behavior on different affect mediators such as affective attitudes and anticipated affect. A further area for future research could be testing whether a focus on an alternative cognitive or affective influence might be more effective in reducing an existing affective influence. The most important future direction for affect science and behavior, however, may be through interventions targeting change in the moderators of the affect-behavior link (Sheeran et al., 2018). This route to intervention highlights affect regulation, based on the assumption that, while affect

Affect-Based Interventions

constructs may be difficult to change (i.e., due to an evolutionary or primal foundation), it may be possible to mitigate the impact of affect on behavior. For example, habit (e.g., McCarthy et al., 2018; Rhodes & Gray, 2018), identity (Rhodes & Gray, 2018), mindfulness (Reese et al., 2018), and pharmaceuticals (Reese et al., 2018) have all been discussed as possible ways to alter the impact of affect-related constructs on health behavior. Implementation intentions have been viewed as potentially effective in this regard (see Appendix 34.1 [supplemental materials] for case studies using implementation intentions to target affect regulation). For example, studies showing implementation intentions can be used both to moderate the impact of affect on behavior and to change affect (Sheeran et al., 2018). Webb et al. (2012) provide a review of the effectiveness of different emotion regulation strategies, including distraction, reappraisals, suppression, and concentration. For example, distraction, reappraisal, and suppression were shown to be an effective means to regulate emotions, while concentration was not. Future research should seek to explore differences between increasing and decreasing the impact of affect on behavior, the extent to which there are differences when considering behaviors that need to be promoted against those that need to be negated, and the value of interventions designed to both change affect and change the impact of affect on behavior simultaneously.

34.5 Conclusions This chapter focused on the use of affect-based interventions to change behavior. Affect was defined and relevant theory and mechanisms of change presented. Step-by-step guides on how techniques targeting two affect-related determinants of action, affective attitudes and anticipated affect, can be used to promote behavior change were presented. In total, three routes by which behavioral interventions targeting affect can

505

impact behavior change were identified: direct modification of the affect constructs; direct modification of other sources of behavioral influence to overcompensate for the impact of affect; and intervention on moderators of the affect-behavior link. Targeting change in affect-related determinants to change behavior is a relatively new area of research. Initial findings are promising, although there is considerable scope for additional research. For example, the combined effects of targeting changes in affective attitudes and anticipated affect have been little studied. In addition, little research has explored the impact of different intensities of intervention or the longterm effects on behavior change.

References Abraham, C., & Sheeran, P. (2004). Deciding to exercise: The role of anticipated regret. British Journal of Health Psychology, 9, 269–278. https:// doi.org/10.1348/135910704773891096 Baranowski, T., Buday, R., Thompson, D. I., & Baranowski, J. (2008). Playing for real: Video games and stories for health-related behavior change. American Journal of Preventive Medicine, 34, 74–82. https://doi.org/10.1016/ j.amepre.2007.09.027 Barnett, A., Cerin, E., & Baranowski, T. (2011). Active video games for youth: A systematic review. Journal of Physical Activity and Health, 8, 724– 737. https://doi.org/10.1123/jpah.8.5.724 Cabanac, M. (1992). Pleasure: The common currency. Journal of Theoretical Biology, 155, 173–200. Carfora, V., Caso, D., & Conner, M. (2016). Randomized controlled trial of “messaging intervention” to increase fruit and vegetable intake in adolescents: Affective versus instrumental messages. British Journal of Health Psychology, 21, 937–955. https://doi.org/10.1111/bjhp.12208 Carfora, V., Caso, D., & Conner, M. (2017). Randomised controlled trial of a text messaging intervention for reducing processed meat consumption: The mediating roles of anticipated regret and intention. Appetite, 117, 152–160. https://doi.org/10.1016/j.appet.2017.06.025

506

MARK CONNER, DAVID M. WILLIAMS, AND RYAN E. RHODES

Conner, M. (2018). Affective attitudes, anticipated affect and health behaviors. In D. M. Williams, R. E. Rhodes, & M. T. Conner (Eds.), Affective Determinants of Health-Related Behavior (pp. 48–68). New York: Oxford University Press. Conner, M., McEachan, R., Taylor, N., O’Hara, J., & Lawton, J. (2015). Role of affective attitudes and anticipated affective reactions in predicting health behaviors. Health Psychology, 34, 642–652. https://doi.org/10.1037/hea0000143 Conner, M. & Norman, P. (Eds.). (2015). Predicting and Changing Health Behaviour: Research and Practice with Social Cognition Models (3rd ed.). Maidenhead: Open University Press. Conner, M., Rhodes, R., Morris, B., McEachan, R., & Lawton, R. (2011). Changing exercise through targeting affective or cognitive attitudes. Psychology and Health, 26, 133–149. https://doi .org/10.1080/08870446.2011.531570 Davidson, R. J., Scherer, K. R., & Goldsmith, H. H. (Eds.). (2009). Handbook of Affective Sciences. New York: Oxford University. Day, A. K., & Coups, E. J. (2018). Affect and tanning behaviors. In D. M. Williams, R. E. Rhodes, & M. T. Conner (Eds.), Affective Determinants of HealthRelated Behavior (pp. 357–376). New York: Oxford University Press. De Houwer, J., Thomas, S., & Baeyens, F. (2001). Associative learning of likes and dislikes: A review of 25 years of research on human evaluative conditioning. Psychological Bulletin, 127, 853–869. https://doi.org/10.1037/00332909.127.6.853 Deterding, S., Dixon, D., & Khaled, R. (2011). From game design mechanics to gamefulness: Defining gamification. Paper presented at the 15th International Academic MindTrek Conference: Envisioning Future Media Environments, Tampere, Finland, September 28–30. Dillard, A. J., Fagerlin, A., Dal Cin, S., ZikmundFisher, B. J., & Ubel, P. A. (2010). Narratives that address affective forecasting errors reduce perceived barriers to colorectal cancer screening. Social Science and Medicine, 71, 45–52. https:// doi.org/10.1016/j.socscimed.2010.02.038 Ellis, E. M., Elwyn, G., Nelson, W. L., Scalia, P., Kobrin, S. C., & Ferrer, R. A. (2018).

Interventions to engage affective forecasting in health-related decision making: A meta-analysis. Annals of Behavioral Medicine, 52, 157–174. https://doi.org/10.1093/abm/kax024 Ferrer, R. A., Fisher, J. D., Buck, R., & Amico, K. R. (2011). Pilot test of an emotional education intervention component for sexual risk reduction. Health Psychology, 30, 656–660. https://doi.org/ 10.1037/a0023438 Fisher, A., Bonner, C., Biankin, A. V., & Juraskova, I. (2012). Factors influencing intention to undergo whole genome screening in future healthcare: A single-blind parallel-group randomised trial. Preventive Medicine, 55, 514–520. https://doi.org/ 10.1016/j.ypmed.2012.08.008 Frijda, N. (2008). The psychologists point of view. In M. Lewis, J. M. Haviland-Jones, & L. F. Barrett (Eds.), Handbook of Emotions (3rd ed.) (pp. 68– 87). New York: Guilford Press. Godin, G., Sheeran, P., Conner, M. et al. (2010). Which survey questions change behavior? Randomized controlled trial of mere measurement interventions. Health Psychology, 29, 636–644. https://doi.org/10.1037/a0021131 Hofmann, W., De Houwer, J., Perugini, M., Baeyens, F., & Crombez, G. (2010). Evaluative conditioning in humans: A meta-analysis. Psychological Bulletin, 136, 390–421. https://doi .org/10.1037/a0018916 Hollands, G. J., Prestwich, A., & Marteau, T. M. (2011). Using aversive images to enhance healthy food choices and implicit attitudes: An experimental test of evaluative conditioning. Health Psychology, 30, 195–203. https://doi.org/ 10.1037/a0022261 Johnston, V. S. (2003). The origin and function of pleasure. Cognition and Emotion, 17, 167–179. https://doi.org/10.1080/02699930302290 Kaos, M. D., Beauchamp, M. R., Bursick, S. K. et al. (2018). Efficacy of online multi-player versus single-player exergames on adherence behaviours among children: A nonrandomized control trial. Annals of Behavioral Medicine, 20, 1–12. Kaos, M. D., Rhodes, R. E., Hämäläinen, P., & Graham, T. C. N. (2019). Social play in an exergame: How the need to belong predicts adherence. In Proceedings of the 2019 CHI Conference on

Affect-Based Interventions Human Factors in Computing Systems (CHI ’19). ACM, New York, Paper No. 430. https://doi.org/ 10.1145/3290605.3300660 Kaushal, N., & Rhodes, R. E. (2014). The home physical environment and its impact on physical activity and sedentary behaviour: A systematic review. Preventive Medicine, 67, 221–237. https:// doi.org/10.1016/j.ypmed.2014.07.026 Kiviniemi, M. T., & Klasko-Foster, L. B. (2018). The behavioral affective associations model. In D. M. Williams, R. E. Rhodes, & M. T. Conner (Eds.), Affective Determinants of Health-Related Behavior (pp. 185–203). New York: Oxford University Press. Kok, G., Gottlieb, N. H., Peters, G.-J. Y. et al. (2016). A taxonomy of behavior change methods: An intervention mapping approach. Health Psychology Review, 10, 297–312. https://doi.org/ 10.1080/17437199.2015.1077155 Larsen, R. J. (2000). Toward a science of mood regulation. Psychological Inquiry, 11, 129–141. https://doi.org/10.1207/S15327965PLI1103_01 LeBlanc, A. G., Chaput, J. P., McFarlane, A. et al. (2013). Active video games and health indicators in children and youth: A systematic review. PLoS ONE, 8, e65351. https://doi.org/10.1371/journal .pone.0065351 Lewis, M., Haviland-Jones, J. M., & Barrett, L. F. (Eds.). (2008). Handbook of Emotions (3rd ed.). New York: Guilford Press. Madigan, J. (2016). Getting Gamers: The Psychology of Videogames and Their Impact on the People Who Play Them. New York: Rowman & Littlefield. McCarthy, D. E., Cook, J. W., Leyro, T. M., Minami, H., & Bold, K. W. (2018). Affective determinants of smoking. In D. M. Williams, R. E. Rhodes, & M. T. Conner (Eds.), Affective Determinants of Health-Related Behavior (pp. 286–312). New York: Oxford University Press. Michie, S., Richardson, M., Johnston, M. et al. (2013). The behavior change technique taxonomy (v1) of 93 hierarchically clustered techniques: Building an international consensus for the reporting of behavior change interventions. Annals of Behavioral Medicine, 46, 81–95. https://doi.org/ 10.1007/s12160-013-9486-6

507

Miller, A. S., Cafazzo, J. A., & Seto, E. (2016). A game plan: Gamification design principles in mHealth applications for chronic disease management. Health Informatics Journal, 22, 184–193. https:// doi.org/10.1177/1460458214537511 Morris, B., Lawton, R., McEachan, R., Hurling, R., & Conner, M. (2015). Changing self-reported physical activity using different types of affectively and cognitively framed health messages, in a student population. Psychology, Health and Medicine, 21, 198–207. https://doi .org/10.1080/13548506.2014.997762 Morris, W. N. (1999). The mood system. In D. Kahneman, E. Diener, & N. Schwarz (Eds.), WellBeing: The Foundations of Hedonic Psychology (pp. 169–189). New York: Russell-Sage. Murgraff, V., McDermott, M. R., White, D., & Phillips, K. (1999). Regret is what you get: The effects of manipulating anticipated affect and time perspective on risky single-occasion drinking. Alcohol and Alcoholism, 34, 590–600. https://doi .org/10.1093/alcalc/34.4.590 O’Carroll, R. E., Shepherd, L., Hayes, P. C., & Ferguson, E. (2016). Anticipated regret and organ donor registration: A randomized controlled trial. Health Psychology, 35, 1169–1177. https://doi .org/10.1037/hea0000363 Parrott, M. W., Tennant, L. K., Olejnik, S., & Poudevigne, M. S. (2008). Theory of planned behaviour: Implications for an email-based physical activity intervention. Psychology of Sport and Exercise, 9, 511–526. https://doi.org/10.1016/ j.psychsport.2007.07.002 Reese, E. D., Yi, J. Y., Bell, R. P., & Daughters, S. B. (2018). The role of negative affect in the course of substance use disorders. In D. M. Williams, R. E. Rhodes, & M. T. Conner (Eds.), Affective Determinants of HealthRelated Behavior (pp. 313–333). New York: Oxford University Press. Rhodes, R. E., Beauchamp, M., Blanchard, C. M., Bredin, S. S. D., Warburton, D. E. R., & Maddison, R. (2019). Predictors of stationary cycling exergame use among inactive children in the family home. Psychology of Sport and Exercise, 41, 181–190. https://doi.org/10.1016/j .psychsport.2018.03.009

508

MARK CONNER, DAVID M. WILLIAMS, AND RYAN E. RHODES

Rhodes, R. E., Fiala, B., & Conner, M. (2009). Affective judgments and physical activity: A review and meta-analysis. Annals of Behavioral Medicine, 38, 180–204. https://doi.org/10.1007/ s12160-009-9147-y Rhodes, R. E., & Gray, S. M. (2018). Affect in the process of action control of health-protective behaviors. In D. M. Williams, R. E. Rhodes, & M. T. Conner (Eds.), Affective Determinants of Health-Related Behavior (pp. 21–47). New York: Oxford University Press. Rhodes, R. E., Gray, S. M., & Husband, C. (2019). Experimental manipulation of affective judgements about physical activity: A systematic review and meta-analysis of adults. Health Psychology Review, 13, 18–34. https://doi.org/ 10.1080/17437199.2018.1530067 Rhodes, R. E., Kaos, M. D., Beauchamp, M. R. et al. (2018). Effects of home-based exergaming on child social cognition and subsequent prediction of behavior. Scandinavian Journal of Medicine and Science in Sports, 28, 2234–2242. Rhodes, R. E., Williams, D. M., & Conner, M. (2018). Affective determinants of health behaviors: Common themes, future direction, and implications for health behavior change. In D. M. Williams, R. E. Rhodes, & M. T. Conner (Eds.), Affective Determinants of Health-Related Behavior (pp. 485–498). New York: Oxford University Press. Riley, D. M. (2010). 2009 U.S. Video Game Industry and PC Game Software Retail Sales Reach $20.2 Billion. Port Washington, NY: NPD Group. Russell, J. A. (1980). A circumplex model of affect. Journal of Personality and Social Psychology, 39, 1161–1178. https://doi.org/10.1037/h0077714 Russell, J. A., & Barrett, L. F. (1999). Core affect, prototypical emotional episodes, and other things called emotion: Dissecting the elephant. Journal of Personality and Social Psychology, 76, 805– 819. https://doi.org/10.1037/0022-3514.76.5.805 Sandberg, T., & Conner, M. (2009). A mere measurement effect for anticipated regret: Impacts on cervical screening attendance. British Journal of Social Psychology, 48, 221–236. https://doi.org/ 10.1348/014466608X347001

Sandberg, T., & Conner, M. (2011). Using selfgenerated validity to promote exercise behaviour. British Journal of Social Psychology, 50, 769– 783. https://doi.org/10.1111/j.20448309.2010.02004.x Sheeran, P., Harris, P., & Epton, T. (2014). Does heightening risk appraisals change people’s intentions and behavior? A meta-analytic review of the experimental evidence. Psychological Bulletin, 140, 511–543. https://doi.org/10.1037/ a0033065 Sheeran, P., Maki, A., Montanaro, E. et al. (2016). The impact of changing attitudes, norms, and selfefficacy on health-related intentions and behavior: A meta-analysis. Health Psychology, 35, 1178– 1188. https://doi.org/10.1037/hea0000387 Sheeran, P., Webb, T. L., Gollwitzer, P. M., & Oettingen, G. (2018). Self-regulation of affect-health behavior relations. In D. M. Williams, R. E. Rhodes, & M. T. Conner (Eds.), Affective Determinants of Health-Related Behavior (pp. 90–114). New York: Oxford University Press. Sirriyeh, R., Lawton, R., & Ward, J. (2010). Physical activity and adolescents: An exploratory randomized controlled trial investigating the influence of affective and instrumental text messages. British Journal of Health Psychology, 15, 825–840. https://doi.org/10.1348/ 135910710X486889 Smerecnik, C. M., & Ruiter, R. A. (2010). Fear appeals in HIV prevention: The role of anticipated regret. Psychology, Health and Medicine, 15, 550–559. https://doi.org/10.1080/13548506.2010.498888 Thayer, R. E. (1978). Factor analytic and reliability studies on the Activation-Deactivation Adjective Check List. Psychological Reports, 42, 747–756. https://doi.org/10.2466/pr0.1978.42.3.747 Vernadakis, N., Zetou, E., Derri, V., Bebetsos, E., & Filippou, F. (2014). The differences between less fit and overweight children on enjoyment of exergames, other physical activity and sedentary behaviours. Procedia – Social and Behavioral Sciences, 152, 802–807. https://doi.org/10.1016/ j.sbspro.2014.09.324 Watson, D., & Tellegen, A. (1985). Toward a consensual structure of mood. Psychological

Affect-Based Interventions

Bulletin, 98, 219–235. https://doi.org/10.1037/ 0033-2909.98.2.219 Webb, T. L., Miles, E., & Sheeran, P. (2012). Dealing with feeling: A meta-analysis of the effectiveness of strategies derived from the process model of emotion regulation. Psychological Bulletin, 138, 775–808. https://doi.org/10.1037/a0027600 Weirs, R. M., Anderson, K. G., Van Bockstaele, B., Salemink, E., & Hommel, B. (2018). Affect, dual-processing, developmental psychopathology, and health behaviors. In D. M. Williams, R. E. Rhodes, & M. T. Conner (Eds.), Affective Determinants of HealthRelated Behavior (pp. 158–184). New York: Oxford University Press. White, B. X., & Albarracín, D. (2017). Investigating belief falsehood. Fear appeals do change behaviour in experimental laboratory studies. A

509

commentary on Kok et al. (2018). Health Psychology Review, 12, 147–150. https://doi.org/ 10.1080/17437199.2018.1448292 Williams, D. M., & Evans, D. R. (2014). Current emotion research in health behavior science. Emotion Review, 6, 277–287. https://doi.org/ 10.1177/175407391452305 Williams, D. M., Rhodes, R. E., & Conner, M. (2018). Overview of affective determinants of health behavior. In D. M. Williams, R. E. Rhodes, & M. T. Conner (Eds.), Affective Determinants of Health-Related Behavior (pp. 1–18). New York: Oxford University Press. Wilson, T. D., & Gilbert, D. T. (2005). Affective forecasting: Knowing what to want. Current Directions in Psychological Science, 14, 131–134. https://doi.org/10.1111/j.09637214.2005.00355.x

35 Autonomy-Supportive Interventions Johnmarshall Reeve and Sung Hyeon Cheon

35.1 Introduction Catalyzing enduring behavior change is difficult. Consider trying to get people to “take the stairs” instead of “take the elevator.” One research team posted “calories burned” signs on each step of a seven-story building and then observed community stair-taking rise (nonsignificantly) from 23.2 percent to 23.8 percent (Crozier, 2019). Feeling disappointed by the meager behavior change, the researchers then added a sign between the elevator and stairs saying, “Join staff and students taking the stairs today! We’re doing it! Are you?” Following this social message, stair-taking “climbed” to 24.4 percent. This was a statistically significant behavior change, but there was clearly room for improvement. The lesson learned from this study is that the effort to catalyze enduring behavior change needs to go beyond just providing information and appealing to social norms.

35.2 Background Behavior change requires planning, new skills, time to experiment with new ways of behaving, and a lot of support (see Chapter 21, this volume). In practice, this often requires a longitudinal intervention comprised of a step-by-step plan of action that applies tried-and-true, evidence-based strategies. When rooted in self-determination theory (see Chapter 8, this volume), the intervention is typically provided to supervisors working in the helping professions. This means working closely with classroom teachers, workplace managers, athletic coaches, health care professionals (e.g., general practitioners, dentists), and others to help them upgrade the quality of

their interpersonal motivating style toward those they supervise. When supervisors learn how to do this, they become increasingly able to help others engage in more adaptive behavior (e.g., learning, prosocial behavior), positive functioning (e.g., skill development, achievement), and well-being and in less maladaptive behavior (e.g., antisocial behavior), deteriorated functioning (e.g., disengagement, problematic relationships), and ill-being. The purpose of this chapter is to provide practitioners with a guide in how to create the conditions under which other people will volitionally change their behavior based on the principles of self-determination theory (see Chapter 8, this volume). To do this, the chapter (1) defines the key constructs and practices featured in the autonomy-supportive intervention program or ASIP (e.g., supervisors’ motivating styles, supervisees’ psychological needs); (2) provides the theoretical basis of the intervention; (3) identifies the specific mechanism that explains why the intervention enables behavior change (i.e., psychological need status); (4) provides an overview of what occurs during an ASIP; (5) outlines the evidence base supporting the efficacy and benefits of the intervention; and (6) offers step-by-step guidelines for how to carry out an ASIP.

35.3 Definitions Two constructs are central to understanding an ASIP: the supervisor’s motivating style and the supervisees’ psychological need status.

https://doi.org/10.1017/9781108677318.035

Autonomy-Supportive Interventions

35.3.1 Motivating Style Motivating style refers to the interpersonal tone (Reeve, 2009, 2016), orientation toward students (Deci et al., 1981), or basic attitude (Aelterman et al., 2019) that social agents (i.e., supervisors) rely on when they try to engage the individuals with whom they work (e.g., supporting students’ learning in the classroom; increasing workers’ productivity). Such a style tends to be fairly consistent across contexts (e.g., from one class and from one semester to the next; Brekelmans, 1989; Cheon et al., 2014) and, as illustrated in Figure 35.1, it ranges from a style that is strongly prescriptive over and insistent about what others (e.g., students, employees, exercisers, patients) should think, feel, and do (a controlling motivating style) through a neutral, indifferent, or mixed style to one that is highly respectful of others’ perspectives and supportive of their initiatives (an autonomy-supportive motivating style). Autonomy support features supervision (e.g., teaching, managing, coaching, parenting) through an interpersonal tone of understanding,

511

acceptance, and empathy that provides the other person with a constant flow of opportunities for volitional action, initiative, and choice (Assor, Kaplan, & Roth, 2002). An autonomy-supportive motivating style is an enduring and consistent relationship characteristic in which one person adopts the other’s perspective, is highly respectful of the other’s initiatives, and welcomes, encourages, and invites the other’s thoughts, feelings, initiatives, and suggestions into the flow of an activity (Deci et al., 1981; Reeve, 2009, 2016). It is referred to as a motivating style because it involves a cluster of positively interrelated and co-occurring behaviors (see Figure 35.1), including taking the other’s perspective, vitalizing their psychological needs, relying on invitational and informational language, providing explanatory rationales for requests, acknowledging and accepting expressions of negative affect, and displaying patience. Interpersonal control features supervision through an interpersonal tone of coercion that pressures the other person to think, feel, and behave in a prescribed way (Reeve, 2016). A

Motivating Style

Autonomy Supportive

Neutral, Indifferent

Take the Individual’s Perspective

Instructional Behaviors to Promote Intrinsic Motivation

• Vitalize Psychological Needs During Instruction

Instructional Behaviors to Promote Internalization

• • • •

Controlling

Behavioral Control

Directly Controlling Instructional Behaviors

Psychological Control

Conditional Regard (Positive, Negative)

Provide Explanatory Rationales Rely on Invitational Language Accept Negative Affect Display Patience

Figure 35.1 Dimensions of motivating style and the specific instructional behaviors within the autonomy-supportive and controlling styles

512

JOHNMARSHALL REEVE AND SUNG HYEON CHEON

controlling motivating style is the enduring and consistent relationship characteristic in which one person tells another what to think, feel, or do and then applies an increasing amount of pressure until the other actually complies with the imposed prescription or proscription (Assor et al., 2005; Reeve, 2009). It is also referred to as a motivating style because, like the autonomysupportive style, controlling acts of supervision tend to be positively interrelated and co-occurring behaviors (see Figure 35.1), including directly controlling behaviors such as yelling or commanding (Assor et al., 2005) and psychologically controlling behaviors such as shaming and conditional regard (Soenens et al., 2012).

35.3.2 Psychological Need Status A psychological need is an inherent and everready motivational state that energizes the individual’s proactive engagement, interest-taking, challenge-seeking, information assimilation (i.e., learning), psychological growth, and wellbeing (Ryan & Deci, 2017; see Chapter 8, this volume). All individuals – regardless of age, gender, nationality, or ability – possess the three basic psychological needs of autonomy, competence, and relatedness (Ryan & Deci, 2017). Autonomy is the desire to experience selfdirection, volition, and personal endorsement of one’s behavior. It is the need to be the origin of one’s behavior and it is experienced as an inner endorsement of one’s self-initiated goals and actions. Competence is the desire to experience effectance and mastery in one’s interactions with the environment. It is the need to seek out optimal challenges, take them on, and exert persistent effort and strategic thinking until mastering them to catalyze personal growth. Relatedness is the desire to experience a close, warm connection in one’s interpersonal relationships. It is the need to be involved in warm, caring relationships characterized by mutual concern, liking, acceptance, and the establishment and enrichments of close

emotional bonds and attachments (Ryan & Deci, 2017). Need status exists in one of three states – satisfaction, dissatisfaction, or frustration – and it is highly responsive to supporting, indifferent, and thwarting environmental conditions and interpersonal relationships (Cheon, Reeve, Lee et al., 2019). Given autonomy support, individuals tend to experience need satisfaction; given indifference, they tend to experience need dissatisfaction; and given interpersonal control, they tend to experience need frustration. Need satisfaction is an uplifting and energy-mobilizing experience that tends individuals toward adaptive functioning (e.g., engagement, learning, skill development, personal growth). Need dissatisfaction is an energy-depleting experience that tends otherwise proactive individuals toward a passive and diminished type of maladaptive functioning (e.g., amotivation, boredom, disengagement). Need frustration is an energy-disorganizing experience that tends individuals toward a type of reactive and defiant functioning (e.g., anger, disruptive behavior, antisocial behavior).

35.4 Theory and Mechanisms of Change Self-determination theory serves as the theoretical basis of an ASIP (see Chapter 8, this volume). More specifically, however, the theoretical basis of an ASIP is the dual-process model within the larger self-determination theory framework.

35.4.1 Theoretical Basis in SelfDetermination Theory Self-determination theory proposes a motivational mediation model (social context ➔ motivation change ➔ behavior change) that explains individuals’ positive functioning well. For instance, in educational contexts, teacher-provided autonomy support enables students’ need satisfaction, which then energizes classroom engagement (Cheon,

Autonomy-Supportive Interventions

Reeve, & Moon, 2012; Jang, Kim, & Reeve, 2016). As researchers turned their attention to understanding maladaptive functioning, however, they found that the primary reason individuals became amotivated, angry, disengaged, and antisocial was not so much because they experienced low need satisfaction from low autonomy support but, instead, because they experienced either high need dissatisfaction from interpersonal indifference or high need frustration from interpersonal control (Cheon, Reeve, Lee et al., 2019; De Meyer et al., 2014; Jang et al., 2016). To explain adaptive functioning, diminished maladaptive functioning, and defiant maladaptive functioning, self-determination theorists now highlight three somewhat parallel explanatory processes by which the interpersonal motivating style of supervisors (e.g., teachers, managers, leaders, mentors) affects individuals’ (e.g., students, employees, athletes) motivation and outcomes, as shown in Figure 35.2. Specifically, interpersonal autonomy support vitalizes the “brighter” side of individuals’ motivation and functioning; interpersonal indifference fosters the “diminished” side of individuals’ motivation

513

and functioning; while interpersonal control galvanizes the “darker” side of individuals’ motivation and functioning (Bartholomew, Ntoumanis, Ryan, Bosch, & Thøgersen-Ntoumani, 2011; Bartholomew, Ntoumanis, Ryan, & ThøgersenNtoumani, 2011; Bhavsar et al., 2019; Cheon, Reeve, & Song, 2016; Cheon, Reeve, Lee et al., 2019; Costa, Ntoumanis, & Bartholomew, 2015; Gunnell et al., 2013; Haerens et al., 2015; Vansteenkiste & Ryan, 2013).

35.4.2 Mechanisms of Action Psychological need status serves as the mechanism by which an autonomy-supportive motivational style relates to behavior change and adaptive outcomes (for general discussions on mechanisms of action, see Chapters 20 and 46, this volume). Specifically, as shown in Figure 35.2, need satisfaction energizes adaptive behavior and functioning; need dissatisfaction precedes diminished or passive behavior and functioning; and need frustration energizes maladaptive behavior and functioning. Recognizing the central explanatory role of individuals’

Social Agent’s Interpersonal Motivating Style

Individual’s Psychological Need Status

Individual’s Outcomes (Including Behavior Change)

AutonomySupportive Motivating Style

Need Satisfaction

Indicators of Adaptive Functioning and Well-Being

Indifferent Motivating Style

Need Dissatisfaction

Indicators of Diminished Functioning and Disinterest

Controlling Motivating Style

Need Frustration

Indicators of Maladaptive Functioning and Ill-Being

Figure 35.2 Mechanisms of action within self-determination theory’s dual-process model

514

JOHNMARSHALL REEVE AND SUNG HYEON CHEON

psychological need status, the focus of the intervention is to help supervisors learn how to create the environmental conditions and interpersonal relationships that (1) catalyze individuals’ need satisfaction and hence support their adaptive functioning and (2) diminish individuals’ need dissatisfaction and need frustration and hence their diminished and maladaptive functioning. In Sidebar 35.1, behavior change techniques from the ASIP are located within Michie et al.’s (2013) taxonomy of behavior change techniques. Critically, the ASIP does not try to change individuals’ behavior. Instead, an indirect approach is adopted by working with supervisors to change the motivation (and hence the behavior) of those they supervise (see Sidebar 35.1).

35.5 Evidence Base Most of the evidence base for the ASIP has been developed and applied in the educational context. Sixteen published studies have empirically tested the capacity of an ASIP to help teachers upgrade the quality of their classroom motivating style (Cheon

& Reeve, 2013, 2015; Cheon, Reeve, Lee, & Lee, 2015, 2018; Cheon et al., 2012; Cheon, Reeve, & Ntoumanis, 2018; Cheon, Reeve, & Song, 2016, 2019; Cheon et al., 2014; Cheon Ntoumani, 2011; Bhavsar et al., 2019; Cheon, Reeve, & Song, 2016; Cheon, Reeve, Lee et al., 2019; Cheon, Reeve, & Vansteenkiste, 2020; Hardre & Reeve, 2009; Reeve, 1998; Reeve & Cheon, 2016; Reeve et al., 2004; Reeve, Jang, & Jang, 2018). In all of these intervention studies, four dependent (outcome) measures were assessed: teachers’ autonomy support, teachers’ interpersonal control, students’ psychological need satisfaction, and students’ psychological need frustration. As reported by students and/or as scored objectively by classroom raters, each of these sixteen published studies has shown ASIPenabled increases in teachers’ autonomy support and students’ need satisfaction and ASIP-enabled decreases in teachers’ interpersonal control and students’ need frustration. Because the intervention has affected students’ psychological need states, each of these studies also reports ASIP-enabled gains in adaptive functioning and declines in maladaptive functioning, as shown in Table 35.1.

Sidebar 35.1 A behavior change technique taxonomy for teachers’ behavior change

A primary domain in which the ASIP has been applied is education. In the ASIP, teachers are trained to create classroom conditions under which students will be able to volitionally change their own behavior. The program assists teachers in developing a more autonomy-supportive motivating style, which, in the language of Michie et al.’s (2013) behavior change technique taxonomy, is a type of social support, including overall (Technique 3.1), practical (Technique 3.2), and emotional (Technique 3.3) support. The adoption and maintenance of such a motivating style represent highly skilled teaching. Accordingly, the current chapter demonstrates how to develop six autonomy-supportive instructional behaviors (Technique 6.1) and provides instruction in how to enact each (Technique 4.1). In the ASIP, teachers are offered skill-building scaffolding in the form of behavioral practice (Technique 8.1), including recommending substitutional strategies in terms of converting ineffective controlling instructional behaviors into effective autonomy-supportive instructional behaviors (Technique 8.2).

Autonomy-Supportive Interventions

515

Table 35.1 Autonomy supportive intervention program (ASIP)-enabled changes in students’ adaptive and maladaptive behavior Indicator of Adaptive Behavior

Supportive Reference

Classroom Engagement What students do to make academic progress, including working harder (behavior), smarter (cognition), enthusiastically (emotion), and proactively (agency). Conceptual Learning Understanding a concept to the extent that one can explain why it occurs and changes. Skill Development Improving domain-specific, performance-enhancing skills. Academic Achievement High grades (students); medals won (Olympic athletes). Prosocial Behavior Benefiting others: helping, encouraging, or cooperating. Positive Self-Concept Positive ability-related beliefs about oneself in a particular domain. Vitality Excess energy, enthusiasm, and sense of feeling alive in a specific domain of activity. Classroom Disengagement What students do to withdraw from opportunities to make academic progress. Amotivation Lack of an intention to engage in a learning activity. Problematic Relationships Interpersonal conflict. Antisocial Behavior Harming others: hurting verbally, physically (hitting), or relationally (intimidating). Burnout Emotional exhaustion in a specific domain of activity.

Reeve et al. (2004)

35.6 Step-by-Step Guide From one ASIP to the next, researchers have identified new ways to improve the intervention and its delivery. Here are some lessons learned.

Jang, Reeve, & Halusic (2016)

Cheon, Reeve, & Moon (2012) Cheon, Reeve, & Moon (2012); Cheon et al. (2015) Cheon, Reeve, & Ntoumanis (2018) Cheon, Reeve, & Song (2019) Cheon et al. (2014)

Cheon, Reeve, & Ntoumanis (2018)

Cheon, Reeve, & Song (2016) Cheon, Reeve, & Song (2019) Cheon, Reeve, & Ntoumanis (2018)

Cheon et al. (2014)

demonstrations, deliberate practice with one-onone tutoring, small-group teaching simulations, role-playing activities, and small-group discussions. The possibility of online delivery of ASIP is currently being explored but, thus far, this has neither been delivered nor been evaluated.

35.6.1 Typical Means of Delivery The ASIP is delivered in person – face-to-face. The program is delivered through a multimediarich lecture, professionally created videotape

35.6.2 Target Audience and Behaviors The ASIP has been applied to teachers in all levels of the education system, including

516

JOHNMARSHALL REEVE AND SUNG HYEON CHEON

elementary school teachers (Reeve, Jang, & Jang, 2018), secondary school teachers (Cheon, Reeve, & Ntoumanis, 2018), and college-level instructors/professors (Matos & Gargurevich, 2019). The intervention targets the six recommended autonomy-supportive instructional behaviors (see Appendixes 35.1 and 35.3, supplemental materials), though it also helps teachers substitute existing controlling behaviors for new autonomysupportive behaviors and integrate the six individual instructional behaviors into an overall, coherent autonomy-supportive motivating style.

35.6.3 Enabling or Inhibiting Factors ASIP participation helps practically every teacher-participant upgrade the quality of their motivating style (Reeve & Cheon, 2019). That said, given teachers’ personality differences, pre-intervention beliefs, deeply held personal values, and contextual factors that surround the teaching profession (e.g., grade level taught, culture in which one teaches), the effect size observed for each teacher-participant varies, at least to some extent. This variance relates mostly to conceptual change issues regarding teachers’ acceptance/rejection of the ASIP message (Reeve, 1998). Consider personality differences that may function as either enabling or inhibiting factors. Teachers relatively high on the autonomy causality orientation (Deci & Ryan, 1985) tend to benefit the most from the intervention experience as their pre-intervention beliefs and values allow them to assimilate the autonomy-supportive message relatively easily and conflict-free, while teachers relatively high on authoritarianism and a control causality orientation tend to benefit the least. They benefit least because their pre-intervention beliefs and values conflict with the autonomy-supportive message and therefore require conceptual accommodation – that is, real cognitive change rather than just cognitive growth (Reeve, Jang, & Jang, 2018). As teachers progress throughout the semester and continue to work actively to upgrade the quality

of their motivating style, several factors catalyze and thus further enable this transition to a more autonomy-supportive motivating style, including an ASIP-enabled increase in their teaching efficacy, the adoption of intrinsic instructional goals, and ongoing experiences of their own psychological need satisfaction while teaching in an autonomy-supportive way (Cheon et al., 2014).

35.6.4 Training or Skills Required ASIP is a sophisticated intervention. It requires extensive, theory-based, and participant-focused preparation, such as highly informative presentation slides and professionally created “how-to” videotapes. Several yardsticks are typically used to evaluate whether the intervention is helping teachers develop the skills needed to upgrade the quality of their classroom motivating style, as listed in Section 35.6.7.

35.6.5 Intensity, Quality, and Durability The ASIP is a three-part, eight-hour professional developmental experience. Teachers need this intense training because changing one’s motivating style requires cognitive conceptual change (i.e., beliefs about teaching, self-concept) and the development of new teaching skills (i.e., the recommended autonomy-supportive instructional behaviors). It further requires actual classroom experience in which teachers try out and refine their capacity to enact each autonomy-supportive instructional behavior in an authentic setting. That said, the intervention is limited to this three-part, eight-hour time commitment because teachers are extremely busy and vulnerable to work overload and because there is a need to minimize attrition from the program. The quality of the intervention matters. A meta-analysis identified several key characteristics of relatively effective ASIPs (Su & Reeve, 2011), including teaching the full range of

Autonomy-Supportive Interventions

autonomy-supportive instructional behaviors rather than only some subset of these behaviors, delivering the intervention in multiple sessions or parts rather than in a single session, including a group discussion component where teachers can express their concerns and share strategies, emphasizing skill-based training (i.e., how to do it) more than content-based knowledge (i.e., what to do), and addressing teachers’ pre-intervention beliefs about motivating style that otherwise might conflict with the message offered by the ASIP. In addition, long-term follow-up studies show impressive endurance both for teachers’ greater autonomy support and for students’ greater positive functioning. For instance, teacher-participants continue to rely on an autonomy-supportive motivating style years after the intervention experience and their new classes of students experience the same benefits shown earlier by the students taught during the intervention (Cheon & Reeve, 2013).

35.6.6 Evaluation of Fidelity Universally, participating teachers report with high ratings that they find the intervention experience to be personally useful and professionally empowering (Cheon & Reeve, 2013). Because practically every teacher finds the intervention experience to be useful, almost all teachers who participate in ASIP do become significantly more autonomy-supportive post-intervention (Reeve & Cheon, 2019). This high success rate occurs because ASIP provides teachers with what they most need and want – namely, new and effective teaching skills that enable them to motivate and engage their students during classroom instruction (Cheon, Reeve, Lee, & Lee, 2018).

35.6.7 Evaluation of Effectiveness Five ways are proposed for teachers to evaluate the extent to which they are becoming more autonomy-supportive and less controlling toward

517

students over the course of the ASIP. First, teachers can invite a trained rater or a trusted colleague to visit their classrooms to score the extent to which they engage in the various autonomy-supportive instructional behaviors. An objective rating sheet used by the rater or colleague to score the teacher’s autonomy-supportive instructional behaviors is provided in the Appendix 35.3 (supplemental materials). Second, teachers can ask their students to complete a questionnaire to report their perceptions of autonomy-supportive (e.g., Learning Climate Questionnaire; Williams & Deci, 1996) and controlling (e.g., Controlling Teacher Questionnaire; Jang et al., 2009) teaching. Third, teachers can look for large and immediate gains in students’ adaptive functioning (e.g., greater engagement). These gains are so large (effect sizes d > 1) as to be obvious classroom events. Fourth, teachers can track their own growing sense of teaching efficacy to enhance their students’ classroom engagement and learning (e.g., Teaching Efficacy Scale – short version; Tschannen-Moran & Woolfolk Hoy, 2001). Because autonomy-supportive teaching is highly effective, teaching efficacy tends to rise in proportion to which teachers upgrade the quality of their motivating style. Fifth, teachers can complete self-report motivating style questionnaires, such as the Situations in Schools Questionnaire (Aelterman et al., 2019).

35.6.8 Example: ASIP Intervention Program The ASIP was created to offer K-12 teachers a professional developmental opportunity to develop the teaching skills needed to become substantially more autonomy-supportive and less controlling during instruction. The procedural timeline of an ASIP is outlined in Figure 35.3. Each of the intervention studies aimed at evaluating the ASIP’s efficacy relies on an experimental research design and on longitudinal data collection. The experimental research design

518

JOHNMARSHALL REEVE AND SUNG HYEON CHEON

Month Before the Semester (or Academic Year) Begins

Teachers (and their Students) Randomly Assigned into an Experimental Condition

Half of the Teachers Placed into the Experimental Condition

ASIP Training Intervention Part I

ASIP Training Intervention Part 2

Half of the Teachers Placed into the Control Condition

Week 1

Week 6

Students Complete T1 Survey

ASIP Training Intervention Part 3

Students Complete T1 Survey

Week 8

Raters Score Teachers’ Insructional Behaviors

Raters Score Teacher’s Instructional Behaviors

Week 10

Week 18

Students Complete T2 Survey

Students Complete T3 Survey

Students Complete T2 Survey

Students Complete T3 Survey

Figure 35.3 Procedural timeline for an autonomy-support intervention program (ASIP)

allows the researcher to address the causal effect of the intervention on a change in teachers’ motivating styles, while the longitudinal data collection permits an evaluation of students’ behavior change over time (e.g., from the beginning to the end of a semester or academic year). Part 1 is a three-hour workshop that begins with reflective, personalized warm-up activities (e.g., teachers complete the Teaching Scenarios Questionnaire; provided in Appendix 35.2, supplemental materials). The workshop introduces autonomy-supportive teaching, contrasts it against teacher control, provides empirical evidence on the benefits of autonomy support and the costs of control, and introduces the six recommended autonomy-supportive instructional behaviors provided in Appendix 35.1 (supplemental materials). Part 2 is a skill-based three-hour workshop that begins with live and videotaped examples/models of the six autonomy-supportive instructional behaviors that then adds one-on-one coaching and practicing on the “how to” of each recommended behavior. An excerpt of the slides used during ASIP to help teachers learn two of the autonomysupportive behaviors is provided in Appendix 35.4 (supplemental materials). The workshop helps teachers, first, develop the skill they need to enact each autonomy-supportive act of instruction and, second, transform existing controlling instructional behaviors (e.g., uttering directives) into autonomy-supportive instructional behaviors

(e.g., providing explanatory rationales). The workshop includes teaching simulations and role-play activities in a small-group format to afford teachers opportunities to experiment with, refine, and personalize each recommended instructional behavior until they are ready to enact each on in their own classroom with their own students. In some recent interventions, add-on modules have been incorporated into Part 2 (see Sidebar 35.2). Part 3 of the ASIP is a two-hour peer-based group discussion that takes place after teachers have had the opportunity to experiment with autonomy-supportive teaching in their own classrooms. In the group discussion, teachers share their experiences, exchange tips and strategies for particular teaching situations, report on how their students reacted, discuss the obstacles they encountered to autonomy-supportive teaching, and learn from their peers some new and perhaps better ways of supporting students’ autonomy.

35.7 Supplemental Materials The supplemental materials in the Appendixes provide resources that interested teachers and practitioners might find useful in implementing an ASIP or in trying to upgrade the quality of their motivating style more generally. Appendix 35.1 provides conceptual definitions for the six autonomy-supportive instructional behaviors; Appendix 35.2 provides the Teaching Scenarios

Autonomy-Supportive Interventions

519

Sidebar 35.2 Refining the autonomy-support intervention program (ASIP)

The ASIP is undergoing constant development and refinement. “Add-on” modules have been introduced during Part 2 of the program, which help teachers learn not only autonomy-supportive teaching but also some additional aspects of expert teaching related to structured teaching. In one intervention, teachers learned how to recommend intrinsic goal pursuit to their students in an autonomy-supportive way (Cheon, Reeve, & Song, 2020). In another intervention, teachers learned how to implement a highly structured motivating style (i.e., clear expectations paired with guidance and feedback in how to meet those expectations) in an autonomysupportive way (Cheon, Reeve, & Vansteenkiste, 2019). In both interventions, teachers were able to offer their students a motivating style that was both highly autonomy-supportive and highly structured.

Questionnaire; Appendix 35.3 provides a rating sheet to score autonomy-supportive and controlling instructional behaviors; and Appendix 35.4 provides an excerpt of slides used during ASIP to help teachers learn how to “acknowledge and accept expressions of negative affect” and “provide explanatory rationales.”

35.8 Conclusion and Reflections Two core conclusions can be drawn from the two dozen ASIP-based interventions conducted over the last decade. First, supervisors – mainly teachers but also athletic coaches (Cheon et al., 2015), workplace managers (Hardre & Reeve, 2009), and parents (Weber-Gasparoni et al., 2013) – can learn how to become significantly more autonomy-supportive and significantly less controlling toward those they care for and supervise. Second, when supervisors become more autonomy-supportive and less controlling, important and long-lasting benefits accrue – both for the supervisor (e.g., greater job satisfaction) and for the supervisees (e.g., adaptive behavior change). It is also important to note that ASIP interventions can always be improved and strengthened (Reeve & Cheon, 2019). For instance, the ASIP has recently been extended to a “multiple

motivating styles” intervention in which supervisors are helped to learn how to integrate autonomy support and structure into a single, coherent motivating style (Cheon, Reeve, & Song, 2019). There is also work being conducted to expand the targeted range of adaptive and maladaptive behaviors (as in Table 35.1). For instance, an initial study helped teachers solve the classroom problems of student amotivation and disengagement (Cheon et al., 2016); a subsequent study helped teachers solve the classroom problem of antisocial behavior (Cheon, Reeve, & Ntoumanis, 2018); and recent studies are helping teachers solve the classroom problem of bullying. Something very important can be learned about the motivational dynamics of promoting adaptive behavior change from autonomy-supportive interventions. If a supervisor can learn how to boost others’ need satisfaction and to diminish others’ need dissatisfaction and frustration, then self-determination theory principles can do the rest in terms of effecting the behavior change. Intervention-enabled gains in need satisfaction are excellent catalysts for many prized adaptive behaviors, just as intervention-enabled declines in need dissatisfaction and frustration are excellent reducers of many worrisome maladaptive behaviors (Ryan & Deci, 2017).

520

JOHNMARSHALL REEVE AND SUNG HYEON CHEON

References Aelterman, N., Vansteenkiste, M., Soenens, B. et al. (2019). Toward a fine-grained understanding of the components of need-supportive and needthwarting teaching: The merits of a gradual approach. Journal of Educational Psychology, 111, 497–521. https://doi.org/10.1037/edu0000293 Assor, A., Kaplan, H., Kanat-Maymon, Y., & Roth, G. (2005). Directly controlling teacher behaviors as predictors of poor motivation and engagement in girls and boys: The role of anger and anxiety. Learning and Instruction, 15, 397–413. https:// doi.org/10.1016/j.learninstruc.2005.07.008 Assor, A., Kaplan, H., & Roth, G. (2002). Choice is good, but relevance is excellent: Autonomyenhancing and suppressing teaching behaviors predicting students’ engagement in schoolwork. British Journal of Educational Psychology, 27, 261–278. https://doi.org/10.1348/ 000709902158883 Bartholomew, K. J., Ntoumanis, N., Ryan, R., Bosch, J. A., & Thøgersen-Ntoumani, C. (2011). Selfdetermination theory and diminished functioning: The role of interpersonal control and psychological need thwarting. Personality and Social Psychology Bulletin, 37, 1459–1473. https://doi.org/10.1177/0146167211413125 Bartholomew, K. J., Ntoumanis, N., Ryan, R. M., & Thøgersen-Ntoumani, C. (2011). Psychological need thwarting in the sport context: Assessing the darker side of athletic experience. Journal of Sport and Exercise Psychology, 33, 75–102. https://doi .org/10.1123/jsep.33.1.75 Bhavsar, N., Ntoumanis, N., Quested, E. et al. (2019). Conceptualizing and testing a new tripartite measure of coach interpersonal behaviors. Psychology of Sport and Exercise, 44, 107–120. https://doi.org/10.1016/j.psychsport.2019.05.006 Brekelmans, M. (1989). Interpersonal Teacher Behavior in the Classroom. Utrecht: W. C. C. Cheon, S. H., & Reeve, J. (2013). Do the benefits from autonomy-supportive PE teacher training programs endure? A one-year follow-up investigation. Psychology of Sport and Exercise, 14, 508–518. https://doi.org/10.1016/ j.psychsport.2013.02.002

Cheon, S. H., & Reeve, J. (2015). A classroom-based intervention to help teachers decrease students’ amotivation. Contemporary Educational Psychology, 40, 99–111. https://doi.org/10.1016/ j.cedpsych.2014.06.004 Cheon, S. H., Reeve, J., Lee, J., & Lee, Y. (2015). Giving and receiving autonomy support in a highstakes sport context: A field-based experiment during the 2012 London Paralympic Games. Psychology of Sport and Exercise, 19, 59–69. https://doi.org/10.1016/j.psychsport.2015.02.007 Cheon, S. H., Reeve, J., Lee, Y., & Lee, J.-W. (2018). Why autonomy-supportive interventions work: Explaining the professional development of teachers’ motivating styles. Teaching and Teacher Education, 69, 43–51. https://doi.org/10.1016/ j.tate.2017.09.022 Cheon, S. H., Reeve, J., Lee, Y. et al. (2019). Expanding autonomy psychological need states from two (satisfaction, frustration) to three (dissatisfaction): A classroom-based intervention study. Journal of Educational Psychology, 111, 685–702. https:// doi.org/10.1037/edu0000306 Cheon, S. H., Reeve, J., & Moon, I. S. (2012). Experimentally based, longitudinally designed, teacher-focused intervention to help physical education teachers be more autonomy supportive toward their students. Journal of Sport and Exercise Psychology, 34, 365–396. https://doi.org/ 10.1123/jsep.34.3.365 Cheon, S. H., Reeve, J., & Ntoumanis, N. (2018). A needs-based intervention to help PE teachers enhance students’ prosocial behavior and diminish antisocial behavior. Psychology of Sport and Exercise, 35, 74–88. https:doi.org/10.1016/ j.psychsport.2017.11.010 Cheon, S. H., Reeve, J., & Song, Y.-G. (2016). A teacher-focused intervention to decrease PE students’ amotivation by increasing need satisfaction and decreasing need frustration. Journal of Sport and Exercise Psychology, 38, 217–235. https://doi.org/10.1123/jsep.2015-0236 Cheon, S. H., Reeve, J., & Song, Y.-G. (2019). Recommending goals and supporting needs: An intervention to help physical education teachers communicate their expectations while supporting students’ psychological needs. Psychology of

Autonomy-Supportive Interventions

Sport and Exercise, 41, 107–118. https://doi.org/ 10.1016/j.psychsport.2018.12.008 Cheon, S. H., Reeve, J., Yu, T. H., & Jang, H.-R. (2014). The teacher benefits from giving autonomy support during physical education instruction. Journal of Sport and Exercise Psychology, 36, 331–346. https://doi.org/10.1123/ jsep.2013-0231 Cheon, S. H., Reeve, J., & Vansteenkiste, M. (2020). Expanding a traditional autonomy-supportive intervention into a multiple motivating styles intervention for PE teachers: Benefits to students, benefits to teachers. Teaching and Teacher Education. http://doi.org/10.1016/j.tate.2019 .103004 Costa, A., Ntoumanis, N., & Bartholomew, K. (2015). Predicting the brighter and darker sides of interpersonal relationships: Does psychological need thwarting matter? Motivation and Emotion, 39, 11– 24. https://doi.org/10.1007/s11031-014-9427-0 Crozier, A. J. (2019). Step up: Exploring the impact of social prompts on stair use in a university setting. Psychology of Sport and Exercise, 41, 99–106. https://doi.org/10.1016/j.psychsport.2018.11.015 Deci, E. L., & Ryan, R. M. (1985). The general causality orientations scale: Self-determination in personality. Journal of Research in Personality, 19, 109–134. http://doi.org/10.1016/0092-6566 (85)90023–6 Deci, E. L., Schwartz, A., Sheinman, L., & Ryan, R. M. (1981). An instrument to assess adults’ orientations toward control versus autonomy in children: Reflections on intrinsic motivation and perceived competence. Journal of Educational Psychology, 73, 642–650. https://doi.org/10.1037/ 0022-0663.73.5.642 De Meyer, J., Tallir, I., Soenens, B. et al. (2014). Does observed controlling teaching behavior relate to students’ motivation in physical education? Journal of Educational Psychology, 106, 541– 554. https://doi.org/10.1037/a0034399 Gunnell, K., Crocker, P. R. E., Wilson, P. M., Mack, D. E., & Zumbo, B. D. (2013). Psychological need satisfaction and thwarting: A test of basic psychological needs theory in physical activity contexts. Psychology of Sport and Exercise, 14, 599–607. https://doi.org/10.1016/ j.psychsport.2013.03.007

521

Haerens, L., Aelterman, N., Vansteenkiste, M., Soenens, B., & Van Petegem, S. (2015). Do perceived autonomy-supportive and controlling teaching relate to physical education students’ motivational experiences through unique pathways? Distinguishing between the bright and dark side of motivation. Psychology of Sport and Exercise, 16, 26–36. https://doi.org/10.1016/ j.psychsport.2014.08.013 Hardre, P. L., & Reeve, J. (2009). Training corporate managers to adopt a more autonomy-supportive motivating style toward employees: An intervention study. International Journal of Training Development, 13, 165–184. https://doi .org/10.1111/j.1468-2419.2009.00325.x Jang, H., Reeve, J., Ryan, R. M., & Kim, A. (2009). Can self-determination theory explain what underlies the productive, satisfying learning experiences of collectivistically-oriented South Korean adolescents? Journal of Educational Psychology, 101, 644–661. https://doi.org/ 10.1037/a0014241 Jang, H., Kim, E.-J., & Reeve, J. (2016). Why students become more engaged or more disengaged during the semester: A selfdetermination theory dual-process model. Learning and Instruction, 43, 27–38. https://doi .org/10.1016/j.learninstruc.2016.01.002 Jang, H., Reeve, J., & Halusic, M. (2016). A new autonomy-supportive way of teaching that increases conceptual learning: Teaching in students’ preferred ways. Journal of Experimental Education, 84, 686–701. https://doi.org/10.1080/ 00220973.2015.1083522 Matos, L., & Gargurevich, R. (2019). An autonomysupportive intervention to help Peruvian college instructors support their students’ autonomy. Unpublished manuscript, Pontifical Catholic University of Peru. Michie, S., Richardson, M., Johnston, M. et al. (2013). The behavior change technique taxonomy (v1) of 93 hierarchically clustered techniques: Building an international consensus for the reporting of behavior change interventions. Annals of Behavioral Medicine, 46, 81–95. https/doi.org/ 10.1007/s12160-013-9486-6 Reeve, J. (1998). Autonomy support as an interpersonal motivating style: Is it teachable? Contemporary

522

JOHNMARSHALL REEVE AND SUNG HYEON CHEON

Educational Psychology, 23, 312–330. https://doi .org/10.1006/ceps.1997.0975 Reeve, J. (2009). Why teachers adopt a controlling motivating style toward students and how they can become more autonomy supportive. Educational Psychologist, 44, 159–178. https://doi.org/ 10.1080/00461520903028990 Reeve, J. (2016). Autonomy-supportive teaching: What it is, how to do it. In J. C. K. Wang, W. C. Liu, & R. M. Ryan (Eds.), Building Autonomous Learners: Perspectives from Research and Practice Using Self-Determination Theory (pp. 129–152). New York: Springer. Reeve, J., & Cheon, S. H. (2016). Teachers become more autonomy supportive after they believe it is easy to do. Psychology of Sport and Exercise, 22, 178–189. https://doi.org/10.1016/ j.psychsport.2015.08.001 Reeve, J., & Cheon, S. H. (2019). Sociocultural influences on teachers’ reactions to an intervention to help them become more autonomy supportive. In G. A. D. Liem & D. M. McInerney (Eds.), School Motivation Intervention: Sociocultural Influences. Greenwich, CT: Information Age Publishing. Reeve, J., Jang, H., Carrell, D., Jeon, S., & Barch, J. (2004). Enhancing high school students’ engagement by increasing their teachers’ autonomy support. Motivation and Emotion, 28, 147–169. https://doi.org/10.1023/B: MOEM.0000032312.95499.6 f Reeve, J., Jang, H.-R., & Jang, H. (2018). Personality-based antecedents of teachers’ autonomy-supportive and controlling motivating styles. Learning and Individual Differences, 62, 12–22. https://doi.org/10.1016/ j.lindif.2018.01.001

Ryan, R. M., & Deci, E. L. (2017). Self-Determination Theory: Basic Psychological Needs in Motivation, Development, and Wellness. New York: Guilford Press. Soenens, B., Sierens, E., Vansteenkiste, M., Goossens, L., & Dochy, F. (2012). Psychologically controlling teaching: Examining outcomes, antecedents, and mediators. Journal of Educational Psychology, 104, 108–120. https:// doi.org/10.1037/a0025742 Su, Y., & Reeve, J. (2011). A meta-analysis of the effectiveness of intervention programs designed to support autonomy. Educational Psychology Review, 23, 159–188. https://doi.org/10.1007/ s10648-010-9142-7 Tschannen-Moran, M., & Woolfolk Hoy, A. (2001). Teacher efficacy: Capturing an elusive construct. Teaching and Teacher Education, 17, 783–805. https://doi.org/10.1016/S0742-051X (01)00036–1 Vansteenkiste, M., & Ryan, R. M. (2013). On psychological growth and vulnerability: Basic psychological need satisfaction and need frustration as a unifying principle. Journal of Psychotherapy Integration, 23, 263–280. https/ doi.org/10.1037/a0032359 Weber-Gasparoni, K., Warren, J. J., Reeve, J., Drake, D. R., Kramer, K. W. O., & Dawson, D. V. (2013). An effective psychoeducational intervention for ECC prevention – Part 2. Pediatric Dentistry, 35, 247–251. Williams, G. C., & Deci, E. L. (1996). Internalization of biopsychosocial values by medical students: A test of self-determination theory. Journal of Personality and Social Psychology, 70, 767–779. https://doi.org/ 10.1037/0022–3514.70.4.767

36 Incentive-Based Interventions Uri Gneezy, Agne Kajackaite, and Stephan Meier

Practical Summary This chapter discusses how and why incentives affect behavior change. Incentives are often introduced into situations in which people are already motivated to change their behaviors but have trouble following through with their intentions. Our framework lays out four “channels” through which incentives can support behavior change: Incentives can create desired or break undesired habits and lead to long-term change even after they have been removed. Well-structured upfront and regular incentives can overcome the dilemma people face when benefits of an activity are too far in the future but costs are immediate, making implementation of behavior change difficult. Incentives can also help overcome setup and switching costs that serve as barriers to behavior change. The four channels and the supporting empirical evidence have implications for how incentive-based interventions work and provide guidance on how best to design them for increased efficacy.

36.1 Introduction Economics is based on the premise that incentives matter. If the cost associated with an activity or product increases, people will consume less of it. Similarly, if the benefit associated with an activity or product increases, people will consume more of it. This reaction to cost-benefit analysis is what economists call the basic law of demand. The law of demand applies not only to products that are standard tangible goods, such as milk or bread, but also to ones that are intangible, such as one’s work performance, exercise, or education. Experimental studies show people adjust their behavior in response to incentives (e.g., Angrist et al., 2002; Ashraf et al., 2006; Bachireddy et al., 2019; Charness & Gneezy, 2009; Thaler & Sunstein, 2008; Friebel et al., 2017; Gneezy et al., 2011; see also Chapter 14,

this volume). Given this behavior, an important question is how incentives can work as a tool for behavior change. Can incentives reduce undesirable behavior (e.g., smoking or drinking) and increase desirable behavior (e.g., exercising or saving for retirement)? The empirical evidence discussed in the following sections shows the effect of incentives on behavior is more complicated than predicted by the basic law of demand in economic theory. In some cases, research finds no effect of incentives on behavior; in others, the incentives backfire and reduce the desired activity (e.g., Falk & Kosfeld, 2006; Fehr & List, 2004; Frey, 1997; Gneezy & Rustichini, 2000; Mellström & Johannesson, 2008). Reconciling these findings, this chapter presents a framework for understanding when https://doi.org/10.1017/9781108677318.036

524

URI GNEEZY, AGNE KAJACKAITE, AND STEPHAN MEIER

Information Implementation

Behavior Change

Motivation

Figure 36.1 Impediments and facilitators of behavior change

and how incentives work to change behavior and, in particular, habits.

36.1.1 Definitions: What Are Incentives? Given the centrality of incentives to the key premise of the current chapter, it is important to provide an operational definition of what an incentive is. Incentives are rewards or punishments that motivate agents to take up an activity and guide the way they perform it. Incentives can be positive or negative and can be tangible or not. For example, positive incentives can take the form of tangible rewards such as money, vouchers, badges, and trophies or be intangible like feedback, praise, or affection. Similarly, negative incentives can be tangible, such as fines, or intangible, like criticism or public berating and so on. These incentives can be used to positively or negatively reinforce behavior (see Chapter 8, this volume).

36.1.2 Incentives and Behavior Change: Theory and Empirical Evidence According to economic theory, people react to changes in relative prices – the cheaper and more beneficial a product or an activity is, the more of it they consume. However, information on these costs and benefits is often limited and may even be asymmetric, with different people having different information. In the economic model, people search for information but only if the cost of acquiring it is not too high. Incentives can change the calculus of

information acquisition by increasing the value of information. Moreover, they can increase or decrease the motivation to take up a certain activity by changing the cost or benefit of the activity. However, neither lack of information nor lack of motivation can explain many of the failures to change behavior. For example, many people want to lose weight but fail to do so. This failure is not due to a lack of information or motivation: They know the benefits and know what they need to do and are motivated, yet they fail. The difficulty lies in finding a way to implement the behavior change (see Figure 36.1) (for a perspective from health psychology, see Chapters 6 and 39, this volume; see also Adams et al., 2014; Giles et al., 2014).

36.2 Four “Channels” for Incentives and Behavior Change In this chapter, four processes or channels are proposed through which incentives can help in implementing behavior change: (1) creating habits; (2) breaking habits; (3) providing regular and upfront incentives; and (4) removing barriers to change. These channels are described in detail in this section.

36.2.1 Creating Habits by Building a “Stock” of Behavior An important goal of incentives is to change behavior not only in the short term, while the incentives are used, but also in the long term, after the

Incentive-Based Interventions

incentives have been removed (see Chapter 8, this volume). Consider people who do not exercise. If someone were to pay them enough for each time they went to the gym, they would likely begin attending, perhaps even enthusiastically. However, the fact that they would attend while the incentives are in place does not necessarily mean they would form a habit, that is, continue to exercise after the incentives are removed. Can incentives help in developing habits effectively? In economics, the leading theory of habit formation is based on the Becker–Murphy model (Becker & Murphy, 1988; see also Becker, 1992). According to this model, past consumption builds up a “capital stock” of behavior, and a person’s current utility from consumption depends on this capital stock. A habit may form when the utility from consuming an activity or product (e.g., exercise, alcohol, cigarettes, or social media) depends on how much the person consumed it in the past (see Chapter 13, this volume). Gym attendance can illustrate the process of building up habitual stock. The first visits to the gym are rough – the body is not ready, visual improvements cannot be immediately seen, and muscles feel sore for days. If one keeps exercising (“consuming this good”), however, the activity becomes more enjoyable as the benefits become tangible, visible, and clear – the body feels stronger and better in daily life, weight loss becomes apparent, and muscles begin to shape. Past consumption of the gym positively affects the utility of present consumption, leading to the formation of habitual stock. Once the process begins, building the habit is easy enough – getting through the first stages is the tough part. Incentives can help with building up this stock of behavior (see Chapter 41, this volume). In a series of field experiments with university students, Charness and Gneezy (2009; see Sidebar 36.1) aimed to increase gym attendance by using monetary incentives. Their participants received informational brochures about the benefits of exercise and, depending on the treatment

525

group, received additional incentives. The first group served as a control and received no financial incentives. The second and the third groups were promised $25 contingent on attending the university’s gym at least once during the next week. After that one week, the experiment was over for the second group but the third group was promised an additional $100 contingent on attending the gym at least eight times in the four following weeks. Overall, participants met the minimum number of visits required in order to receive the incentives, confirming the notion that, if people are incentivized to go to the gym, they will attend. The study found that, although simply going to the gym once did not change long-run behavior, a four-week incentive increased gym attendance significantly. The increase in the number of gym visits did not significantly decline in the weeks following the removal of the incentives. This result shows that helping students get through the initial gym visits by giving them money for attending had an effect not only in the short term but also in the long term, aiding in what seems to be habit formation. It suggests that an incentive stretching over four weeks can have an impact on the behavior past the incentivized period.1 Royer et al. (2015) ran a similarly incentivized gym-attendance experiment with employees in a Fortune 500 company. They also found a positive effect of incentives on forming habits but only for people who did not go to the gym at all prior to the experiment. People who had gone to the gym before reverted back to their previous level of gym attendance after the incentives were removed. However, when accompanied by a self-funded commitment device – willing individuals pledged not to skip more than fourteen days 1

See also Acland and Levy (2015), who replicated the results of the experiment by Charness and Gneezy (2009) but used different instructions and emphasis. They found a stronger decline in gym attendance over time after the intervention than Charness and Gneezy (2009).

526

URI GNEEZY, AGNE KAJACKAITE, AND STEPHAN MEIER

Sidebar 36.1 Incentives to exercise

Monetary incentives can motivate people to exercise and can continue to have an effect even after the incentives are removed. Charness and Gneezy (2009) introduced a one-month monetary incentive for college students contingent on gym attendance (attending the gym at least eight times) and found the incentives led to an increase in exercising both during and in the months following the experiment (“Eight-times” group in Figure 36.2). Figure 36.2 presents the main results of the experiment.

Figure 36.2 Average gym visits with and without incentives

The horizontal axis in Figure 36.2 represents time, starting before the incentives to the participants were introduced, continuing after incentives had ceased. The dashed vertical lines in Figure 36.2 indicate the period of incentivized gym attendances, and the vertical axis is the average number of times a student visits the gym per week. Results indicated that students who were incentivized to go to the gym for a month were significantly more likely to continue going in the following weeks, after the incentive program ended. Incentivizing the students to go to the gym at least eight times was more effective than incentivizing them to go to the gym at least one time (“Onetime” group in Figure 36.2). The findings suggest that introducing monetary incentives to exercise can increase gym attendance in the long term – after the incentives are discontinued.

Incentive-Based Interventions

of exercising in a row in the following eight weeks – incentives led to habit formation for both previous non–gym-goers and gym-goers. Weight-loss programs provide another example of how incentives can be used to change habits through habitual stock. Volpp et al. (2008) conducted a field experiment using two kinds of monetary incentives to change diet and exercise habits in obese participants. The control group received no incentives, whereas the first incentives group played a lottery at the beginning of the intervention and received the winnings if they lost at least the given target weight over the following sixteen weeks. The second incentives group invested their own money at the beginning of the intervention and lost it if they failed to lose at least the target weight over the following sixteen weeks. Both interventions were successful – participants lost more weight in the two incentives groups than in the control group. Specifically, whereas around half of the participants in both incentive groups reached the target weight, only 10 percent did so in the control group. These results show that paying participants to lose weight or, alternatively, making them pay for failing to lose weight changes behavior significantly. To see whether this behavior change would persist after the incentives were removed and the habit of a new diet and exercise was successfully formed, some participants were asked to return for a seven-month follow-up weigh-in. Volpp et al. (2008) found both incentive groups did regain some weight but they regained less weight than the control group and weighed significantly less at the seven-month follow-up than at the beginning of the study. Thus, even though the effects declined over time, some habits in diet and exercise were successfully formed. This evidence exemplifies how incentives can be used to get past the challenging “first stages” of the behavior change process and successfully help form habitual “stock” for diet and exercise; but can incentives be used to aid in habit

527

formation in arguably less salient domains? For example, can they be used to help form a philanthropic habit? Meier (2007) studied this question in a field experiment with university students in Switzerland, who had to decide anonymously whether to contribute to two social funds each semester. Randomly selected students in the treatment group had their donations partially subsidized by an anonymous donor, whereas none of the students in the control group received incentives. Meier (2007) found students with subsidized donations donated more to the social funds during the experiment. However, after removal of the incentive, these treatment-group students donated even less than those who were in the control group. In other words, the incentive not only failed to create the desired habit but also backfired. A possible explanation for this result is that the incentive “crowded out” or “undermined” intrinsic motivation to give. The “crowding-out” literature suggests, after incentives are removed, levels of behavioral participation may worsen compared to when no incentives are introduced (see Chapters 8 and 35, this volume). Another example of research in creating stock of habitual behavior comes from the education context. Levitt et al. (2016) ran a large-scale natural experiment with low-performing schools in Chicago Heights. They introduced multiple performance-based monetary incentives paid based on attendance, behavior, grades, and standardized test scores. The incentives were either fixed or lottery-based, were received either by the students or by their parents, and were introduced for the entire school year. In the fixed-incentive treatment, students (or parents, depending on which fixed treatment) who met the goals of the month received $50. In the lottery-incentive treatment, 10 percent (10 of about 100 people) received a $500 prize. Levitt et al. (2016) found the positive effect of such incentives to be overall small; however, the incentives worked on students who were just below the threshold of meeting the defined goals. These students continued to

528

URI GNEEZY, AGNE KAJACKAITE, AND STEPHAN MEIER

outperform their control-group peers a year after the incentives ended, which suggests they formed an effective study habit. However, this positive effect disappeared two years after the intervention, implying the habit may fade over time. Together, these studies on health, philanthropy, and education suggest habits can be developed when people receive incentives to start a new activity, presumably by encouraging people to build a stock of behavior. Notably, however, these incentives can backfire if they crowd out intrinsic motivation to engage in an activity, as seen in the experiment by Meier (2007). Furthermore, the studies discussed in this section show that, even if a habit is successfully formed, it can still be only temporary. More research is needed in order to understand how incentives can be used to create habits that persist over longer periods of time.

36.2.2 Breaking Habits The experiments discussed in Section 36.2.1 describe interventions aimed at creating new habits. The conceptual model prescribes that creating a stock of behavior increases the marginal utility from consuming an activity. If past consumption can create habits, could reducing consumption “kill” habits by reducing the stock of behavior? This notion of successfully “killing” a habit is supported by the plausible assumption that the stock of behavior decays over time; or, according to the Becker–Murphy model, there is a “disappearance of the physical and mental effects of past consumption” over time. Consider the exercising example – enjoyment from going to the gym today is likely influenced more by gym visits from the past month than by gym visits from a year ago. If the habitual stock of relevant behavior decays over time, incentivizing people to stop a certain activity for a while can reduce the probability that they will return to their old habits once the incentives are removed. Looking at smoking

cessation during pregnancy and postpartum, Higgins et al. (2012) used incentives in the form of vouchers redeemable for retail items. Their fifty-eight participants were assigned to either contingent- or noncontingent-voucher conditions. In the contingent condition, participants earned vouchers for biochemically verified smoking abstinence, whereas, in the noncontingent condition, receiving the vouchers did not depend on smoking behavior. They found the contingent vouchers increased seven-day point-prevalence abstinence at the end of pregnancy to 37 percent (compared to 9 percent for the noncontingent group) and twelve-week postpartum (33 percent vs. 0 percent, respectively). This effect was sustained through the twenty-four-week postpartum assessment (27 percent vs. 0 percent), which was twelve weeks after the discontinuation of the voucher incentive. In another study looking at smoking cessation, Volpp et al. (2009) introduced incentives spread over a period of time. Specifically, participants were paid to attend a smoking-cessation program, were promised $250 if they refrained from smoking in the following six months, and were promised $400 if they stopped for an additional six months. The authors found smoking decreased significantly during the intervention period, and the quitting rate after the intervention was 9 percent. Sometimes, incentives may be used to break a certain habitual behavior in order to create a new one in its place. Habit-driven purchase patterns are one common illustration. When buyers first purchase an item, for example shampoo, they likely deliberate for some time. If they are happy with the results after use, they continue buying this same shampoo. They will continue this purchasing habit visit after visit to the store and will not spend time searching for alternatives (Ehrenberg, 1991; Khare & Inman, 2006; Seetharaman, 2004; Wood et al., 2002; see the discussion in Carden et al., 2017). Even if other brands produce a superior shampoo that the

Incentive-Based Interventions

customers may prefer to the current one they use (e.g., a new shampoo that was not on the market when the consumer originally chose their current shampoo), they will not notice the advantage because of their now ingrained habit. As Carden et al. (2017) argue, incentives can override this habit, making people activate a “control” mindset. For example, if the superior shampoo brand were to offer an incentive (e.g., buy-one-get-one-free), it might catch the customer’s eye and stop their habitual behavior by making them reconsider their purchase decision. Once the new brand makes it into the shopping cart, the customer will use it for a while, build up a stock of behavior, and create a habit of using this new shampoo even when the discount is discontinued. Together, these studies suggest incentives can help “kill” habits. A successful way to “kill” a habit is to incentivize quitting an activity for a while. Once a person stops an activity, the habitual stock will start decreasing. The goal is to deplete this stock such that, by the time the incentives are removed, the stock will have decreased to the point that the activity is discontinued, as seen in the smoking studies.

36.2.3 Providing Upfront Incentives Implementing a behavior change can be particularly challenging when the costs and benefits of the behavior change are temporally separated, as is often the case. For example, the benefit of eating healthier, going to the gym regularly, or saving for retirement is costly in the present and is only beneficial in the future. Future benefits are discounted differently by different individuals. If the individual’s discount rate is constant and they care about the future in a consistent manner, implementing behavior change depends on whether the future discounted benefits are large enough to outweigh the present costs. However, many individuals do not discount the future in a dynamically consistent manner;

529

instead, their discount rate changes depending on whether they make an intertemporal decision. For example, Read and van Leeuwen (1998) found that 49.5 percent of their participants chose to have a healthy snack when asked what they wanted to eat in a week but 83 percent chose the unhealthy snack when asked what they wanted to eat that day. Similarly, Sadoff et al. (2014) gave participants in their field experiment the choice to change their grocery order, placed a week ago, at the time of delivery. Twenty-one percent of their sample took up this offer and 96 percent of their choices made at the time of delivery led to unhealthier choices. These studies suggest the calculation of present cost versus discounted future benefit changes when the future becomes the present. One way to formalize such dynamic inconsistency is to assume that, when the present is involved in an intertemporal decision, individuals discount future payoffs even more than when two points in the future are involved in that same decision. This inconsistency is called present bias (e.g., Laibson, 1997; O’Donoghue & Rabin, 1999; Strotz, 1956). In a study on credit card debt, Meier and Sprenger (2010) found around 36 percent of participants exhibited present-biased preferences in a purportedly unrelated task designed to capture dynamic inconsistencies. They then found the individuals who were present-biased in this experimental task had a 16 percent higher probability of being in debt. Such present-biased preferences can therefore lead to constant violation of plans made for the future and make the implementation of behavioral change difficult. Present-biased preferences help explain the difficultly people have with changing behavior and also shed light on how incentives could be structured and timed in order to overcome this obstacle. Given present bias, the intuitive and theoretical approach to the timing and structure of incentives is to make them front-loaded and not too far in the future (e.g., Aggarwal et al.,

530

URI GNEEZY, AGNE KAJACKAITE, AND STEPHAN MEIER

2018; O’Donoghue & Rabin, 1999). For an impatient person, providing delayed incentives is less effective in encouraging activity take-up, such as going to the gym, or in increasing activity frequency, such as walking more. To increase efficacy for these types of people, incentives must be given earlier and more frequently. For example, in encouraging an impatient person to walk more, theory predicts incentives given after a month of taking “X” steps per day would not be as effective as incentives given after a week of reaching the goal. Although the theoretical predictions are clear and intuitive, only a small number of empirical studies have explicitly examined the efficacy of different timing and structure of incentives, and the evidence they provide is inconclusive. Carrera et al. (2017) studied the efficacy of a front-loaded incentive structure compared to a smaller constant regular incentive for about 1,000 employees in a Fortune 500 firm. In particular, employees were assigned to receive either $25 per gym visit for the first two weeks and $5 in the remaining six weeks or $10 per visit for the full eight weeks. In both treatments, employees received the incentive for a maximum of two visits per week, that is, the total incentive budget was $160 per person. The results do not support the conjecture that “front-loaded” incentives increase gym membership. Additionally, frontloaded incentives also did not increase the number of visits per member. Bachireddy et al. (2019) found similar results: They manipulated the structure of a financial incentive scheme aimed at increasing physical activity through more daily steps for about 3,500 participants. They compared three different allocations of a given budget across time: (1) a constant payment across time periods; (2) a decreasing incentive schedule, with incentives that started high and decreased over time; and (3) an increasing incentive schedule, with incentives that started low and increased over time. In their study, the constant-incentive schedule

worked better than both the decreasing and the increasing schedule both during the intervention and post-intervention. Overall, although both Carrera et al. (2017) and Bachireddy et al. (2019) showed that incentives work, a frequent and constant incentive structure was more efficacious in increasing exercise than distributing the incentive unevenly across time, even when this distribution amounted to a “front-loaded” scheme. Aggarwal et al. (2018) ran a field experiment with diabetes patients to test whether increasing the frequency of payment for physical activity has any additional impact. Specifically, they looked at whether daily payments would be more effective than monthly payments. They did not find support for this prediction. However, in additional treatments, they found that offering an incentive was only successful in increasing physical activity if impatient individuals reached the goal multiple days in a row. They pose that, if individuals discount monetary incentives less than they discount effort costs of future behavior (as shown in Augenblick et al., 2015), an incentive structure that pays for an action in a given period contingent on an action in other periods can be effective. Notably, their finding was dependent on how individuals discounted the future. Because large heterogeneity exists in discounting between different people, individually tailored incentives may prove to be more effective. Andreoni et al. (2016) show in a work setting that incentive schemes individually tailored to an individual’s discounting pattern are more effective than schemes that do not consider an individual’s impatience. Tailoring the structure of incentives for behavioral change could prove to be very fruitful but more research in the area is needed. An additional difference between delayed incentives and earlier, and more regular, incentives could be that the latter are more salient because people experience them more frequently. John et al. (2018) showed in their field

Incentive-Based Interventions

experiment that incentives, independent of their structure, were ineffective if the incentives were not made conspicuous. Incentives were only effective at increasing physical activity when accompanied by regular reminders about the incentive structure. Related to salience, Kast et al. (2018) found an increase in the interest rate from 0.3 percent to 5 percent had little influence on the savings rate. However, their more immediate and very salient nonmonetary incentive of having to share whether one saved or not with a peer group on a weekly basis almost doubled the savings rates. Of course, multiple differences exist between the different incentive schemes; future research on the salience aspect of incentives, independent of their structure, is necessary. A more challenging but arguably more sustainable way to overcome a high discount factor is to try to lower the immediate cost of the activity. For example, Milkman et al. (2013) proposed a way for people to incentivize themselves, using what they call “temptation bundling.” In their field experiment, “want” activities (page-turner audiobooks) were bundled with “should” behaviors that had delayed benefits (exercising); consider, for example, allowing yourself to watch your favorite TV show only while exercising. Bundling this “should” behavior with this “want” activity could help make the immediate experience of the “should” behavior less painful.

36.2.4 Removal of Barriers and Reducing Switching Costs Before choosing which company to buy a product from, a customer may be indifferent between two products of the same type. However, once the customer chooses a given company, inertia often imposes a switching cost. These switching costs can come in different forms, as this section discusses, but a defining characteristic is their tendency to create inertia in behavior that affects choices and equilibrium in markets. Some of these costs, such as learning about the product,

531

are natural, whereas others are artificial costs imposed by firms, such as repeat-purchase discounts or loyalty programs. The economics literature (e.g., Klemperer, 1987a, 1987b) shows equilibrium in markets with switching costs may be the same as the collusive outcome in an otherwise identical market without switching costs. The higher the switching costs are, the more market power companies have. Burnham et al. (2003) distinguish between three types of switching costs in their typology. The first, procedural switching costs, come in the form of time and effort. Consider a person who is switching from Windows to Mac. That person faces the cost of having to learn a new operating system and interface, something that takes both time and effort. Similarly, a person thinking about changing gyms faces the procedural switching cost of learning how the new gym works, what the timetable of classes is, when to pay, and so on. If the person is already part of a gym, the switching costs may act as a barrier and the member may simply stick with the less desirable old gym. The second type of switching cost is financial and may be artificially created by companies. For example, fees may be associated with switching banks, or lost reward points may be associated with switching airlines. Consider a person who is planning a trip. All else being equal, that person might not choose the cheapest flight available but may instead choose the flight with the airline with which they have a loyalty program. Burnham et al. (2003) term the third type of switching cost “relational.” This cost comes in the form of any psychological or emotional discomfort a person might feel from either the loss of identity or the breaking of bonds. Consider brand loyalty. People get attached to brands to the point that certain products can become a piece of their identity – switching to another brand due to price or availability might thereby bring about psychological and emotional discomfort. The data support this theoretical prediction. For example, Neiman and Vavra (2018) studied

532

URI GNEEZY, AGNE KAJACKAITE, AND STEPHAN MEIER

Sidebar 36.2 Switching costs and market power

Sometimes switching costs are so high that they lead to market power in the hands of a few companies. For instance, in the early days of cell phones, customers who wanted to switch carriers had to also change their phone number. Recognizing this switching cost, the Federal Communications Commission (FCC) in the United States mandated that all wireless carriers had to offer number portability by 2004. Park (2011) looked at the response of wireless pricing to the introduction of number portability. Examining around 100,000 calling plans, Park found that the prices of wireless plans dropped by 6.8 percent in the seven months after the FCC ruling. In other words, the switching cost had been a significant barrier to switching and, once this barrier was removed, companies turned to decreasing prices to acquire new customers and retain existing ones.

data on 700 million purchases in more than 160,000 households over the span of a decade. They found that people are increasingly buying only their one favored product in a given category. The implication of this habit is that companies are no longer forced to diversify and compete for business, creating consolidated market power. To increase competition in markets, legislators often create policies that attempt to reduce the market power of companies associated with switching costs (for an example, see Sidebar 36.2). Incentives can be used to both create and reduce these costs. For example, many companies offer great upfront deals to attract customers and get them “hooked” on a product. Consider a person who does their taxes using a certain software, for example TurboTax. That person needs to invest time and effort into entering information, from addresses to workplace names. TurboTax software can save this information such that the effort of doing their taxes the next year is considerably lower. To avoid procedural switching costs, customers might be willing to pay more just to stick with this software, even if cheaper or better options are available the following year. Amazon’s 1-Click patent epitomizes this cost;

customers only ever need to enter their information once for the system to remember it. Although how much money the patent has brought Amazon is unclear, estimates indicate billions annually. Incentives can also be used to reduce switching costs and help change behavior by removing barriers. Cappelen et al. (2019) ran a large randomized controlled trial in which they tested the hypothesis that incentives that reduce barriers for physical activity can improve academic performance. In the study, college students in Bergen, Norway, were given a free gym membership (worth about $140) for a semester. The authors found this removal of barriers encouraged students to attend the gym more often, which in turn led to an improvement in their academic performance. The prevalent concern regarding contingent incentives in an educational context is that they can crowd out the intrinsic motivation to study. The removal-of-barriers-to-exercise approach has the added bonus of being a more politically feasible, and perhaps more fruitful, way to increase educational outcomes than incentivizing the educational outcomes directly. Homonoff et al. (2019) studied a large-scale wellness program at a university that offered gym membership reimbursements for students who

Incentive-Based Interventions

attended the gym at least fifty times in a sixmonth period. Their data consisted of individual-level administrative data on daily gym attendance over a five-year period: one year before policy implementation, the three years the policy was in place, and one year after policy termination. In other words, they used a natural experiment conducted by the university, considering the before, during, and after effects of the removal-of-barriers approach. Their approach provides a much larger sample (100,000 student/year observations and 1.5 million gym visits) than most experiments. As expected, the authors found an effect at around the fifty-visit threshold and an overall 20 percent increase from the mean. Importantly, they also found that approximately 50 percent of the program effect persisted after policy termination, evidence indicating habit formation. The barrier-removal approach also succeeded in changing behavior when used in the Home Energy Report, which the Opower company sends by mail to millions of households in America regularly. In contrast to previous examples that used financial incentives, this intervention used a social comparison – the report told consumers where they stood in their energy consumption relative to their neighbors. In addition, the report included individual tips on how to reduce energy consumption. Allcott and Rogers (2014) show the reports reduced energy consumption significantly and the effect persisted even for households that stopped receiving the reports postintervention. The positive effect decayed over time but, importantly, stayed significant. Brandon et al. (2017) analyzed the Opower data and concluded the behavior change was mostly due not to changes in habits but rather to investments in capital. In other words, the social-comparison incentives drove participants to overcome the costs associated with switching to better technologies, such as purchasing more efficient appliances.

533

36.3 Incentivizing Behavioral Change: Practical Implications In the current chapter, four ways incentives can affect behavior change have been presented, with a number of supporting empirical examples outlined. The four channels, along with the steps required for successful implementation of behavior change using incentives, are as follows: 1. Incentives can help create habits by building up the stock of behavior. Increasing recent experience makes current behavior less costly and more enjoyable. 2. Incentives can help kill habits by reducing the stock of behavior. Decreasing recent experience makes current behavior costlier and less enjoyable. 3. Incentives can help counter present bias. Using frequent and regular incentives helps change behavior. 4. Incentives can help remove barriers to change. Using incentives to reduce switching costs makes activity take-up cheaper or free.

36.4 Summary and Conclusion This chapter has provided a framework for, and outlined empirical evidence on, how incentives can affect behavior change. As the impediment to behavior change is often not a lack of motivation or a lack of information but difficulties in implementing the change, four channels in which incentives can affect the implementation of behavior change were reviewed. Incentives can create positive habits and can break negative habits. Incentives may also help counteract present-biased preferences. Finally, incentives can help by removing barriers to change. These four channels and the supporting empirical evidence have implications for how incentivebased interventions work and provide guidance on how best to design the incentives for optimal efficacy.

534

URI GNEEZY, AGNE KAJACKAITE, AND STEPHAN MEIER

References Acland, D., & Levy, M. R. (2015). Naiveté, projection bias, and habit formation in gym attendance. Management Science, 61, 146–160. https://doi .org/10.1287/mnsc.2014.2091 Adams, J., Giles, E., McColl, E., & Sniehotta, F. (2014). Carrots, sticks and health behaviours: A framework for documenting the complexity of financial incentive interventions to change health behaviours, Health Psychology Review, 8, 286–295. https://doi .org/10.1080/17437199.2013.848410 Aggarwal, S., Dizon-Ross, R., & Zucker, A. (2018). Incentivizing behavioral change: The role of time preferences. Unpublished manuscript, Chicago Booth University, Chicago, IL. http://faculty.chi cagobooth.edu/rebecca.dizon-ross/research/ papers/Incentive_Design.pdf Allcott, H., & Rogers., T. (2014). The short-run and long-run effects of behavioral interventions: Experimental evidence from energy conservation. American Economic Review, 104, 3003–3037. https://doi.org/10.1257/aer.104.10.3003 Andreoni, J., Callen, M., Khan, Y., Jaffar, K., & Sprenger, S. (2016). Using preference estimates to customize incentives: An application to polio vaccination drives in Pakistan. NBER Working Paper No. 22019. www.nber.org/papers/w22019 Angrist, J., Bettinger, E., Bloom, E., King, E., & Kremer, M. (2002). Vouchers for private schooling in Colombia: Evidence from a randomized natural experiment. American Economic Review, 92, 1535–1558. https://doi.org/ 10.1257/000282802762024629 Ashraf, N., Karlan, D., & Yin, W. (2006). Tying Odysseus to the mast: Evidence from a commitment savings product in the Philippines. Quarterly Journal of Economics, 121, 635–672. https://doi.org/10.1162/qjec.2006.121.2.635 Augenblick, N., Niederle, M., & Sprenger, C. (2015). Working over time: Dynamic inconsistency in real effort tasks. Quarterly Journal of Economics. 130, 1067–1115. https://doi.org/10.1093/qje/qjv020 Bachireddy, C., Joung, A., John, L. K. et al. (2019). Comparing financial incentive structures for promoting physical activity: A randomized controlled trial. Unpublished manuscript,

University of Pennsylvania, Philadelphia, PA. https://static1.squarespace.com/static/ 5353b838e4b0e68461b517cf/t/5b5760c82b6a28 ce3b6a9745/1532453064822/ Incentives_for_Physical_Activity_RCT_Manusc ript_PNAS.pdf Becker, G. S. (1992). Habits, addictions and traditions, Kyklos, 45, 327–345. https://doi.org/10.1111/ j.1467-6435.1992.tb02119.x Becker, G. S., & Murphy, K. M. (1988). A theory of rational addiction. Journal of Political Economy, 96, 675–700. https://doi.org/10.1086/261558 Brandon, A., Ferraro, P., List, J., Metcalfe, R., Price, M., & Rundhammer, F. (2017). Do the effects of social nudges persist? Theory and evidence from 38 natural field experiments. NBER Working Paper No. 23277. www.nber.org/papers/w23277 Burnham, T. A., Frels, J. K., & Mahajan, V. (2003). Consumer switching costs: A typology, antecedents, and consequences. Journal of the Academy of Marketing Science, 31, 109–126. Cappelen, A., Charess, G., Ekström, M., Gneezy, U., & Tungodden, B. (2019). Exercise improves academic performance. https://brage.bibsys.no/ xmlui/bitstream/handle/11250/2452391/DP% 2008.pdf?sequence=1 Carden, L., Wood, W., Neal, D. T., & Pascoe, A. (2017). Incentives activate a control mind-set: Good for deliberate behaviors, bad for habit performance. Journal of the Association for Consumer Research, 2, 279–290. https://doi.org/10.1086/695325 Carrera, M., Royer, H., Stehr, M., & Sydnor, J. (2017). The structure of health incentives: Evidence from a field experiment. NBER Working Paper No. 23188. www.nber.org/papers/w23188 Charness, G., & Gneezy, U. (2009). Incentives to exercise. Econometrica, 77, 909–931. https://doi .org/10.3982/ECTA7416 Ehrenberg, A. S. C. (1991). New brands and the existing market. Journal of the Market Research Society, 33, 285–299. https://doi.org/10.1177/ 147078539103300402 Falk, A., & Kosfeld, M. (2006). The hidden costs of control. American Economic Review, 96, 1611–1630. https://doi.org/10.1257/aer.96.5.1611 Fehr, E., & List, J. A. (2004). The hidden costs and returns of incentives: Trust and trustworthiness

Incentive-Based Interventions

among CEOs. Journal of the European Economic Association, 2, 743–771. https://doi.org/10.1162/ 1542476042782297 Frey, B. (1997). Not Just for the Money: An Economic Theory of Personal Motivation. Cheltenham: Edward Elgar. Friebel, G., Heinz, M., Krueger, M., & Zubanov, N. (2017). Team incentives and performance: Evidence from a retail chain. American Economic Review, 107, 2168–2203. https://doi.org/10.1257/ aer.20160788 Giles, E. L., Robalino, S., McColl, E., Sniehotta, F. F. & Adams, J. (2014). The effectiveness of financial incentives for health behaviour change: Systematic review and meta-analysis. PLoS ONE, 9, e90347. https://doi.org/10.1371/journal .pone.0090347 Gneezy, U., Meier, S., & Rey-Biel, P. (2011). When and why incentives (don’t) work to modify behavior. Journal of Economic Perspectives, 25, 191–209. https://doi.org/10.1257/jep.25.4.191 Gneezy, U., & Rustichini, A. (2000). A fine is a price. Journal of Legal Studies, 29, 1–18. https://doi.org/ 10.1086/468061 Higgins, S. T., Washio, Y., Heil, S. H. et al. (2012). Financial incentives for smoking cessation among pregnant and newly postpartum women. Preventive Medicine, 55, S33–40. https://doi.org/ 10.1016/j.ypmed.2011.12.016 Homonoff, T., Willage, B., & Willén, A. (2020). Rebates as incentives: The effects of a gym membership reimbursement program. Journal of Health Economics, 70, 102285. https://doi.org/ https://doi.org/10.1016/j.jhealeco.2019.102285 John, L. K., Milkman, K. L., Gino, F., Tuckfield, B., & Foschini, L. (2018). The ineffectiveness of inconspicuous incentives: A field experiment on inattention. Unpublished manuscript, Harvard Business School, Cambridge, MA. https://static1 .squarespace.com/static/5353b838e4b0e68461 b517cf/t/5966320a86e6c01e1f7bdffe/149986971 747/IncIncentivesSubmittedMSciBEDec2017.pdf Kast, F., Meier, S., & Pomeranz, D. (2018). Saving more in groups: Field experimental evidence from Chile. Journal of Development Economics, 133, 275–294. https://doi.org/10.1016/ j.jdeveco.2018.01.006

535

Khare, A., & Inman, J. J. (2006). Habitual behavior in American eating patterns: The role of meal occasions. Journal of Consumer Research, 32, 567–575. https://doi.org/10.1086/500487 Klemperer, P. (1987a). Markets with consumer switching costs. Quarterly Journal of Economics, 102, 375–394. https://doi.org/10.2307/1885068 Klemperer, P. (1987b). The competitiveness of markets with switching costs. RAND Journal of Economics, 18, 138–150. https://doi.org/10.2307/ 2555540 Laibson, D. (1997). Golden eggs and hyperbolic discounting. Quarterly Journal of Economics 112, 443–477. https://doi.org/10.1162/ 003355397555253 Levitt, S. D., List, J. A., & Sadoff, S. (2016). The effect of performance-based incentives on educational achievement: Evidence from a randomized experiment. NBER Working Paper No. 22107. www.nber.org/papers/w22107 Meier, S. (2007). Do subsidies increase charitable giving in the long run? matching donations in a field experiment. Journal of the European Economic Association, 5, 1203–1222. https://doi .org/10.1162/JEEA.2007.5.6.1203 Meier, S., & Sprenger, C. (2010). Present-biased preferences and credit card borrowing. American Economic Journal: Applied Economics, 2, 193– 210. https://doi.org/10.1257/app.2.1.193 Mellström, C., & Johannesson, M. (2008). Crowding out in blood donation: Was Titmuss right? Journal of the European Economic Association, 6, 845– 863. https://doi.org/10.1162/JEEA.2008.6.4.845 Milkman, K. L., Minson, J., & Volpp, K. (2013). Holding the hunger games hostage at the gym: An evaluation of temptation bundling. Management Science, 60, 283–299. https://doi.org/10.1287/ mnsc.2013.1784 Neiman, B., & Vavra, J., (2018). The rise in household spending concentration. Available at SSRN. https://papers.ssrn.com/sol3/papers.cfm? abstract_id=3137782 O’Donoghue, T., & Rabin, M. (1999). Doing it now or later. American Economic Review, 89, 103–124. https://doi.org/10.1257/aer.89.1.103 Park, M. (2011). The economic impact of wireless number portability. The Journal of Industrial

536

URI GNEEZY, AGNE KAJACKAITE, AND STEPHAN MEIER

Economics, 59, 714–745. https://doi.org/10.1111/ j.1467-6451.2011.00471.x Read, D., & van Leeuwen, B. (1998). Predicting hunger: The effects of appetite and delay on choice. Organizational Behavior and Human Decision Processes, 76, 189–205. https://doi.org/ 10.1006/obhd.1998.2803 Royer, H., Stehr, M., & Sydnor, J. (2015). Incentives, commitments, and habit formation in exercise: Evidence from a field experiment with workers at a Fortune-500 company. American Economic Journal: Applied Economics, 7, 51–84. https:// doi.org/10.1257/app.20130327 Sadoff, S., Samek, A., & Sprenger, C. (2014). Dynamic inconsistency in food choice: Experimental evidence from a food desert. Becker Friedman Institute for Research in Economics Working Paper No. 2572821; CESR-Schaeffer Working Paper 2015-027. Available at SSRN. https:// papers.ssrn.com/sol3/papers.cfm? abstract_id=2572821## Seetharaman, P. B. (2004). Modeling multiple sources of state dependence in random utility models: A distributed lag approach. Marketing

Science, 23, 263–271. https://doi.org/10.1287/ mksc.1030.0024 Strotz, R. H. (1956). Myopia and inconsistency in dynamic utility maximization. Review of Economic Studies, 23, 165–180. https://doi.org/ 10.2307/2295722 Thaler, R. H., & Sunstein, C. R. (2008). Nudge: Improving Decisions About Health, Wealth, and Happiness. New Haven, CT: Yale University Press. Volpp, K. G., John, L. K., Troxel, A. B., Norton, L., Fassbender, J., & Loewenstein, G. (2008). Financial incentive-based approaches for weight loss: A randomized trial. Journal of the American Medical Association, 300, 2631–2637. https://doi .org/10.1001/jama.2008.804 Volpp, K. G., Troxel, A. B., Pauly, M. V. et al. (2009). A randomized, controlled trial of financial incentives for smoking cessation. New England Journal of Medicine, 360, 699–709. https://doi.org/10.1056/ NEJMsa0806819 Wood, W., Quinn, J. M., & Kashy, D. A. (2002). Habits in everyday life: Thought, emotion, and action. Journal of Personality and Social Psychology, 83, 1281–1297.

37 Monitoring Interventions Thomas L. Webb and Marijn de Bruin

Practical Summary When people set a goal (e.g., to try to lose 5 kg of bodyweight), prompting them to monitor their progress toward the goal (e.g., through keeping track of their sugar consumption, physical activity, and body weight) can help them to make effective changes to their behavior and achieve their goal. However, attaining these benefits is not as easy as simply giving people a pedometer, asking them to read the packets of the food that they eat, and using a weighing scale. This chapter therefore offers a brief overview of the research and psychological theory in this area. A step-by-step guide is also provided to assist the development of effective monitoring interventions. The guide includes considerations with respect to setting a goal that is suitable for monitoring (step 1), designing the monitoring strategy (step 2), and ensuring that people are able to make use of the information that is derived from monitoring (step 3).

37.1 Introduction Evidence suggests that once someone has decided to make a change to their behavior (e.g., to eat less sugar, to reduce their household electricity consumption), monitoring their progress toward this goal makes achieving it more likely (for a review, see Harkin et al., 2016). This is because monitoring allows people to identify discrepancies between goals and actual performance, identify the contextual factors that affect progress, and thereby inform consecutive actions to increase the likelihood of goal achievement. Monitoring thus informs both the need to take and maintain action and when and how it might be best to do so. Unfortunately, evidence also suggests that people often do not monitor their progress (e.g., Webb, Chang, & Benn, 2013), hence monitoring interventions can represent a viable means of promoting changes in behavior. The aim of this chapter is to provide an overview of how

monitoring has been used as an intervention to change behavior, along with practical guidelines for how to use monitoring.

37.2 Definitions Monitoring involves an entity (e.g., a person, group, or organization) taking stock of the current situation (e.g., how much sugar has been consumed that day, when, and where), comparing this to some goal or reference value (e.g., a maximum of 25 g or 6 teaspoons of sugar per day; WHO, 2015), and identifying whether or not there is a discrepancy. Monitoring may also include cognitive and affective appraisals of that discrepancy (e.g., that it is upsetting and needs to be addressed) and identifying contextual factors that are associated with the discrepancy or lack thereof (e.g., that more sugary https://doi.org/10.1017/9781108677318.037

538

THOMAS L. WEBB AND MARIJN DE BRUIN

snacks are consumed when watching television than at other times). Interventions that prompt people to monitor their progress are referred to as monitoring interventions. Given that the primary aim of monitoring is to assess where people are relative to their desired goal, monitoring is probably most relevant for goals that require repeated efforts over time (e.g., reducing sugar to lose weight or unplugging the TV to reduce electricity consumption) rather than goals that can be achieved via a single action (e.g., having bariatric surgery or installing energy-saving light bulbs). Monitoring can take many shapes or forms. Monitoring can be done by the self (termed “selfmonitoring”) or by others, people may monitor behavior or outcomes, the rate of progress versus the absolute size of discrepancy, or monitoring may be active versus passive (see Table 1 in Harkin et al., 2016). For example, people may be asked to use a pedometer to (self-)monitor their physical activity (behavior), the idea being that the person could use the data from a pedometer to evaluate whether they had done sufficient exercise on a given day, to examine progress toward or distance from a reference value (e.g., a target of 10,000 steps a day) or whether progress is sufficient to be “in shape” for a marathon that they have entered three months from now. An example of monitoring by others might involve a practice nurse checking blood sugar levels (i.e., an outcome) in a patient with Type 2 diabetes. Despite the abundant evidence that interventions that prompt monitoring can be effective (see the review by Harkin et al., 2016), a means or method for monitoring is not a panacea for behavior change (Finkelstein et al., 2016; Jakicic et al., 2016) and applying it effectively usually requires more than just giving people a diary or device (see Section 37.5 for potential considerations in this regard).

37.3 Theory and Mechanisms of Change A number of models posit a central role for monitoring, including feedback intervention theory

(Kluger & DeNisi, 1996), goal setting theory (Latham & Locke, 1991; see Chapter 38, this volume), field theory (Lewin, 1951), models of self-awareness (e.g., Duval & Wicklund, 1972), Kanfer and Karoly’s (1972) account of selfregulation, the test-operate-test-exit system (Miller, Galanter, & Pribram, 1960), the “living systems perspective” (Ford, 1987), the model of multiple-goal pursuit (Louro, Pieters, & Zeelenberg, 2007), and models of self-control (see Chapter 11, this volume). Monitoring is also a core part of cybernetic models such as control theory (Carver & Scheier, 1982; Powers, 1973; see Chapter 9, this volume). Figure 37.1 draws on these theories to present a framework showing how monitoring can lead to behavior change. Monitoring is typically viewed as one of the processes by which intentions are translated into behavior (de Bruin et al., 2012; Banas et al., 2017). Monitoring therefore follows from the decision to strive for a desired goal or outcome. However, monitoring can also influence motivation – both because it can inform decisions about whether the goal needs to be revised in the light of current progress (Campion & Lord, 1982) and because monitoring can influence the strength of motivation to act to further goal pursuit (e.g., Reynolds et al., 2018). It is also important to note that people may inadvertently obtain information that prompts them to set new goals (e.g., someone realizes that their trousers no longer fit them and so sets the goal to try to lose some weight). Monitoring the current state and comparing it with the desired state, as specified by a goal or reference value, should serve to identify discrepancies (e.g., that one has consumed more calories than intended). Experienced affect is hypothesized to signal the outcome of monitoring (Carver & Scheier, 1990), such that people experience positive affect when their (rate of) progress is better than expected and negative affect when their (rate of) progress is poorer than expected, although the precise nature of the affective states that people experience is also hypothesized to depend on whether the

Monitoring Interventions

Desired state or reference value

Monitoring

539

Anticipated affect

Experienced affect

Intention to achieve desired state

Behavior (goal-directed action or response)

Current state

Figure 37.1 A framework showing how monitoring may lead to changes in behavior

goal is oriented toward approaching a desired outcome or avoiding an undesired outcome (Carver, 2003). Control theory predicts that affect drives goal-directed behavior, such that negative affect leads people to increase goal-directed efforts in order to reduce the discrepancy, while positive affect may lead people to “coast,” switch attention to other goals, or set a more ambitious goal (Carver, 2003; for a detailed review, see Chapter 9, this volume). While there is good evidence that progress has the anticipated affective consequences (i.e., people feel good when they are doing well and bad when progress is not so good; Carver, Lawrence, & Scheier, 1999; Moberly & Watkins, 2010; Reynolds et al., 2018) and that affect is associated with intentions to reduce the discrepancy (Reynolds et al., 2018), the evidence that experienced affect drives actual goaldirected behavior, as hypothesized by control theory, is not strong. Indeed, there may even be stronger evidence for the opposite prediction, namely that negative affect undermines rather than impels goal pursuit because it leads people to prioritize dealing with the aversive state rather than tackling the discrepancy that may have produced that state (Tice, Bratslavsky, & Baumeister, 2001). Furthermore, when negative affect is seen as indicating that it is unlikely that the goal will be attained, it can also lead

people to revise the goal (e.g., increase the allowed number of calories a day) or abandon it altogether (for discussions of goal revision, see Campion & Lord, 1982; Wang & Mukhopadhyay, 2012). Thus, the affect that accrues as a function of monitoring is informative; however, it is unlikely to directly influence goal-directed action. Rather, experienced affect likely helps the person to predict how they are likely to feel should the same situation occur again in the future (Brown & McConnell, 2011) and it is this anticipated affect that drives intentions to take action (e.g., to avoid the aversive state before it occurs; Baumeister et al., 2007). If a suitable action is chosen and implemented (e.g., the person switches to a low-calorie soft drink), then it should have an impact on the current state (e.g., calorie consumption), alongside any disturbances (e.g., an invitation to have a glass of champagne to celebrate a colleague’s success; Beck et al., 2017). Monitoring thus enables people to recognize when additional effort or self-control is needed (Fishbach et al., 2012; Myrseth & Fishbach, 2009). Monitoring should also signal the success, or otherwise, of current goal-directed actions and thus shape motivation to continue with the current means of goal pursuit or switch to an alternative (i.e., influence the means by which

540

THOMAS L. WEBB AND MARIJN DE BRUIN

people strive for the goal) or abandon the goal altogether (Campion & Lord, 1982).

37.4 Evidence Base Harkin and colleagues (2016) quantified the impact of monitoring interventions on rates of goal attainment. A systematic literature search identified 138 studies (N = 19,951) that randomly allocated participants to an intervention designed to promote monitoring of goal progress versus a control condition. All of the studies reported the effects of the treatment on (1) the frequency of progress monitoring and (2) subsequent goal attainment. A random effects model revealed that, on average, the interventions increased the frequency of monitoring goal progress (d+ = 1.98, 95% CI [1.71, 2.24]) and promoted goal attainment (d+ = 0.40, 95% CI [0.32, 0.48]; d+ = 0.44 for objective and d+ = 0.34 for self-reported outcomes). Furthermore, changes in the frequency of progress monitoring mediated the effect of the interventions on goal attainment. There have also been reviews of monitoring interventions in specific domains. For example, there are reviews on the impact of monitoring on household energy efficiency (Abrahamse et al., 2005), adult problem behaviors (Febbraro & Clum, 1998), and behavior management in special education classrooms (Webber et al., 1993); multiple reviews on the monitoring of clinical indicators such as weight, blood pressure, blood glucose, or coagulation to support self-management and treatment decisions (Bray et al., 2010; Butryn et al., 2007; Clar et al., 2010; Heneghan et al., 2012; Richardson et al., 2008); and reviews of the effects of monitoring – particularly, seeking feedback on performance – in organizational settings (Anseel et al., 2015; Ashford, 2003). All of these conclude that monitoring can be an effective way to promote changes in behavior. To the best of our knowledge, there has not yet been a direct comparison between the relative effectiveness of monitoring for promoting the initiation versus maintenance of behavior; for

example, it is not yet known whether monitoring household electricity consumption is more effective in initiating new behaviors (e.g., encouraging people to buy energy-efficient appliances) or maintaining existing behaviors (e.g., encouraging them to continue to take showers instead of baths). Studies do, however, suggest that the nature of monitoring may differ depending on whether people are far from or close to the end state (e.g., Bonezzi et al., 2011) and that assessments of progress may be biased so as to maintain motivation (e.g., Howansky, Dominick, & Cole, 2018; Huang, Zhang, & Broniarczyk, 2012).

37.5 Step-by-Step Guide Table 37.2 summarizes relevant considerations when using monitoring to promote changes in behavior and outcomes. Three steps are relevant for effective monitoring interventions: (1) setting a goal that is suitable for monitoring; (2) designing the monitoring strategy; and (3) processing the obtained information. Table 37.2 describes the tasks within each of these steps, along with relevant considerations and examples. A summary of how monitoring intervention techniques feature in various taxonomies of behavior change techniques is presented in Sidebar 37.1.

37.5.1 Typical Means of Delivery Monitoring interventions can be used by individuals but also by groups of people and organizations to monitor progress toward a shared goal (e.g., access to resources, revenue). A variety of means and methods of monitoring can be used, such as written or electronic diaries, mobile phone apps, or medical records. While evidence indicates variability in the apparent effectiveness of different techniques (e.g., Harkin et al., 2016), techniques are often confounded with the nature of the goal or outcome being monitored (e.g., a blood pressure monitor can only be used to measure blood pressure). Therefore, studies are

Monitoring Interventions

541

Sidebar 37.1 Behavior change technique(s) related to monitoring from existing taxonomies

Monitoring techniques feature prominently in existing taxonomies of behavior change. For example, the widely used behavior change technique taxonomy version 1 (BCTTv1; Michie et al., 2013) differentiates between techniques that involve monitoring behavior versus the outcomes of behavior, self-monitoring versus monitoring by others, and monitoring with versus without feedback. The BCTTv1 also describes a technique that involves explicitly drawing attention to discrepancies between a person’s current behavior (in terms of the form, frequency, duration, and/ or intensity of that behavior) and the person’s previously set outcome goals, behavioral goals, or action plans. This is described as going beyond monitoring – likely in the sense that the technique explicitly focuses on identifying discrepancies, while monitoring could indicate that there is no discrepancy. Similar techniques are defined in the intervention mapping taxonomy (Kok et al., 2016, see also Chapter 19, this volume), which differentiates between self-monitoring and methods that involve receiving feedback from others on monitored behavior and/or outcomes (i.e., the methods “Feedback” and “Consciousness raising”). The intervention mapping taxonomy explicitly defines self-monitoring in terms of monitoring behavior rather than a physiological state or outcomes; however, self-monitoring could concern states or outcomes when it is clear what people can do to achieve those outcomes. The taxonomy also suggests that, for self-monitoring to be effective, the data must be used and interpreted and that achieving the goal must be reinforcing to the individual. Table 37.1 How behavior change taxonomies incorporate self-monitoring techniques, along with related techniques and descriptions Taxonomy BCTTv1 (Michie et al., 2013)

Technique and Cluster (if any)

Primary/ Closely Related Description

Closely Related Draw attention to discrepancies between a person’s current behavior (in terms of the form, frequency, duration, or intensity of that behavior) and the person’s previously set outcome goals, behavioral goals, or action plans (goes beyond self-monitoring of behavior) Cluster: 2. Feedback and Closely Related Observe or record behavior with the person’s monitoring knowledge as part of a Technique: 2.1. behavior change strategy Monitoring of behavior by others without feedback Cluster: 1. Goals and planning Technique: 1.6. Discrepancy between current behavior and goal

Continued

542

THOMAS L. WEBB AND MARIJN DE BRUIN

Table 37.1 (Cont.) Taxonomy

Intervention Mapping Taxonomy (Kok et al., 2016)

Technique and Cluster (if any)

Primary/ Closely Related Description

Cluster: 2. Feedback and Closely Related Monitor and provide informative or evaluative monitoring feedback on performance Technique: 2.2. Feedback of the behavior (e.g., form, on behavior frequency, duration, intensity) Establish a method for the Cluster: 2. Feedback and Primary person to monitor and monitoring record their behavior(s) as Technique: 2.3. part of a behavior change Self-monitoring of strategy behavior Establish a method for the Cluster: 2. Feedback and Primary person to monitor and monitoring record the outcome(s) of Technique: 2.4. their behavior as part of a Self-monitoring of behavior change strategy outcome(s) of behavior Cluster: 2. Feedback and Closely Related Observe or record outcomes of behavior with the person’s monitoring knowledge as part of a Technique: 2.5. behavior change strategy Monitoring of outcome(s) of behavior [by others] without feedback Cluster: 2. Feedback and Closely Related Provide feedback about the body (e.g., physiological or monitoring biochemical state) using an Technique: 2.6. external monitoring device Biofeedback as part of a behavior change strategy Cluster: 2. Feedback and Closely Related Monitor and provide feedback on the outcome of monitoring performance of the Technique: 2.7. Feedback behavior on outcome(s) of behavior Self-monitoring of Primary Prompting the person to keep behavior a record of specified behavior(s)

Continued

Monitoring Interventions

543

Table 37.1 (Cont.) Taxonomy

Technique and Cluster (if any)

Primary/ Closely Related Description

Feedback

Closely Related Giving information to individuals and environmental agents regarding the extent to which they are accomplishing learning or performance goals or the extent to which performance is having an impact Consciousness raising Closely Related Providing information, feedback, or confrontation about the causes, consequences, and alternatives for a problem or a problem behavior Organizational diagnosis Closely Related Assessing of organizational and feedback structures and employees’ beliefs and attitudes, desired outcomes, and readiness to act, using surveys and other methods Note. BCTTv1 = Michie et al.’s (2013) behavior change technique taxonomy version 1.

needed that compare the effects of different methods of monitoring progress for the same goal (e.g., Helsel, Jakiac, & Otto [2007] compared the impact of completing detailed vs. abbreviated diaries on monitoring food intake). Technological advances make monitoring behavior and outcomes increasingly easy; however, research suggests that many people quickly discontinue monitoring after it has been initiated (Hargreaves, Nye, & Burgess, 2010; Webb et al., 2014). Hence, embedding monitoring strategies in an engaging program that also involves using and discussing the monitored information is likely more effective in ensuring continued monitoring and achieving the desired outcomes. Although monitoring can be fully automated to reduce participant burden (e.g., people can

monitor their electricity consumption using a digital meter at home), it is likely that expert input is – for the average person – beneficial for setting appropriate goals, designing and sticking to the monitoring strategy, processing the monitored information, and deciding on an effective course of action.

37.5.2 Target Audience and Behaviors The benefits of monitoring for promoting goal achievement have been demonstrated across a range of goals and samples, although there is some evidence that effect sizes are smaller for people with particular medical conditions compared with members of the general public (Harkin et al., 2016). Monitoring is particularly suitable

Table 37.2 Steps and considerations for using monitoring to promote changes in behavior Task

Description

Step 1: Set a goal that is suitable for monitoring Goal setting Set or agree a goal defined in terms of the behavior or outcome to be achieved

Considerations

Example

The goal should be concrete, measurable, time-bound, and personally relevant Additionally, for outcome monitoring, the person should know what behavior(s) would help to achieve the desired outcome

Form the intention to eat 2 pieces of fruit each day for 1 month Reduce home electricity consumption by 20 percent in the next 3 months

Step 2: Design the monitoring strategy Monitoring may focus on different aspects, such as Decide what to monitor Monitoring may focus on the performance (or not) of behavior, the context in outcome, the behaviors which behavior is (or is not) performed, changes in leading to the outcome, and/ behavior or outcomes, rates of progress, or absolute or contextual variables discrepancy from a reference value associated with performance Identify a reference value

Decide on the frequency of monitoring

Monitor calorie consumption in an attempt to reduce BMI to < 30 within 12 months (the end goal) Take note of internal (e.g., stress) or external (e.g., social event) factors associated with excessive calorie consumption The person must be able to observe progress against A person first aims to eat 1 piece of fruit per day The reference value may be the reference value within a reasonable time frame (end goal: 2 pieces per day) the desired future goal, A person compares current home electricity intermediate steps toward that consumption to that over the past year; or goal, a current or past state, or electricity consumption in the average the state of other people household Monitoring can be continuous Monitoring may need to be more frequent for behaviors In the early stages of an intervention, calorie intake is measured daily to identify the impact or outcomes that change quickly versus slowly; and (e.g., to allow the person to of adjustments to diet. Once an appropriate frequency may differ between the early stages of goal make immediate diet has been identified, monitoring is done striving (behavior initiation) and later stages (behavior adjustments) or intermittent only at monthly intervals to assess long-term maintenance) (e.g., to capture longer-term maintenance The advantages of frequent monitoring should be patterns) weighed against the costs (e.g., effort), as people may quickly disengage from monitoring if the costs do not seem to outweigh the benefits Frequent monitoring of slowly changing processes may be de-motivating Continued

Decide on the required duration of monitoring

Decide whether immediate and/or delayed feedback is needed

Identify a means or method for monitoring

Monitoring should cover the period of time in which the person performs (or intends to perform) the desired behavior and/or might expect to observe changes in respective outcomes. However, monitoring could also continue after the goal has been achieved (e.g., to assess whether the change is sustained and to monitor for potential relapse) Feedback can be provided immediately following behavior or assessment of the desired outcome (immediate feedback) or at a later point in time (delayed feedback)

The advantages of monitoring over longer time periods A person monitors electricity consumption until the desired energy savings have been should be weighed against the costs of so doing achieved and then for another six months to Monitoring behavior or outcomes over longer time ensure that the desired outcomes have been periods can allow the person to detect patterns of sustained success/failure and associated contextual factors, as well as allowing reflection on performance over time and whether the current strategy should be maintained or adapted

The approach might involve self-monitoring, monitoring by another or others, or relevant behaviors and/or outcomes may be monitored automatically (e.g., technology such as a pedometer or app)

The method for monitoring should be reliable, valid, and easy to use Evidence suggests that information obtained from monitoring that is physically recorded has a greater impact on goal striving

Immediate feedback allows the person to take immediate corrective action when behavior has not been performed Delayed feedback may be more appropriate when immediate feedback could interfere with goal striving (e.g., be demotivating or distracting)

An app signals to a person that lights in their house are still on, consuming energy (immediate feedback) A patient checks medication reports over the last three months and identifies that most medication doses are missed in the morning during the weekend (delayed feedback) People are asked to record what they have eaten in a diary, using items from a validated food frequency questionnaire A nurse weighs a patient each time that they attend a medical appointment, using a calibrated scale

Continued

Table 37.2 (Cont.) Task

Description

Considerations

Example

Present the information The monitored data are fed back Data must be presented in a familiar, informative format An app presents a monthly summary of physical to the user to the user that allows the user to evaluate their progress activity as monitored with a pedometer, including number of days per week with exercise, duration and intensity of exercise, and the discrepancy between these indicators and the desired state. Additional data on geographical location or psychological state (e.g., short questionnaires completed over the last 4 weeks) could be linked with physical activity indicators Step 3: Make use of the information derived from monitoring Ensure that the person is willing and able to objectively An app designed to promote physical activity Users need to reflect on the Ensure that users invites a person to self-affirm or complete a and accurately appraise the information obtained by monitored information in actively engage with positive psychology exercise, before the monitoring procedure. This may require relation to their goals and and respond to the prompting them to compare their actual intervention strategies for overcoming defensive decide on a course of action information progress with their desired progress. The app reactions (i.e., step up efforts, change guides the user through an exercise of strategy, adjust or disengage Expert input (either in person or via eHealth or self-help identifying personal and contextual drivers of interventions) may be required for people to decide on from the goal) (un)successful performance, common an appropriate route of action if goal progress or solutions to well-known physical activity maintenance is insufficient barriers, and finally setting an action and coping plan for the coming month

Monitoring Interventions

for goals that cannot be immediately achieved by a person but require repeated effort over time. As goal striving can be considered a volitional process – and the process of monitoring, processing the information, and responding to discrepancies is likely to be relatively effortful – it is important that the goal is important to the individual (e.g., personally relevant). Furthermore, when outcomes are monitored, it is important that people know what behaviors they can perform in order to change those outcomes. It is also important that the information derived from monitoring is adequately processed and that potential negative consequences of “failing” are mitigated. For example, if people struggle to take their diabetes medication as prescribed due to, for example, depressive symptoms (Gonzalez et al., 2007), then presenting information about exactly how many doses of medication have been missed could lead to negative affect, disengagement, and/or denial. Another instance where monitoring can be dysfunctional is when it draws attention to processes that are both negative and difficult to control. For example, Ickes et al. (2003) found that monitoring relationships for signs of infidelity undermined trust and satisfaction in the relationship. Hence, it is important that the monitoring strategy is carefully designed and piloted with the target group.

37.5.3 Enabling or Inhibiting Factors Evidence suggests that interventions that combine (self-)monitoring with other behavior change techniques (see Chapter 20, this volume) derived from control theory (e.g., goal setting, action planning, and some forms of feedback) tend to engender larger effects than interventions that did not incorporate these additional behavior change techniques (e.g., Dombrowski et al., 2012; Febbraro & Clum, 1998; Greaves et al., 2011; Harkin et al., 2016; Michie et al., 2009). These combinations of techniques are likely to be effective because they target different self-regulatory processes that serve to bolster the impact of progress monitoring. That

547

is, goal setting may help people to set appropriate reference values (Bandura, 1991; Carver & Scheier, 1982; see Chapter 38, this volume), immediate feedback on behavior facilitates attention to and reinforces ongoing performance (Ashford, 1986; Della Libera & Chelazzi, 2006; Kluger & DeNisi, 1996), and planning helps people to act on discrepancies (for reviews, see Carraro & Gaudreau, 2013; Gollwitzer & Sheeran, 2006; see Chapter 6, this volume). The finding that discovering that lack of progress can undermine motivation (e.g., Reynolds et al., 2018) suggests that additional techniques like self-affirmation (prompting people to self-affirm their strengths prior to evaluating progress) might be used to protect motivation to pursue the goal. In the Adherence Improving self-Management Strategy (AIMS) intervention (de Bruin et al., 2010; 2017; see Sidebar 37.2), techniques derived from attribution theory (e.g., attributing failure to controllable causes) and motivational interviewing (e.g., rolling with resistance; see Chapter 45, this volume) helped participants maintain motivation and focus on what could be learned from the relative lack of progress that could be used to improve future performance.

37.5.4 Training and Skills Required Interventions designed to prompt people to monitor their progress typically have very large effects on the frequency of monitoring (Harkin et al., 2016), suggesting that (1) people typically do not monitor their progress without being prompted to do so (Webb et al., 2013) and (2) training and skills are probably not needed to prompt people to monitor their progress. However, the same review did observe smaller effects of interventions designed to promote monitoring on goal achievement, suggesting that the primary challenge is ensuring that monitoring, or the information derived from monitoring, is processed and used to promote goal achievement. In other words, training, skills, and/ or experience may be required to address the

548

THOMAS L. WEBB AND MARIJN DE BRUIN

Sidebar 37.2 Self-monitoring in the Adherence Improving self-Management Strategy (AIMS)

The Adherence Improving self-Management Strategy (AIMS) is a program that is designed to assist people receiving medication for a long-term condition to take their medication regularly and persistently. In many long-term conditions, irregular intake of medication (e.g., deciding to skip a dose or simply forgetting to take the medication) or discontinuation of the medication against the physician’s advice is quite common. The AIMS intervention is a short intervention delivered by trained nurses during routine visits to the clinic that has been found to be effective and costsaving. Next, a summary of the intervention and its evaluation is provided but a more elaborate description is given in de Bruin et al. (2005, 2010, 2017). When offering AIMS, nurses first ensure that patients have sufficient understanding of the disease and treatment, then support patients’ motivation for adhering to the treatment, followed by guiding patients through setting challenging yet feasible adherence goals (e.g., take the medication every day at 8:00 a.m. and 8:00 p.m.). After that, patients are given an electronic medication bottle (e.g., the Medication Event Monitoring System, or MEMS; Aardex, Ltd.) that registers the date and time that the bottle is opened. Patients take their medication from this bottle for a period of, for example, three months until the next visit to the clinic. During this second visit, the patient and nurse download the data, which are then displayed in plots that are easy to interpret (see Figure 37.2). This allows the patient and nurse to examine whether medication use has been consistent with the patient’s goals for adherence. In case of suboptimal adherence (e.g., missed doses of medication), the patient and nurse examine whether there are any recognizable patterns to the 3 AM

Time of day

11 PM 7 PM 3 PM 11 AM 7 AM 3 AM 01/01/2019

15/01/2019

29/01/2018 Consecutive days

12/02/2018

26/02/2019

Figure 37.2 A chronology plot depicting a patient’s medication intake from a Medication

Event Monitoring System Note. The x-axis shows the twenty-four hours of the day; the y-axis shows the consecutive days; the “dots” show when the pill bottle has been opened; the “triangles” show when morning or evening doses have been missed. In this example, the patient initially used the medication regularly but this rapidly deteriorated to levels that are too low and possible even risky. Within three months of the AIMS intervention, this patient showed near-perfect levels of adherence. The Medication Event Monitoring System (MEMS®; AARDEX Group, Belgium).

Monitoring Interventions

549

nonadherence (e.g., whether more evening or weekend doses have been missed) and explore what the cause(s) of missing medication might have been (e.g., lack of routines, not willing to take medication in the presence of colleagues or friends). The patient and the nurse then identify solutions to these barriers to adherence, which are written down in action and/or coping plans. The patient then sets a goal for adherence for the next three months, uses the pill bottle again, the data are downloaded and discussed during the next clinic visit, and so forth. This continues until patients have achieved and maintained their adherence goals for at least three to six months. considerations detailed in Table 37.2 – particularly steps 1 and 3 that involve helping people to set appropriate goals and process and use the monitored information.

37.5.5 Intensiveness Harkin et al. (2016) found that the duration of interventions designed to promote monitoring had no impact on the frequency of progress monitoring, suggesting that even relatively brief interventions can be effective in prompting people to monitor their progress. However, those interested in using monitoring to promote behavior change should consider the frequency and duration with which people are prompted to monitor, balancing the need to collect enough information to be informative without overburdening the person. For example, weekly measurements may be sufficient to monitor the effects of a diet on weight loss, as no relevant changes would be expected in a shorter time span and frequently measuring weight without seeing much progress can be discouraging. However, for other outcomes, like home energy consumption, it may be informative to monitor electricity use daily and immediately see the effects of changes in behavior (e.g., switching equipment off instead of keeping them on standby). After monitoring the impact of these initial changes in energy consumption, it may then be sufficient to monitor energy consumption only monthly or quarterly to check whether consumption levels have been sustained. Indeed, this

may explain evidence that the frequency with which people engage with and process the information generated by some monitoring devices (e.g., home electricity monitors) declines rapidly (Hargreaves et al., 2010; Webb et al., 2014), perhaps as the information becomes less novel (termed “the fallback effect”; Wilhite & Ling, 1995), successful changes have been made, or people have disengaged from the goal.

37.5.6 Evaluation of Fidelity Monitoring can take many shapes and forms, and measures for assessing fidelity need to be tailored to the focal goal or domain and the means or method used. As a general rule, engagement with the means or method for monitoring will be an important indicator of monitoring. This can be more easily evaluated for methods that automatically and continuously collect information (e.g., an app, pedometer, or electronic pill bottle), as written diaries cannot be monitored from a distance and – if people want to – can simply be completed when the fidelity of the intervention is assessed. As Table 37.2 indicates, for monitoring to be effective people must first set a goal (step 1) – which should be relatively easy to assess in most interventions. However, whether people have processed and engaged with the monitored information (step 3) is likely harder to assess, although this may depend on the mode of delivery (e.g., it may be easier to assess whether information has been processed and used in digital or face-to-face interventions

550

THOMAS L. WEBB AND MARIJN DE BRUIN

than in more private forms of monitoring such as written diaries).

37.5.7 Evaluation of Effectiveness The primary criterion for determining the effectiveness of an intervention designed to promote monitoring is changes in the respective behavior and/or desired outcomes. Additionally, it is informative to examine the degree to which changes can be attributed to changes in monitoring (Harkin et al., 2016). The effectiveness of monitoring could be evaluated through collecting baseline data, introducing the monitoring intervention, and then observing whether the behaviors or outcomes of interest change in the wake of the intervention (this is termed an “N-of-1 study”; Shaffer et al., 2018). It can also be valuable to measure putative mediators (e.g., how people feel about their progress, their ability to identify the need to act, their motivation to continue to pursue the goal, and/or their confidence in their ability to achieve the relevant outcome).

37.6 Concluding Remarks Monitoring interventions have a strong and robust evidence base that supports their use in promoting changes in behavior. However, monitoring is not a panacea for behavior change, perhaps because there are a wide variety of means and methods for promoting monitoring. Those interested in using monitoring for promoting behavior therefore need to consider the most appropriate strategy, bearing in mind that this may differ as a function of the individual who needs to change, the focal behavior, and the context. It is envisaged that the step-by-step guide presented in this chapter will help researchers and practitioners to work through – and, importantly, empirically test – the relevant considerations and, ultimately, develop effective behavior change interventions.

References Abrahamse, W., Steg, L., Vlek, C., & Rothengatter, T. A. (2005). Review of intervention studies aimed at household energy conservation. Journal of Environmental Psychology, 25, 273–291. https:// doi.org/10.1016/j.jenvp.2005.08.002 Anseel, F., Beatty, A., Shen, W., Lievens, F., & Sackett, P. R. (2015). How are we doing after 30 years? A meta-analytic review of the antecedents and outcomes of feedback-seeking behavior. Journal of Management, 41, 318–348. https://doi.org/ 10.1177/0149206313484521 Ashford, S. J. (1986). Feedback-seeking in individual adaptation: A resource perspective. Academy of Management Journal, 29, 465–487. https://doi .org/10.2307/256219 Ashford, S. J. (2003). Reflections on the looking glass: A review of research on feedback-seeking behavior in organizations. Journal of Management, 29, 773–799. https://doi.org/ 10.1016/S0149-2063(03)00079-5 Banas, K., Lyimo, R. A., Hospers, H. J., van der Ven, A., & de Bruin, M. (2017). Predicting adherence to combination antiretroviral therapy for HIV in Tanzania: A test of an extended theory of planned behavior model. Psychology & Health, 32, 1249–1265. https://doi.org/10.1080/ 08870446.2017.1283037 Bandura, A. (1991). Social cognitive theory of selfregulation. Organizational Behavior and Human Decision Processes, 50, 248–287. https://doi.org/ 10.1016/0749-5978(91)90022-L Baumeister, R. F., Vohs, K. D., DeWall, C. N., & Zheng, L. (2007). How emotion shapes behavior: Feedback, anticipation, and reflection, rather than direct causation. Personality and Social Psychology Review, 11, 167–203. https://doi.org/ 10.1177/1088868307301033 Beck J. W., Scholer, A. A., & Hughes, J. (2017). Divergent effects of distance versus velocity disturbances on emotional experiences during goal pursuit. Journal of Applied Psychology, 102, 1109–1123. https://doi.org/10.1037/apl0000210. Bonezzi, A., Brendl, C. M., & De Angelis, M. (2011). Stuck in the middle: The psychophysics of goal pursuit. Psychological Science, 22,

Monitoring Interventions

607–612. https://doi.org/10.1177/ 0956797611404899 Bray, E. P., Holder, R., Mant, J., & McManus, R. J. (2010). Does self-monitoring reduce blood pressure? Meta-analysis with meta-regression of randomized controlled trials. Annals of Medicine, 42, 371–386. https://doi.org/10.3109/ 07853890.2010.489567 Brown, C. M., & McConnell, A. R. (2011). Discrepancy-based and anticipated emotions in behavioral self-regulation. Emotion, 11, 1091–1095. https://doi.org/10.1037/a0021756 Butryn, M. L., Phelan, S., Hill, J. O., & Wing, R. R. (2007). Consistent self-monitoring of weight: A key component of successful weight loss maintenance. Obesity, 15, 3091–3096. https://doi .org/10.1038/oby.2007.368 Campion, M. A., & Lord, R. G. (1982). A control systems conceptualization of the goal-setting and changing process. Organizational Behavior and Human Performance, 30, 265–287. https://doi .org/10.1016/0030-5073(82)90221-5 Carraro, N., & Gaudreau, P. (2013). Spontaneous and experimentally induced action planning and coping planning for physical activity: A meta-analysis. Psychology of Sport and Exercise, 14, 228–248. http://dx.doi.org/10.1016/ j.psychsport.2012.10.004 Carver, C. S. (2003). Pleasure as a sign you can attend to something else: Placing positive feelings within a general model of affect. Cognition and Emotion, 17, 241–261. https://doi.org/10.1080/ 02699930302294 Carver, C. S., & Scheier, M. F. (1982). Control theory: A useful conceptual framework for personalitysocial, clinical, and health psychology. Psychological Bulletin, 92, 111–135. https://doi .org/10.1037/0033-2909.92.1.111 Carver, C. S., & Scheier, M. F. (1990). Origins and functions of positive and negative affect: A control-process view. Psychological Review, 97, 19–35. https://doi.org/10.1037/0033-295X.97.1.19 Carver, C. S., Lawrence, J. W., & Scheier, C. S. (1999). Self-discrepancies and affect: Incorporating the role of feared selves. Personality and Social Psychology Bulletin, 25, 783–792. https://doi.org/ 10.1177/0146167299025007002

551

Clar, C., Barnard, K., Cummins, E., Royle, P., & Waugh, N. (2010). Self-monitoring of blood glucose in type 2 diabetes: Systematic review. Health Technology Assessment, 14, 1–140. https:// doi.org/10.3310/hta14120 de Bruin, M., Hospers, H. J., van Breukelen, G. J., Kok, G., Koevoets, W. M., & Prins, J. M. (2010). Electronic monitoring-based counseling to enhance adherence among HIV-infected patients: a randomized controlled trial. Health Psychology, 29, 421–428. https://doi.org/10.1037/a0020335 de Bruin, M., Hospers, H. J., van den Borne, H. W., Kok, G., & Prins, J. M. (2005). Theory and evidence-based intervention to improve adherence to antiretroviral therapy among HIV-infected patients in the Netherlands: a pilot study. AIDS Patient Care and STDs, 19, 384–394. https://doi .org/10.1089/APC.2005.19.384 de Bruin, M., Oberjé, E. J. M., Viechtbauer, W. et al. (2017). Effectiveness and cost-effectiveness of a nurse-delivered intervention to improve adherence to treatment for HIV: A pragmatic, multicentre, open-label, randomised clinical trial. The Lancet Infectious Diseases, 17, 595–604. https://doi.org/ 10.1016/S1473-3099(16)30534-5 de Bruin, M., Sheeran, P., Kok, G. et al. (2012). Self-regulatory processes mediate the intentionbehavior relation for adherence and exercise behaviors. Health Psychology, 31, 695–703. https://doi.org/10.1037/ a0027425 Della Libera, C., & Chelazzi, L. (2006). Visual selective attention and the effects of monetary rewards. Psychological Science, 17, 222–227. https://doi.org/10.1111/ j.1467-9280.2006.01689.x Dombrowski, S. U., Sniehotta, F. F., Avenell, A., Johnston, M., MacLennan, G., & Araújo-Soares, V. (2012). Identifying active ingredients in complex behavioural interventions for obese adults with obesity-related co-morbidities or additional risk factors for co-morbidities: A systematic review. Health Psychology Review, 6, 7–32. https://doi.org/ 10.1080/ 17437199.2010.513298 Duval, S., & Wicklund, R. (1972). A Theory of Objective Self-Awareness. New York: Academic Press. Febbraro, G. A. R., & Clum, G. A. (1998). Meta-analytic investigation of the effectiveness of self-regulatory

552

THOMAS L. WEBB AND MARIJN DE BRUIN

components in the treatment of adult problem behaviors. Clinical Psychology Review, 18, 143–161. https://doi.org/10.1016/S0272-7358 (97)00008-1 Finkelstein, E. A., Haaland, B. A., Bilger, M. et al. (2016). Effectiveness of activity trackers with and without incentives to increase physical activity (TRIPPA): A randomised controlled trial. The Lancet Diabetes and Endocrinology, 4, 983–995. https://doi.org/10.1016/S2213-8587(16)30284-4 Fishbach, A., Touré-Tillery, M., Carter, T. J., & Sheldon, O. J. (2012). The problem with self-control. Paper presented at the Society for Personality and Social Psychology Conference, January, San Diego, CA, January 26–28. Ford, D. H. (1987). Humans As Self-Constructing Living Systems: A Developmental Perspective on Behavior and Personality. Hillsdale, NJ: Lawrence Erlbaum Associates. Gollwitzer, P. M., & Sheeran, P. (2006). Implementation intentions and goal achievement: A meta-analysis of effects and processes. Advances in Experimental Social Psychology, 38, 69–120. http://dx.doi.org/10.1016/ S0065-2601 (06)38002–1 Gonzalez, J. S., Safren, S. A., Cagliero, E. et al. (2007). Depression, self-care, and medication adherence in Type 2 Diabetes: Relationships across the full range of symptom severity. Diabetes Care, 30, 2222–2227. https://doi.org/10.2337/dc07-0158 Greaves, C. J., Sheppard, K. E., Abraham, C. et al. (2011). Systematic review of reviews of intervention components associated with increased effectiveness in dietary and physical activity interventions. BMC Public Health, 11, 119. https://doi.org/10.1186/1471-2458-11-119 Hargreaves, T., Nye, M., & Burgess, J. (2010). Making energy visible: A qualitative field study of how householders interact with feedback from smart energy monitors. Energy Policy, 38, 6111–6119. https://doi.org/10.1016/ j.enpol.2010.05.068 Harkin, B., Webb, T. L., Chang, B. P. I. et al. (2016). Does monitoring goal progress promote goal attainment? A meta-analysis of the experimental evidence. Psychological Bulletin, 142, 198–229. https://doi.org/10.1037/bul0000025

Helsel, D. L., Jakicic, J. M., & Otto, A. D. (2007). Comparison of techniques for self-monitoring eating and exercise behaviors on weight loss in a correspondence-based intervention. Journal of the American Dietetic Association, 107, 1807–1810. https://doi.org/10.1016/j.jada.2007.07.014 Heneghan, C., Ward, A., Perera, R. et al. (2012). Selfmonitoring of oral anticoagulation: Systematic review and meta-analysis of individual patient data. The Lancet, 379. https://doi.org/10.1016/ S0140-6736(11)61294-4 Howansky, K., Dominick, J. K., & Cole, S. (2018). The look of success or failure: Biased self-perceptions serve as informational feedback during goal pursuit. Motivation Science. https://doi.org/ 10.1037/mot0000121 Huang, S.-C., Zhang, Y., & Broniarczyk, S. M. (2012). So near and yet so far: The mental representation of goal progress. Journal of Personality and Social Psychology, 103, 225–241. https://doi.org/ 10.1037/a0028443 Ickes, W., Dugosh, J. W., Simpson, J. A., & Wilson, C. L. (2003). Suspicious minds: The motive to acquire relationship-threatening information. Personal Relationships, 10, 131–148. https://doi .org/10.1111/1475-6811.00042 Jakicic, J. M., Davis, K. K., Rogers, R. J., et al. (2016). Effect of wearable technology combined with a lifestyle intervention on long-term weight loss: The idea randomized clinical trial. Journal of the American Medical Association, 316, 1161–1171. https://doi.org/10.1001/jama.2016.12858 Kanfer, F. H., & Karoly, P. (1972). Self-control: A behavioristic excursion into the lion’s den. Behavior Therapy, 3, 398–416. https://doi.org/ 10.1016/S0005-7894(72)80140-0 Kluger, A. N., & DeNisi, A. (1996). The effects of feedback interventions on performance: A historical review, a meta-analysis, and a preliminary feedback intervention theory. Psychological Bulletin, 119, 254–284. https://doi .org/10.1037/0033-2909.119.2.254 Kok, G., Gottlieb, N. H., Peters, G.-J. Y. et al. (2016). A taxonomy of behaviour change methods: An intervention mapping approach. Health Psychology Review, 10, 297–312. https://doi.org/ 10.1080/17437199.2015.1077155

Monitoring Interventions

Latham, G. P., & Locke, E. A. (1991). Selfregulation through goal setting. Organizational Behavior and Human Decision Processes, 50, 212–247. https://doi.org/10.1016/0749-5978 (91)90021-K Lewin, K. (1951). Field Theory in Social Science. New York: Harper. Louro, M. J., Pieters, R., & Zeelenberg, M. (2007). Dynamics of multiple-goal pursuit. Journal of Personality and Social Psychology, 93, 174–193. https://doi.org/10.1037/0022-3514.93.2.174 Michie, S., Abraham, C., Whittington, C., McAteer, J., & Gupta, S. (2009). Effective techniques in healthy eating and physical activity interventions: A meta-regression. Health Psychology, 28, 690–701. https://doi.org/10.1037/a0016136 Michie, S., Richardson, M., Johnston, M. et al. (2013). The behavior change technique taxonomy (v1) of 93 hierarchically clustered techniques: Building an international consensus for the reporting of behavior change interventions. Annals of Behavioral Medicine, 46, 81–95. https://doi.org/ 10.1007/s12160-013-9486-6 Miller, G. A., Galanter, E., & Pribram, K. H. (1960). Plans and the Structure of Behavior. New York: Holt, Rinehart, & Winston. https://doi.org/ 10.1037/10039-000 Moberly, N. J., & Watkins, E. R. (2010). Negative affect and ruminative self-focus during everyday goal pursuit. Cognition and Emotion, 24, 729–739. https://doi.org/ 10.1080/02699930802696849 Myrseth, K. O. R., & Fishbach, A. (2009). Self-control: A function of knowing when and how to exercise restraint. Current Directions in Psychological Science, 18, 247–252. https://doi.org/10.1111/ j.1467-8721.2009.01645.x Powers, W. T. (1973). Behavior: The Control of Perception. Chicago: Aldine. Reynolds, J. P., Webb, T. L., Benn, Y., Chang, B. P. I., & Sheeran, P. (2018). Feeling bad about progress does not lead people to want to change their health behaviour. Psychology and Health, 33, 275–291. https://doi.org/10.1080/08870446.2017.1310862 Richardson, C. R., Newton, T. L., Abraham, J. J., Sen, A., Jimbo, M., & Swartz, A. M. (2008). A meta-

553

analysis of pedometer-based walking interventions and weight loss. Annals of Family Medicine, 6, 69–77. https://doi.org/10.1370/ afm.761 Shaffer, J. A., Kronish, I. M., Falzon, L., Cheung, Y. K., & Davidson, K. W. (2018). N-of-1 randomized intervention trials in health psychology: A systematic review and methodology critique. Annals of Behavioral Medicine, 52, 731–742. https://doi.org/10.1093/abm/kax026. Tice, D. M., Bratslavsky, E., & Baumeister, R. F. (2001). Emotional distress regulation takes precedence over impulse control: If you feel bad, do it! Journal of Personality and Social Psychology, 80, 53–67. https://doi.org/10.1037/ 0022-3514.80.1.53 Wang, C., & Mukhopadhyay, A. (2012). The dynamics of goal revision: A Cybernetic Multiperiod TestOperate-Test-Adjust-Loop (TOTAL) model of self-regulation. Journal of Consumer Research, 38, 815–832. https://doi.org/10.1086/660853 Webb, T. L., Benn, Y., & Chang, B. P. I. (2014). Antecedents and consequences of monitoring domestic electricity consumption. Journal of Environmental Psychology, 40, 228–238. https:// doi.org/10.1016/j.jenvp.2014.07.001 Webb, T. L., Chang, B. P. I., & Benn, Y. (2013). “The ostrich problem”: Motivated avoidance or rejection of information about goal progress. Social and Personality Psychology Compass, 7, 794–807. https://doi.org/10.1111/ spc3.12071 Webber, J., Scheuermann, B., McCall, C., & Coleman, M. (1993). Research on selfmonitoring as a behavior management technique in special-education classrooms: A descriptive review. Remedial and Special Education, 14, 38–56. https://doi.org/10.1177/ 074193259301400206 WHO (World Health Organization). (2015). Sugars Intake for Adults and Children. Geneva: WHO. www.who.int/nutrition/publications/guidelines/ sugars_intake/en Wilhite, H., & Ling, R. (1995). Measured energy savings from a more informative energy bill. Energy and Buildings, 22, 145–155.

38 Goal Setting Interventions Tracy Epton and Christopher J. Armitage

Practical Summary This chapter describes the behavior change technique of goal setting. It is an established and popular technique that has been used successfully in multiple contexts, behaviors, and populations from making a personal goal to eat out only once a month to governmental goals of reducing unemployment figures. Different types of goals have been identified in academic theories of goal setting, such as behavioral goals (e.g., “in the next week, I will take a prepacked lunch to work on four weekdays”) and outcome goals (e.g., “in the next week, I will save £20”), and those in the public domain, such as SMART goals (i.e., a goal that incorporates best practice and is specific, measurable, achievable, relevant, and time-bound; e.g., “do my homework after class on Monday, Wednesday, and Friday”). The chapter focuses on goal setting theory (Locke & Latham, 1990), which is described in detail alongside evidence on how effective goal setting is, for whom goal setting works best, and under what circumstances. A step-by-step guide is included in Section 38.5 to help people set goals.

38.1 Introduction Goal setting is a behavior change technique that most people have experienced. For example, they have either set goals themselves or had goals set for them by other people such as in education (e.g., do two hours of revision per day; achieve an “A” on your next assignment), in their work life (e.g., achieve a promotion within two years; package 100 products per hour), or in their personal life (e.g., run a marathon in under four hours; clean your bedroom every two weeks). Although the first experimental study on goal setting was undertaken in 1935 Work on this chapter was supported by the NIHR Manchester Biomedical Research Centre and NIHR Greater Manchester Patient Safety Translational Research Centre. https://doi.org/10.1017/9781108677318.038

(Carson, Carson, & Heady, 1994), some theorists argue that goal setting is much older than that and is the basis of all behavior (e.g., Powers, 1973/2005, 2008; Powers, Clark, & McFarland, 1960a, 1960b). Goal setting is also used by global organizations such as the United Nations (UN), which has specified sustainable development goals (UN, 2015). The sustainable development goals cover seventeen different fields, including poverty, education, and health, and each is made up of several subgoals (the UN calls these “targets”). For example, in the health and well-being field there is the subgoal “By 2030, reduce the global maternal mortality ratio to less than 70 per 100,000 live births” and, in the no poverty field, one of the subgoals is “By 2030, eradicate extreme poverty

Goal Setting Interventions

for all people everywhere, measured as people living on less than $1.90 a day” (UN, 2015). Goal setting, therefore, is a well-used behavior change technique that has been applied to promote and modify behavior in multiple contexts and populations. It is also one with which many people have had firsthand experience. This chapter (1) defines goals and describes different types of goal; (2) discusses theoretical perspectives on goal setting; (3) explores how goal setting fits with the behavior change technique taxonomy version 1 (Michie et al., 2013); and (4) provides practical guidelines for how to set goals.

38.2 Definition and Types of Goal Setting A goal has been defined as “the object or aim of an action” (Locke & Latham, 2002, p. 705). However, because goals are ubiquitous, they are described in many different ways and thus there are many different types of goals that are in common usage, including the UN “targets” described in Section 38.1 and SMART goals that try to characterize a “successful goal” (see Table 38.1). Some of the important distinctions made in goal setting theory and research are described in the following sections.

38.2.1 Specific Goals vs. Aspirational Targets In the scientific literature, goals are defined as specific objectives or targets that are contrasted with general intentions (see Chapters 2 and 7, this volume), aims, or aspirational targets (Locke & Latham, 1990). In experimental studies, a goal setting condition is often compared with a control condition in which participants are asked to work toward an aspirational target (e.g., a “do your best at X” goal). The difference between specific goals and aspirational targets can be illustrated by the UN sustainable development goals – the seventeen goals can be considered more as aspirational targets than specific goals because they are wide-ranging in aims

555

and, as a consequence, goal achievement is difficult to evaluate (e.g., “zero hunger,” “quality education,” “take urgent action to combat climate change and its impacts”). In contrast, the subgoals or targets the UN has set within the sustainable development goals vary in specificity. A good example of a specific goal is from the health and well-being goal: “By 2030, reduce the global maternal mortality ratio to less than 70 per 100,000 live births,” which is definitively measurable. An example of an aspirational goal from the same sustainable development goal is to “strengthen the prevention and treatment of substance abuse, including narcotic drug abuse and harmful use of alcohol” – “strengthen” is not quantified and so it lacks specificity. Perhaps, not surprisingly, given its ubiquity, the concept of goal setting reaches beyond the realm of scientific research and into popular culture where there are numerous websites offering guidance as to how to set goals. One particularly popular approach is the “SMART” goal specification, which was devised by Doran (1981) and provides guidance as to how to make goals specific. Doran (1981, p. 36) clearly defines each part of the SMART goal acronym: “Specific” (“target a specific area for improvement”); “Measurable” (“quantify or at least suggest an indicator of progress”); “Assignable” (“specify who will do it”); “Realistic” (“state what results can realistically be achieved, given available resources”); and “Time-related” (“specify when result(s) can be achieved”). Other versions replace “Assignable” with “Attainable” and “Realistic” with “Relevant” (see Appendix 38.3, supplemental materials; Behavioral Medicine Lab, 2018).

38.2.2 Behavior vs. Outcome Goals Goal setting theory (Locke & Latham, 1990, 2019) differentiates between behaviors (an action that is performed) and outcomes (an outcome is the result of an action being performed). For example, the UN sustainable development goal to reduce inequalities sets goals related to behaviors, such as “Adopt policies, especially fiscal,

Table 38.1 Use of goal setting in everyday life Field

Source

Social inequalities

UN sustainable development goal #10 is to reduce inequality “By 2030, progressively achieve and sustain income growth of the within and among countries bottom 40 per cent of the population at a higher rate than the national average” (UN, 2015) The Duke of Edinburgh charity helps to prepare young people “Increase the number of young people starting a DofE programme per for life and for work year to 350,000 (18/19 figure 287,937), with 20% (70,000) from disadvantaged backgrounds” (Duke of Edinburgh, 2019) Public Health England – Stoptober Stop smoking for 28 days during the month of October (PHE, 2012) NHS England – GP appointment targets “Roll out evening and weekend GP appointments, to 50% of the public by March 2018 and 100% by March 2019” (NHS England, 2017) UK Government – eliminating the national deficit “To aim for a balanced budget by the middle of the next decade” (Conservatives, 2017) Runner’s World “Race under 40 minutes in a 10 k by September” (Runner’s World, 2003) Couch to 5 k An app with a goal to run 5 k within 9 weeks (NHS, 2017) Monday.com Set and track work goals for yourself and team (Monday.com, 2019) Employee performance reviews – recommends goal setting to “Use the review process as an opportunity to set attainable goals” improve problem areas (Business News Daily, 2019) UK Government – goals for school curriculum “We will expect every 11-year-old to know their time tables off by heart” (Conservatives, 2017) Study Goal app – allows you to set goals for study “Study Goals borrows ideas from fitness apps, allowing students to see their learning activity, set targets, record their own activity amongst other things. It also has a social side, allow students to share their activity and goals with their peers” (Study Goal, 2019) Athletes set goals “Jessica Ennis says her next heptathlon goal will be to break the 7,000point barrier after an incredible year” (Sky Sports, 2012) UK Government “To reduce absolute carbon emissions by 34% by 2020/21” (Parliament UK, 2019)

Charities

Health

Financial targets Fitness

Workplace

Education

Sport Environment

Example

Goal Setting Interventions

wage and social protection policies, and progressively achieve greater equality,” as well as outcomes, such as “By 2030, progressively achieve

557

and sustain income growth of the bottom 40 per cent of the population at a rate higher than the national average.”

Sidebar 38.1 Goal setting behavior change technique(s) from existing taxonomies

Goal setting features in existing taxonomies of behavior change. In the behavior change technique taxonomy version 1 (BCTTv1; Michie et al., 2013) and the CALO-RE taxonomy (Michie et al., 2011), a precursor to the BCTTv1, goal setting is broken down into two isolated techniques dependent on whether the goal is targeting behavior or outcomes. The intervention mapping taxonomy (Kok et al., 2016) does not make this distinction. There are several behavior change techniques that are related to goal setting; some require a goal to be set before the technique can be implemented (e.g., review of goals), others are closely related to goal setting as they are similar techniques (e.g., graded tasks), and others are related by theory (e.g., feedback). A summary of the techniques and associated descriptions from the taxonomies is provided in Table 38.2. Table 38.2 Behavior change techniques related to goal setting from existing taxonomies Taxonomy

Technique and Cluster (if any)

Primary/ Description (taken from Closely Related taxonomies)

BCTTv1 (Michie et al., 2013)

Cluster: 1. Goals & Primary Planning Technique: 1.1 Goal setting (behavior)

Technique: 1.3 Goal setting (outcome)

Cluster: 1 Goals & Planning Technique: 1.5. Review behavior goals

Primary

“Set or agree a goal defined in terms of the behavior to be achieved (e.g., set the goal of eating 5 pieces of fruit per day)” “Set or agree a goal defined in terms of a positive outcome (e.g., set a weight loss goal, such as 0.5 kg over one week, as an outcome of changed eating patterns)” “Review behavior goals jointly with the person and consider modifying the goals in light of achievement (e.g., examine how well someone’s performance corresponds to the goal, whether they consumed less than 1 unit of alcohol per day, and then modify up or down according to success or failure)” Continued

558

TRACY EPTON AND CHRISTOPHER J. ARMITAGE

Table 38.2 (cont.)

Taxonomy

Technique and Cluster (if any)

Primary/ Description (taken from Closely Related taxonomies)

“Review outcome goals jointly with the person and consider modifying the goals in light of achievement (e.g., examine how much weight has been lost and adjust the goals accordingly)” Technique: 1.6 “Draw attention to discrepancies Discrepancy between a person’s current between current behavior and the person’s pregoal and viously set goals (e.g., point out behavior that the recorded exercise fell short of the goal set)” Cluster: 1 Goals & Closely Related “Create a written specification of Planning the behavior to be performed, Technique: 1.8 agreed by the person, and witBehavioral nessed by another (e.g., sign a contract contract that they will not drink alcohol for a week)” Cluster: 8 Repetition “Set easy to perform tasks, mak& Substitution ing them increasingly difficult, Technique: 8.7 but achievable, until behavior is Graded tasks performed (e.g., ask the person to walk 100 yards a day, then half a mile a day once achieved 100 yards, then one mile a day once achieved half a mile)” Cluster: 1 Goals & Primary “Ask the person to affirm or Planning reaffirm statements indicating Technique: 1.9 commitment to change the Commitment behavior (e.g., ask the person to say ‘I will take my medication as prescribed’)” Cluster: 2 Feedback “Monitor and provide informa& Monitoring tive or evaluative feedback on Technique: 2.2 performance of the behavior (e. Feedback on g., inform the person how many behavior steps they walked each day)” Technique: 1.7 Review outcome goals

Continued

Goal Setting Interventions

559

Table 38.2 (cont.)

Taxonomy

Technique and Cluster (if any)

Primary/ Description (taken from Closely Related taxonomies)

“Monitor and provide feedback on the outcome of performance of the behavior (e.g., inform the person of how much weight they have lost)” Technique: 5. Goal Primary “The person is encouraged to setting behavior make a behavioral resolution (e.g., take more exercise next week)” Technique: 6. Goal “The person is encouraged to set setting outcome a general goal that can be achieved by behavioral means but is not defined in terms of behavior (e.g., to reduce blood pressure)” Technique: 10 Closely Related “Involves a review or analysis of Prompt review of the extent to which previously behavioral goals set behavioral goals were achieved” Technique: 11 “Involves a review or analysis of Prompt review of the extent to which previously outcome goals set outcome goals were achieved” Technique: 25 Closely Related “Must involve written agreement Agree behavioral on the performance of an contract explicitly specified behavior so that there is a written record of the person’s resolution witnessed by another” Technique: 9 Set “Breaking down the target graded tasks behavior into smaller easier to achieve tasks and enabling the person to build on small successes to achieve the behavior” Technique: 19 Closely Related “This involves providing the parProvide feedback ticipant with data about their on performance own recorded behavior” Technique 2.7 Feedback on outcomes of behavior

CALO-RE (Michie et al., 2011)

Continued

560

TRACY EPTON AND CHRISTOPHER J. ARMITAGE

Table 38.2 (cont.)

Taxonomy

Technique and Cluster (if any)

Intervention Technique: Goal Mapping setting Taxonomy (Kok et al., 2016) Technique: Set graded tasks Technique: Public commitment Technique: Feedback

Primary/ Description (taken from Closely Related taxonomies) “Prompting planning what the person will do, including a definition of goal-directed behaviors that result in the target behavior” Closely Related “Setting easy tasks and increase difficulty until target behavior is performed” Closely Related “Stimulating pledging, promising, Primary

or engaging oneself to perform the healthful behavior and announcing that decision to others” “Giving information to individuals and environmental agents regarding the extent to which they are accomplishing learning or performance, or the extent to which performance is having an impact”

Note. BCTTv1 = Michie et al.’s (2013) behavior change technique taxonomy version 1.

38.3 Theory and Mechanisms of Change In keeping with Powers and colleagues’ view that goal setting is the basis of all behavior (Powers, 1973/2005, 2008; Powers, Clark, & McFarland, 1960a, 1960b; see also Chapter 9, this volume), the different types of goals described in Sections 38.2.1 and 38.2.2 have largely evolved over time in applied settings and have only later been classified into different goal types. This section describes the main theory of goal setting and the mechanisms through which it predicts that goal setting will achieve its effects. We further compare goal setting theory with other theories that include goal setting.

38.3.1 Goal Setting Theory Goal setting theory (Locke & Latham, 1990, 2002, 2006, 2019) provides a good explanation of how to set goals optimally and can be contrasted with related theories such as goal systems theory (Kruglanski et al., 2002), which focuses on the processes leading up to the setting of a goal, and control theory (Powers, 1973/2005, 2008; Powers, Clark, & McFarland, 1960a, 1960b; see Chapter 9, this volume), which focuses on the processes to monitor goal progression. Goal setting theory (Locke & Latham, 1990, 2002, 2006, 2019) was built around a series of studies conducted in the field of organizational psychology that explored the use of goal setting in improving human

Goal Setting Interventions

performance (Locke & Latham, 2002). The evidence on which the theory was based showed that the most effective goals were specific and sufficiently difficult (Locke & Latham, 1990). As noted in Section 38.2.1, the most specific goals are those that are quantifiable (e.g., the health and well-being sustainable development goal: “By 2030, reduce the global maternal mortality ratio to less than 70 per 100,000 live births”). Nonquantifiable goals tend to be less effective than comparable quantifiable goals (Locke & Latham, 1990). However, even with nonquantifiable goals, there is some evidence that difficult ones are more effective than easier goals; for example, “go as fast as you can” would be more effective in achieving a fast pace over a set distance than an easier “go at a relaxed pace” (Locke & Latham, 1990). With regards to difficulty, goal setting theory states that there is a positive linear relationship between difficulty and performance (i.e., the more difficult a goal, the greater the improvement in performance) but only up to the limit of a person’s abilities, at which point improvements in performance level off (Locke & Latham, 1990). Goals that are too difficult may lead to people disengaging with the goal in order to avoid future failure or even as an excuse for not trying. Consistent with this point, a recent critical review suggested that specific difficult goals are not appropriate when an individual is learning a new and complex task as this can lead to stress that impairs performance and self-efficacy and can even result in cheating or misreporting of performance (Swann et al., 2019). Goal setting theory proposes that goals are most likely to be effective when (1) the person is committed to the goal, (2) feedback on the goal progress is provided, (3) the task is low in complexity (i.e., it involves a small number of acts and decisions), and (4) there are adequate resources and few situational constraints (Latham & Seijts, 2016; Locke & Latham, 1990, 2002, 2006, 2019). More recently, Wood et al. (2017) have suggested that strategy use (e.g., using other behavior change techniques) is also

561

a moderator of the effect of goals on performance. They suggest that people with high knowledge of strategies (e.g., are familiar with other behavior change techniques) are likely to perform better than people with low knowledge of additional strategies (Wood et al., 2017). Eberly et al. (2017) further propose a role for attribution theory in terms of understanding additional potential moderators of the relationship between goal setting and performance on the behaviors or actions required to achieve the goals. Attribution theory proposes that people analyze the events and feedback to which they are subjected in order to determine why it occurred (Heider, 1958). In other words, in response to an event, people assess whether its causes were internal (e.g., personality) or external (e.g., situation), as well as the degree of stability and controllability of those causes. Eberly et al. (2017) contend, therefore, that performance is dependent on people’s views of their progress as caused by internal or external factors and how stable and controllable individuals perceive these factors to be. Specifically, Eberly et al. (2017) suggest that lack of progress attributed to internal factors (e.g., the person lacks the capability) can lead to a reduction in their confidence or self-efficacy (i.e., a person’s perception of their capability) to perform the behavior and a consequent adjustment of the goal downwards (see also Chapter 3, this volume). Likewise, if lack of progress is seen as stable (e.g., they need to work full-time so cannot devote time to the goal), this can lead to goal adjustment or disengagement. Finally, if the cause of the lack of progress is seen as uncontrollable (e.g., they are too old to reach a sporting personal best), then this can also lead to disengagement. Locke et al. (1981) proposed four mechanisms through which goal setting works to improve performance. First, goals help to direct attention, that is, they assist people in paying attention to goal-relevant activities and directing attention away from those that are not relevant. Second, once attention is directed toward goal-relevant

562

TRACY EPTON AND CHRISTOPHER J. ARMITAGE

tasks, people exert effort to complete them. The amount of effort expended is regarded as being in proportion to the size of the goal, meaning that greater effort is put in for more difficult goals than less difficult goals. Third, related to directed attention and effort, goals motivate persistence so the effort is extended over longer time periods. Finally, goal setting leads to the person developing strategies, such as action plans, and problem solving to achieve their goal. These strategies may form the basis of other types of goal: process goals that involve using existing strategies (e.g., add vegetables to your dinner) and learning goals that focus on identifying new strategies (e.g., identify three ways you can improve your diet; Swann et al., 2019). Wood and colleagues (2017) elaborate on the idea of the effects of goal setting on performance being mediated by strategy development. They suggest that, when a goal is activated, then strategies that have been used before to achieve similar goals are activated in long-term memory. However, if the task is too complex or novel, then the goal itself, or fear of failing to attain it, will motivate people to create new task strategies to accomplish the goal (Wood et al., 2017). Wood et al. (2017) identify four categories of strategies, namely (1) task-specific strategies derived from knowledge (e.g., a strategy that had been used in the past or one that had been prescribed by the interventionist); (2) strategy development (e.g., amount of self-reported effort used in either refining task-specific strategies or developing new strategies, or reported use of imagination); (3) search and information processing strategies (e.g., measures of cognitive processes used such as hypothesis testing and critical thinking); and (4) self-regulatory strategies (e.g., planning, positive self-talk, not self-handicapping). Eberly et al. (2017) suggest that affect may also be a mediator of the goal–performance relationship, based on the general principle that, if positive affect is experienced, then the goal might be adjusted upward, whereas, if negative affect is experienced,

then goal may be adjusted downward. However, Eberly et al. (2017) propose a level of nuance to this pattern whereby lower effort toward a goal is associated with shame but increased effort is associated with guilt. Goal setting theory has attracted considerable attention in the research literature on behavior change and provides insight into why the different types of goals outlined in Sections 38.2.1 and 38.2.2 work in promoting behavioral engagement and goal pursuit. It also proposes mechanisms of action, although these have, by comparison, received lower levels of attention in the research literature. One possible limitation, however, is that goal setting theory views goals as conscious, singular endeavors with a finite end and without context such as the presence of competing goals. However, goal setting theory is focused solely on goal setting, whereas related theories such as goal systems theory (Kruglanski et al., 2002) and control theory (Powers, Clark, & McFarland, 1960a, 1960b; Powers, 1973/2005, 2008; see Chapter 9, this volume) address issues such as multiple complementary or competing goals with unspecified end points and mechanisms of action that may be nonconscious (see Chapters 12 and 15, this volume).

38.4 Evidence Base There is a considerable body of empirical evidence showing that goal setting works in changing behavior and there have been numerous reviews looking at various aspects of goal setting such as the effect of goal difficulty and specificity (Mento, Steel, & Karren, 1987), adding feedback to goal setting (Neubert, 1998), and setting group goals (Kleingeld, van Mierlo, & Arends, 2011) on behavior and outcomes. Recently, Locke and Latham (2017, 2019) illustrated the numerous fields in which goal setting has been successfully applied to change behavior and maximize goal pursuit, including in resource management, creativity, leadership, sports, negotiation, health

Goal Setting Interventions

behavior, entrepreneurship, education, and therapy. Furthermore, a recent meta-analysis (Epton, Currie, & Armitage, 2017) found that goal setting was efficacious in promoting behavior change in environmental (e.g., increasing recycling), health (e.g., eating more fruit and vegetables), sports (e.g., increasing archery performance), production (e.g., building models), education (e.g., completing math problems), cognition (e.g., spotting errors), and scheduling and appointment-keeping contexts. In this section, further details of the evidence base for goal setting interventions in changing behavior are provided.

38.4.1 Effectiveness of Specific, Difficult Goals Locke and Latham (1990) provided evidence from five meta-analyses that supported the premise that, for goal setting to be optimally effective, goals should be specific and difficult. Average effects of specific and difficult goals on behavioral performance compared to “do your best” or “no goals” varied between small-tomedium (d = 0.42; Cohen [1992], considers d = 0.20 “small,” 0.50 “medium,” 0.80 “large”) and large (d = 0.80). An up-to-date meta-analysis, which included studies that adopted randomized controlled trials, often considered the most “robust” study design to test intervention efficacy (see Chapter 21, this volume), found that setting specific goals compared to “no goals” or “do your best goals” resulted in small average effects on behavior change, d = 0.34 (Epton et al., 2017). The meta-analysis also confirmed that the average effect of difficult goals on behavior (d = 0.45) was larger than goals of moderate (d = 0.25) or easy (d = 0.25) difficulty.

38.4.2 Goal-Type Moderation Effects Neither goal setting theory nor the behavior change technique taxonomy version 1 (Michie et al., 2013) makes predictions about the

563

comparative effectiveness of behavior and outcome goals on change in performance (i.e., goal setting studies include measures that are either behavior and/or outcomes of behavior). Furthermore, Epton et al.’s (2017) meta-analysis found that goals that focused on behavior or outcome were equally effective in changing performance. However, the goals were more effective when they corresponded with the measure used; behavioral goals were more effective at increasing behavior than outcome goals and vice versa. For example, the behavioral goal of “eating five portions of fruit or vegetables per day” was more effective when the measure of success was number of portions of fruit and vegetables eaten than when the measure was weight loss. The goal of “lose one pound of weight per week” was more effective when the measure was weight loss than when the measure was the number of portions of fruit and vegetables eaten. This suggests that differences between setting behavioral goals relative to outcome goals may be semantic rather than of theoretical importance; however, no studies to date have directly compared behavior and outcome goals within a single experiment (cf. outcome vs. process mental simulations; Pham & Taylor, 1999; Chapter 33, this volume).

38.4.3 Theoretical Moderation Effects According to goal setting theory, commitment should be high for individuals to achieve their set goals (Locke & Latham, 1990). This implies that individuals need to find the goal worthy of pursuing. However, a recent meta-analysis found that studies that included a behavior change technique to increase goal commitment (e.g., asking participants to make affirming statements such as “I will save £20 per week” to increase commitment to the goal) alongside goal setting actually decreased the effectiveness of the goal on performance (d = 0.20) compared to “do your best” goals, no goal, or the same intervention minus goal setting and commitment (d = 0.34; Epton

564

TRACY EPTON AND CHRISTOPHER J. ARMITAGE

et al., 2017). This suggests that commitment techniques may be detrimental to behavior change and, in some circumstances, manipulating commitment could highlight a lack of commitment to the goal and backfire. For example, a person who was asked to set a goal to save money may make some token efforts to do this (e.g., buy fewer takeout coffees) but, when asked to think about their level of commitment to the goal or asked to affirm their commitment to saving money, they may start to think about other things they need to do to achieve the goal (e.g., stop going out with friends, unsubscribe from Netflix) and realize that they do not want to put a great deal of effort into the goal and therefore disengage from it. There is inconclusive evidence that providing feedback (i.e., telling people their current standing or progress toward a goal) in conjunction with goal setting improves performance. Metaanalyses have found that goal setting plus feedback interventions led to similar-sized effects on behavior change relative to a goal setting–only interventions (d = 0.63; Neubert, 1998; d = 0.56; Tubbs, 1986). Epton et al. (2017) found that studies that included interventions using goal setting with feedback were no more effective in changing behavior (d = 0.01) than studies on interventions using “do your best” goals, no goal, or the same intervention minus goal setting plus feedback. When meta-analyses have coded for the presence or absence of feedback, rather than only including studies that had manipulated feedback (Neubert, 1998; Tubbs, 1986), results were inconclusive with one finding no differences between feedback and no feedback (d = 0.57 vs. d = 0.54, respectively; Mento et al., 1987) and another finding strong differences between feedback and no feedback (d = 0.73 vs. d = 0.44, respectively; Chidester & Grigsby, 1984). Evidence of task complexity as a moderator of goal setting on behavioral performance is also inconclusive. Wood et al. (1987) found that increasing task complexity decreased the effect of goal setting on individual behavioral

performance, while research has demonstrated no effect of task complexity in the effectiveness of goal setting in changing behavior in groups (Kleingeld et al., 2011) and groups and individuals (Epton et al., 2017). However, later versions of goal setting theory suggest that task complexity interacts with ability in determining behavioral performance. Specifically, task complexity is only a moderator of the effects of goal setting on behavior when ability is exceeded (Locke & Latham, 2006). Finally, although other elements of goal setting theory have high face validity, including the prediction that a lack of resources and high situational constraints would hinder the success of goals, to date, there have been no tests of the effects of these variables. Future research is needed to systematically test whether or not these factors are effective in augmenting the effectiveness of goal setting interventions on behavior change.

38.4.4 Mediators Locke et al. (1981) provided empirical evidence of the effect of goal setting on the directed attention, effort, and persistence. However, the hypotheses in these early studies were not tested directly through manipulation of the variables, or by using statistical mediation analysis, and it seems that these mediators have not been studied in detail since (Wood et al., 2017). These seem like important avenues for future research in order to provide comprehensive tests of establishing how goal setting interventions “work” in changing behavior. There have been studies that have examined the effect of goals on directing attention so people can resist distractions from competing goals and stay focused on the goal of interest. For example, Shah and colleagues (2002) found that people who were asked to think about a goal to which they were committed could list fewer alternative goals than those who were asked to think about a goal to which they were less committed. Further experiments showed that the inhibitory process

Goal Setting Interventions

was automatic rather than a conscious choice. When primed with a personally important goal (i.e., subliminally shown a goal word, e.g., activities such as running that they wanted to attain in the next week), participants were slower to identify other less important goals (i.e., desirable activities they did not want to do in the next week) in a lexical decision task where they had to decide if a target word was an activity or not (Shah et al., 2002). Shah and colleagues also found that the inhibitory effect was more pronounced for alternative goals that could be substituted for the focal goal (e.g., running could substitute cycling in the goal of fitness) and that alternative goals that could facilitate performance of the focal goal (e.g., the goal of not drinking alcohol could facilitate studying more as the person will have more time to study) were less inhibited. These studies suggest that focusing on a goal, by, for example, setting a goal, directs attention away from alternative goals, that is, goals that are unrelated to the focal goal and those that are unnecessary (i.e., similar goals), and that this is done without conscious effort. Effort and persistence as mediators have face validity but, again, few studies have looked at their direct role in explaining goal setting intervention effects directly. Eberly et al. (2017) suggest that effort and persistence in achieving goals are linked with feedback. In one experiment, participants were required to multitask to perform three simultaneous online tasks each with a specific goal (i.e., track a point on-screen using a joystick, respond to gauges when the pointer moved a notch, and keep two fuel tanks topped up); over the course of a few blocks of tasks, they were provided with either increasingly positive or increasingly negative feedback on their task accuracy. Physiological measures showed that initially negative feedback increased effort, and participants reported expending more effort than those in the positive feedback group; however, over time, those who received negative

565

feedback showed a trend to disengage with the goal (Venables & Fairclough, 2009). This suggests that negative feedback may be beneficial in increasing effort to achieve goals in the short term but, in the long term, the absence of positive feedback can lead to a reduction of effort and disengagement of the goal. So, a person who is trying to improve their grades at university will work hard after receiving a couple of bad grades but, if they continue to get bad grades despite the perceived effort they have invested, they may subsequently commit less effort to their studies. Wood et al. (2017) conducted a review of the literature to determine the types of strategies used in goal setting studies. These strategies include the aforementioned potential mediators and additional strategies should those fail. When someone first adopts a goal, the automatic strategy of directing specific attention to the focal goal and inhibiting alternative goals is likely to be applied (e.g., an academic might decide that they want to get promoted at work so they find themselves focusing on their work a bit more and having fewer coffee breaks with their friends). In the event of the task being too complex for directed attention to succeed in completing the goal, increased effort and persistence are applied (e.g., it seems that the minor changes the academic made in paying more attention to work and spending less time with their work colleagues are not paying off so they decide to up the effort and put more effort into the work they do by reducing procrastination and working longer hours). If effort and persistence do not result in achievement of the goal, then other strategies are applied such as monitoring progress and developing new skills (e.g., the academic decided to ask their mentor about what they would do and then develop a new strategy: asking for advice and reprioritizing their work; so, after taking the advice from a colleague, the academic develops a new strategy of applying for research grants and writing research articles rather than other activities that were less valuable for promotion).

566

TRACY EPTON AND CHRISTOPHER J. ARMITAGE

Task-specific strategies used to attain goals were those that people already knew about and might have been used in the past (e.g., the academic tried to work longer hours) had a medium effect size (r = 0.302). The effort put into developing new strategies (e.g., the academic focusing their efforts on writing grant applications and papers) had a small-to-medium effect (r = 0.237). Self-regulatory strategies that focus on allocating effort (e.g., reducing procrastination to get more work done) had small-to-medium effects (r = 0.233). Finally, the strategy of searching and information processing (e.g., the academic asking for advice on strategies from their mentor) had a small effect (r = 0.095) on behavioral performance (Wood et al., 2017). This indicated that strategy use, especially tried-andtested strategies, was related to achieving goals and that spending time researching and developing new strategies and allocating effort were also effective in achieving goals. To summarize the evidence base, there is strong evidence that specific sufficiently difficult goals are effective for changing behavior and outcomes. Regarding other types of goals, there is burgeoning evidence that behavior goals are most effective at changing behavior and outcome goals are most effective at changing outcomes. There is also burgeoning evidence that commitment may be detrimental to goal setting in some circumstances. There is inconclusive evidence that feedback enhances goals and that task complexity affects goal performance. There is no research on the effect of lack of resources and high situational constraints on goal performance. There is some evidence that the proposed mediators of goals (i.e., directing attention, increased effort, increased persistence, and strategy use) are the mechanisms responsible for behavior change after goal setting.

38.5 Step-by-Step Guide There are nine key elements to consider when developing interventions in which people are encouraged to set goals (see Table 38.3 for a

summary): means of delivery, target audience and behaviors, enabling and inhibiting factors, training and skills required, intensiveness, evaluation of fidelity, evaluation of effectiveness, typical materials needed, and typical examples of implementation. This section outlines the key elements of goal setting interventions and sets out how goal setting can be used to effectively change behavior based on the currently available evidence.

38.5.1 Typical Means of Delivery Goal setting as a behavior change technique does not require a trained practitioner for delivery. Goals can be set with minimal instructions that can be delivered by print or verbally so that the individual can “self-direct” their goal setting (see Knittle et al., 2019). However, there is some evidence that setting goals in interventions delivered face-to-face by a practitioner is the most effective mode of goal setting. For example, a recent meta-analysis demonstrated that there were no differences in the effectiveness of goals on changing behavior and outcomes dependent on the mode of delivery (e.g., goal presented in writing or verbally) but goals that were set faceto-face with someone (e.g., whereby someone verbally presents a goal or hands someone a sheet or leaflet with a written goal on it) were more effective than goals that were not set faceto-face (e.g., online; Epton et al., 2017). Goals are more typically set by or for individuals but could be set by or for groups – there is evidence that group goals (i.e., one goal set for a group of people rather than each person being given a goal) are more effective than individual goals (Epton et al., 2017). Therefore, where possible, goals should be set for groups rather than individuals and delivered face-to-face rather than remotely. Personal preference can be used to determine whether the person whose behavior is to be changed, participatively or by another person, sets goals.

Goal Setting Interventions

567

Table 38.3 The dos, don’ts, and personal preferences for goal setting interventions Content

Intensiveness

Delivery & format

Materials needed

Do Make the goal specific Make the goal difficult but not improbable Consider pairing with monitoring of behavior by others without feedback Make the goal measurable Include a time frame of when to complete the goal Don’t Pair with commitment Personal preference Target behavior or outcome Target current standing or external standard End point of goal distal or proximal Personal preference Number of different goals made Repetition of goals Do Set the goal publicly Set the goal as a group Personal preference Written or verbal Self-set, other-set, or collaboratively set Goal setting is best done with materials to facilitate SMART goal setting (see supplementary materials developed by the Behavioral Medicine Lab, 2018)

38.5.2 Target Audience and Behaviors Goal setting works for many different behaviors and across multiple populations. They can be delivered to individuals or to groups and there is evidence that group goals are particularly effective (Epton et al., 2017). There is evidence that goal setting is particularly efficacious in younger people, males, and people of Asian ethnicity (Epton et al., 2017). Comparing studies that have used different populations has shown that goal setting was also particularly effective in children and the general population and least effective in university students (Epton et al., 2017). It is not clear, from current research, why goal setting is most effective in young people, males, people of Asian ethnicity, and general populations. However, research may show that goal setting is least effective in university students as this population are often sampled

in laboratory studies so the goals set may lack relevance in those situations; this indicates that it is important to ensure goals are personally relevant to the person targeted.

37.5.3 Enabling or Inhibiting Factors Research has suggested that goal setting enhances the effectiveness of the behavior change technique of monitoring of the target behavior or outcome by others without feedback so would be a good addition to those interventions (Epton et al., 2017). For example, rather than a supervisor surreptitiously monitoring the number of products a sales assistant has sold, it would also be useful for the employee to set a goal. A recent meta-analysis found that adding commitment to goal setting could inhibit the effectiveness of the intervention when compared with goal setting alone (Epton et

568

TRACY EPTON AND CHRISTOPHER J. ARMITAGE

al., 2017). As the review also found that merely measuring commitment also reduced the effectiveness of goal setting, this could indicate that, especially in people who are not committed, highlighting commitment could backfire by stopping people making efforts toward achievement of the goal and reducing their goal engagement further. There is insufficient evidence to make robust conclusions but the additional techniques of feedback, behavioral contract, and review of goals may not enhance the effectiveness of goal setting so, if time constraints are an issue, goal setting is an effective standalone behavior change technique (Epton et al., 2017). Regarding the setting in which the goal setting intervention is delivered, interventions conducted in schools and workplaces have been found to be more effective than those conducted in universities (Epton et al., 2017). This may be due to the fact that goals set by schoolchildren and workers are typically directly relevant to their current situation, whereas goals set by university students in laboratory studies tend to be more esoteric. In addition, setting a goal publicly (e.g., telling someone about a goal) is more effective than a privately set goal that might more easily be ignored – this could be partly explained by the fact that group goals, which have been shown to be particularly effective, are also public goals.

Appendix 38.1 (supplemental materials) for examples of SMART and “non-SMART” goals.

38.5.5 Intensiveness The number of goals set or the number of times the same goal is repeated has been shown to have no effect on the effectiveness of goals (Epton et al., 2017). However, it would be sensible to ensure that only a few goals are set to keep the burden on participants or clients low. In addition, it may be useful to provide means for individuals to be able to recite and recall accurately the key features of their set goals – for example, in terms of the SMART acronym. See Appendix 38.2 (supplemental materials) for a practical worksheet on how to effectively set SMART goals.

38.5.6 Evaluation of Fidelity It is possible to gauge the quality of a goal set more effectively if it is written down. Specifically, this enables the goal to be assessed in terms of its adherence to the components of Dorman’s SMART acronym such as specificity, difficulty, measurability, and time-relatedness. Measures such as the length of time spent setting a goal are unlikely to be a good indicator of effectiveness, as goal setting is an elegant behavior change technique to administer.

38.5.4 Training and Skills Required From the research literature on goal setting, it is clear that minimal instructions that describe a specific goal (i.e., “try to quantify the outcome or behavior that you want to achieve”) and difficult goal (i.e., “is more than you would normally expect to achieve in terms of outcome or behavior”) are sufficient to change people’s behavior. To further increase the effectiveness of the goal, instructions should be provided to the person who is setting the goal to ensure that the goal is measurable (e.g., “eat five portions of fruit and vegetables per day”) and that a time frame is set (e.g., “each week for the next fourteen weeks”). See

38.5.7 Evaluation of Effectiveness Effectiveness of goal setting can be measured by directly measuring the behavior or outcome specified in the goal. It is more difficult to assess the mechanisms of goal setting because the mechanism is unlikely to be reported during specification of the goal. Self-report scales could be used to measure the proposed mediators of goal setting (i.e., attention, effort, persistence, and strategy use); however, these mechanisms have not been subject to extensive research and there are no agreed means by which to measure these constructs.

Goal Setting Interventions

38.5.8 Typical Materials Needed Goals can be set without using extensive materials; however, to set the most effective goals it is useful to use a worksheet to make SMART goals. See Appendixes 38.2 and 38.3 (supplemental materials) for practical resources on how to set SMART goals. Goal setting could be done verbally, asking people to think of a goal. However, writing down a goal is good practice as it can be assessed for fidelity and, if it was shared with someone else (e.g., a practitioner), it would enhance the success due to making it public (Epton et al., 2017).

38.5.9 Typical Examples of Implementation Goals are set in many different contexts by many different kinds of people and organizations (see Sidebar 38.1 for examples). There is no one typical way of setting a goal. However, it would be useful to know the demographics of the sample as goal setting is less effective among women than men, older rather than younger participants, and nonAsian ethnicities. Particular care may be needed to ensure that the goals made are relevant for all populations but especially those where goal setting is less effective (e.g., women, older participants, university students, people of non-Asian ethnicity). It is important to make the goal a SMART goal (specific, measurable, achievable, relevant, timely). See Appendix 38.4 (supplemental materials) for an example of how goal setting techniques have been embedded in an intervention.

38.6 Summary and Conclusion Goal setting has been researched over several decades. In this time, research and practice on goal setting have established that specific and optimally challenging goals are the most effective type of goals. Despite a considerable body of evidence on goal setting, there are still many avenues for research. For example, it would be valuable to

569

establish the mechanisms of action of setting a goal. This would involve assigning participants to either a goal setting or a control condition and measuring, observing, or manipulating each of the four proposed mediators from goal setting theory (i.e., directed attention, effort, persistence, and other strategy use) and mediators from other theories (e.g., affect). A second under-researched area is exploring the distinction between behavior and outcomes in goal setting. To do this, an experiment could assign people to one of three groups – no goal, a behavior goal, or an outcome goal in a related area (e.g., avoiding junk food and losing weight) – to determine if there are differences in the effectiveness in behavior and outcome goals. A potentially valuable addition to this would be not only to measure behavior in terms of instrumental acts (e.g., walk 10,000 steps per day) but also to include compound behaviors (e.g., increase physical activity) and outcomes (e.g., lose weight). From a more practical perspective, it would be valuable to establish that the effects of goal setting persist beyond twelve months, what happens when goal setting is paired systematically with other behavior change techniques, and the effect of goal setting in a range of fields (e.g., primary care) and samples (e.g., people with low socioeconomic status).

References Behavioral Medicine Lab. (2018). Health promotion practitioners: Infographic resources. https://onli neacademiccommunity.uvic.ca/mpac/resources2/infographic-resources/ Business News Daily (2019). 6 tips for writing an effective performance review. Business News Daily. www.businessnewsdaily.com/5760-writegood-performance-review.html Carson, P. P., Carson, K. D., & Heady, R. B. (1994). Cecil Alec Mace: The man who discovered goalsetting. International Journal of Public Administration, 17, 1679–1708. https://doi.org/ 10.1080/01900699408524960 Chidester, T. R., & Grigsby, C. (1984). A meta-analysis of the goal setting performance literature. Proceedings

570

TRACY EPTON AND CHRISTOPHER J. ARMITAGE

of the Academy of Management, 1, 202–206. https:// doi.org/10.5465/AMBPP.1984.4978779 Cohen, J. (1992). A power primer. Psychological Bulletin, 112, 155–159. https://doi.org/10.1037/ 0033-2909.112.1.155 Conservatives. (2017). Conservative party manifesto 2017. https://s3.eu-west-2.amazonaws.com/con servative-party-manifestos/Forward+Together +-+Our+Plan+for+a+Stronger+Britain+and+a +More+Prosperous….pdf Doran, G. T. (1981). There’s a S.M.A.R.T. way to write management’s goals and objectives. Management Review, 70, 35–36. Duke of Edinburgh. (2019). Our strategic objectives. Duke of Edinburgh. Website. https://www.dofe .org/about/ Eberly, M. B., Liu, D., Mitchell, T. R., & Lee, T. W. (2017). Attributions and emotions as mediators and/or moderators in the goal striving process. In E. A. Locke & G. P. Latham (Eds.), New Developments in Goal Setting and Task Performance (pp. 35–50). Abingdon: Routledge. Epton, T., Currie, S. & Armitage, C. J. (2017). Unique effects of setting goals on behavior change: Systematic review and meta-analysis. Journal of Consulting and Clinical Psychology, 85, 1182– 1198. https://doi.org/10.1037/ccp0000260 Heider, F. (1958). The Psychology of Interpersonal Relations. Hillsdale, NJ: Lawrence Erlbaum Associates. Kleingeld, A., van Mierlo, H., & Arends, L. (2011). The effect of goal setting on group performance: A meta-analysis. Journal of Applied Psychology, 96, 1289–1304. https://doi.org/10.1037/ a0024315 Knittle, K., Heino, M. T. J., Marques, M. et al. (2020). The compendium of self-enactable techniques to change and self-manage motivation and behaviour v1. 0 Nature Human Behavior. https://doi.org/ 10.1038/s41562-019-0798-9 Kok, G., Gottlieb, N. H., Peters, G.-J. Y. et al. (2016). A taxonomy of behavior change methods: An intervention mapping approach. Health Psychology Review, 10, 297–312. https://doi.org/ 10.1080/17437199.2015.1077155 Kruglanski, A. W., Shah, J. Y., Fishbach, A., Friedman, R., Chun, W. Y., & Sleeth-Keppler, D. (2002). A theory of goal systems. Advances in Experimental

Social Psychology, 34, 331–378. https://doi.org/ 10.1016/S0065-2601(02)80008–9 Latham, G. P., & Seijts, G. H. (2016). Distinguished scholar invited essay: Similarities and differences among performance, behavioral and learning goals. Journal of Leadership and Organizational Studies, 23, 225–233. https://doi.org/10.1177/ 1548051816641874 Locke, E. A., & Latham, G. P. (1990). A Theory of Goal Setting and Task Performance. Englewood Cliffs, NJ: Prentice Hall. Locke, E. A., & Latham, G. P. (2002). Building a practically useful theory of goal setting and task motivation: A 35-year odyssey. American Psychologist, 57, 705–717. https://doi.org/ 10.1037/0003-066X.57.9.705 Locke, E. A., & Latham, G. P. (2006). New directions in goal setting theory. Current Directions in Psychological Science, 15, 265–268. https://doi .org/10.1111/j.1467-8721.2006.00449.x Locke, E. A., & Latham, G. P. (Eds.) (2017). New Developments in Goal setting and Task Performance. Abingdon: Routledge. Locke, E. A., & Latham, G. P. (2019). The development of goal setting theory: A half century retrospective. Motivation Science, 5, 93–105. https://doi.org/10.1037/mot0000127 Locke, E. A., Shaw, K. N., Saar, L. M., & Latham, G. P. (1981). Goal setting and task performance: 1969– 1981. Psychological Bulletin, 90, 125–152. https://doi.org/10.1037/0033-2909.90.1.125 Mento, A. J., Steel, R. P., & Karren, R. J. (1987). A meta-analytic study of the effects of goal setting performance 1966–1984. Organizational Behavior and Human Decision Processes, 39, 52–83. https://doi.org/10.1016/0749-5978(87) 90045–8 Michie, S., Ashford, S., Sniehotta, F. F., Dombrowski, S. U., Bishop, A., & French, D. P. (2011). A refined taxonomy of behaviour change techniques to help people change their physical activity and healthy eating behaviours: The CALO-RE taxonomy. Psychology and Health, 26, 1479– 1498. https://doi.org/10.1080/08870446.2010 .540664 Michie, S., Richardson, M., Johnston, M. et al. (2013). The behavior change technique taxonomy (v1) of 93 hierarchically clustered techniques: Building

Goal Setting Interventions

an international consensus for the reporting of behavior change interventions. Annals of Behavioral Medicine, 46, 81–95. https://doi.org/ 10.1007/s12160-013-9486-6 Monday.com (2019). A new way to manage your work. Monday.com. Website. https://monday.com/ Neubert, M. J. (1998). The value of feedback and goal setting over goal setting alone and potential moderators of this effect: A meta-analysis. Human Performance, 11, 321–335. https://doi.org/ 10.1207/s15327043hup11042 NHS (2017). Couch to 5k: week by week. NHS. Website. www.nhs.uk/live-well/exercise/couchto-5k-week-by-week/ NHS England (2017). Urgent and emergency care. NHS England. Website. www.england.nhs.uk/five-yearforward-view/next-steps-on-the-nhs-five-year-for ward-view/urgent-and-emergency-care/. Parliament UK (2019). Targets and performance. Parliament UK. Website. www.parliament .uk/about/sustainability/targets-andperformance/ Pham, L. B., & Taylor, S. E. (1999). From thought to action: Effects of process- versus outcomebased mental simulations on performance. Personality and Social Psychology Bulletin, 26, 250–260. https://doi.org/10.1177/01461672990 25002010 PHE (Public Health England). (2012). Stoptober. Public Health England. Website. https://cam paignresources.phe.gov.uk/resources/campaigns/ 6-stoptober/overview Powers, W. T. (1973/2005). Behavior: The Control of Perception. New York: Hawthorne. Powers, W. T. (2008). Living Control Systems III: The Fact of Control. Benchmark Publications. Powers, W. T., Clark, R. K., & McFarland, R. L. (1960a). A general feedback theory of human behavior: Part I. Perceptual and Motor Skills, 11, 71–88. https://doi.org/10.2466/pms.1960 .11.1.71 Powers, W. T., Clark, R. K., & McFarland, R. L. (1960b). A general feedback theory of human behavior: Part II. Perceptual and Motor Skills, 11, 309–323. https:// doi.org/10.2466/pms.1960.11.1.71

571

Runner’s World. (2003). Effective goal setting. www .runnersworld.com/advanced/a20819775/effec tive-goal-setting/ Shah, J. Y., Friedman, R., & Kruglanski, A. W. (2002). Forgetting all else: On the antecedents and consequences of goal shielding. Journal of Personality and Social Psychology, 83, 1261–1280. https://doi.org/10.1037/0022-3514.83.6.1261 Sky Sports. (2012). Jessica Ennis is setting new goals after her success at the London Olympics. Sky Sports, November 14. www.skysports.com/moresports/athletics/news/14935/8247229/jessicaennis-is-setting-new-goals-after-her-success-atthe-london-olympics Study Goal. (2019). Study Goal. App. https://itunes .apple.com/gb/app/study-goal/id1187626482?mt=8. Swann, C., Rosenbaum, S., Lawrence, A., Vella, S., McEwan, E., & Ekkekakis, P. (2019). Updating goal setting theory in physical activity promotion: A critical conceptual review. Health Psychology Review. Tubbs, M. E. (1986). A meta-analytical examination of the empirical evidence. Journal of Applied Psychology, 71, 474–483. https://doi.org/10.1037/ 0021-9010.71.3.474 UN (United Nations). (2015). Sustainable development goals. United Nations. Website. www.un.org/sus tainabledevelopment/sustainable-developmentgoals/ Venables, L., & Fairclough, S. H. (2009). The influence of performance feedback on goal setting and mental effort regulation. Motivation and Emotion, 33, 63–74. https://doi.org/10.1007/s11031-0089116-y Wood, R. E., Mento, A. J., & Locke, E. A. (1987). Task complexity as a moderator of goal effects: A meta-analysis. Journal of Applied Psychology, 72, 416–423. https://doi.org/10.1037/00219010.72.3.416 Wood, R. E., Whelan, J., Sojo, V., & Wong, M. (2017). Goals, goal orientations, strategies and performance. In E. A. Locke & G. P. Latham (Eds.), New Developments in Goal Setting and Task Performance (pp. 91–114). Abingdon: Routledge.

39 Planning and Implementation Intention Interventions Ryan E. Rhodes, Stina Grant, and Gert-Jan de Bruijn

Practical Summary This chapter looks at ways to change behavior that focus on planning. Different types of planning are defined and the evidence showing how to change behavior through planning is reviewed. Step-by-step guides are provided in relation to changing action plans (how, when, and where plans), preparatory plans (how plans leading to the action), implementation intentions (if-then plans), and coping plans (plans to overcome barriers). Worked examples for each type of plan are provided. Finally, research on when and how the application of these various planning techniques may be most or least successful is provided from the current evidence base in practice.

39.1 Introduction People attempt to incorporate many health, business, environmental, and prosocial behaviors into their daily and weekly lives to achieve desired outcomes. For example, new year’s resolutions are an annual ritual of setting standards for selfbetterment. People use these resolutions to pursue lofty goals such as exercising more, losing weight, eating healthier, learning a new skill or starting a new hobby, and spending more time enhancing

Ryan E. Rhodes was supported by a Kennedy Y. H. Wong Distinguished Visiting Professorship from Hong Kong Baptist University for the writing of this chapter. He is also supported by funds from the Canadian Cancer Society, the Social Sciences and Humanities Research Council of Canada, the Heart and Stroke Foundation of Canada, and the Canadian Institutes for Health Research. Gert-Jan de Bruijn was supported by grants from the Netherlands Heart Foundation, the Netherlands Organization for Health Research and Development, and the Diabetes Foundation. https://doi.org/10.1017/9781108677318.039

their personal well-being (The Telegraph, 2017). What is particularly interesting about this ritual, which is by no means limited to new year’s resolutions, is that the anticipated benefits are already acknowledged by the person who is engaging in the resolution-making process. That is, the outcome or behavior is already desired. This contrasts with motivational approaches (for overviews of these approaches, see the chapters in Part I of this handbook) that serve to educate a person about the potential outcomes of a given behavior in the hope of increasing that person’s motivation toward that behavior. Such motivational approaches include the reasoned action approach (see Chapter 2, this volume), social cognitive theory (see Chapter 3, this volume), and self-determination theory (see Chapter 8, this volume), as well as others detailed within this handbook. For example, in the case of physical exercise, 95 percent of adults acknowledge that they are already aware of the many health benefits of engaging in the behavior regularly (Martin et al., 2000;

Planning and Implementation Intention Interventions

O’Donovan & Shave, 2007). Given that health is a valued outcome, it is thus unsurprising that the vast majority of adults (approx. 80 percent) hold positive intentions to be regularly active (Rhodes & de Bruijn, 2013a) and explains why exercising more is the most common new year’s resolution (The Telegraph, 2017). What is interesting, however, is that only half of those with positive intentions actually follow through with engagement in exercise (Rhodes & de Bruijn, 2013a; Chapter 6, this volume). Further, most of those who do follow through are already active people and not the people who recently formed new intentions (Rhodes, 2015), such as resolutions. This general finding is not unique to physical activity, as it has been replicated in numerous health, educational, and prosocial behaviors (e.g., Armitage & Conner, 2001; Conner & Norman, 2015; McEachan et al., 2016; Sheeran, 2002; Webb & Sheeran, 2006). The modest relation between intentions and behavior, sometimes referred to as the intention-behavior “gap” (Orbell, 2004; Sheeran & Webb, 2016; see Chapter 6, this volume), represents a very real and important marker for intervention in many of the most soughtafter behaviors in daily life. It suggests that interventions that can help focus good intentions or initial aims to change one’s behavior, that can help organize and regulate motivation, or assist in preventing lapses in memory and focus attention are likely crucial to actual behavior change. In order to address this aspect of behavior change, several theories focus on action control, the translation of intentions into behavior (e.g., de Vries et al., 2005; Hagger & Chatzisarantis, 2014; Hall & Fong, 2007; Heckhausen & Gollwitzer, 1987; Kuhl, 1984; Rhodes, 2017; Schwarzer, 2008). These theoretical approaches all consider intention as a prerequisite for motivated action but presume that intention is a necessary but not always sufficient condition for actual behavior. While action control theories focus on several different constructs that may bridge the intention-

573

behavior gap, planning strategies are among the most common element across all of these theories (Rhodes & Yao, 2015). Thus, the purpose of this chapter is to provide an overview of a select group of strategies that fall under this broad category of planning (i.e., strategies to achieve a desired goal) (Hagger et al., 2016). The theory and context of various planning concepts and techniques are briefly overviewed (see Table 39.1), followed by a discussion with the instructions needed to formulate effective planning. These planning strategies can take different forms, such as prompts or instructions provided by a practitioner, or even as a written or online print communication, and thus resources are also provided that include these various forms (see Sidebar 39.1). As noted, theories that include planning consider intention (and thus some motivation) as a prerequisite, so the approaches in this chapter are considered most appropriate for people who want to achieve a behavior and are seeking strategies to assist in behavior change. Goal setting is also a particularly important strategy that is considered a prerequisite to planning (see Chapter 38, this volume). A goal is “the object or aim of an action … to attain a specific standard of proficiency, usually within a specified time limit” (Locke & Latham, 2002, p. 705), and goal setting is the process of agreement on a goal to be achieved that results in a target behavior. Setting goals serves four functions: (1) to direct attention and effort to the goal-based behavior and away from other activities; (2) to increase motivation toward the behavior specified in the goal; (3) to increase persistence to achieve the behavior; and (4) to prompt action leading to the development of self-regulation strategies to achieve the behavior (Locke & Latham, 2006). The planning strategies included in this chapter are particularly useful to achieve the latter function of goal setting and thus represent a logical next step in this sequence. Specifically, setting a behavioral or outcome goal is a prerequisite for a planning intervention.

Table 39.1 Planning concepts and their relationship with intervention taxonomies Taxonomy

Concept

Technique and Cluster (if any) Primary/Related

BCTTv1 (Michie et al., 2013)

Action Planning

Cluster: 1Goals & Planning Technique: 1.4 Action Planning Cluster: 1Goals & Planning Technique: 1.2 Problem Solving Cluster: 1 Goals & Planning Technique: 1.4 Action Planning

Coping Planning

Implementation Intentions

Intervention Mapping Action Planning Taxonomy (Kok et al., 2016) Coping Planning

Implementation Intentions

Implementation Intentions (Methods to Change Habitual, Automatic and Impulsive Behaviors) Planning Coping Responses (Methods to Change Habitual, Automatic and Impulsive Behaviors; Methods to Change Skills, Capability, and SelfEfficacy and to Overcome Barriers) Implementation Intentions (Methods to Change Habitual, Automatic and Impulsive Behaviors)

Description

Primary

Prompt detailed planning of performance of the behavior

Primary

Analyze, or prompt the person to analyze, factors influencing the behavior and generate or select strategies that include overcoming barriers and/or increasing facilitators Prompt detailed planning of performance of the behavior. Context may be environmental (physical or social) or internal (physical, emotional, or cognitive) (includes “Implementation Intentions”) Prompting planning what the person will do, including a definition of goal-directed behaviors that result in the target behavior

Primary

Related

Related

Getting the person to identify potential barriers and ways to overcome these

Primary

Prompting the making of if-then plans that link situational cues with responses that are effective in attaining goals or desired outcomes

Note. BCTTv1 = Michie et al.’s (2013) behavior change technique taxonomy version 1.

Planning and Implementation Intention Interventions

575

Sidebar 39.1 Approaches to planning

There are a number of planning approaches that have been used in behavior change interventions. Four prominent approaches are action planning, implementation intentions, preparatory planning, and coping planning. Action planning. Formulation of an articulated set of procedures to assist in behavioral action (Leventhal, Singer, & Jones, 1965). Preparatory planning. Formulation of plans that enhance the availability and accessibility of resources needed to obtain the goal (Bryan, Fisher, & Fisher, 2002). Implementation intentions. “If-then” action plans, whereby a link is forged between a cue and a subsequent behavioral response (Gollwitzer & Brandstatter, 1997; see also Chapter 6, this volume). Coping planning. Formulation of plans to overcome important barriers when action initiation and/or maintenance is challenged (Schwarzer, 2016).

39.2 Action and Preparatory Planning Intentions represent an individual’s motivation and the degree of effort they are prepared to invest in pursuing a given behavior to achieve a specific outcome or goal (Fishbein & Ajzen, 1975; Rhodes & Rebar, 2017). Intentions can therefore be thought of as “what” individuals pursue. By contrast, action planning involves the further specification of the intention encompassing the “when,” “where,” and “how” elements of the behavior (see Table 39.1; Hagger & Luszczynska, 2014; Schwarzer, 2008; Sniehotta, 2009; see also Chapter 7, this volume). Specifically, “when” represents the temporal elements of a plan (i.e., the time the behavior will be enacted), “where” represents the contextual aspects of the plan (i.e., the place or location in which it will be performed), and “how” represents the specific sub-actions needed to accomplish the goal (i.e., the types of action that need to be performed in order to fulfill the behavioral goal) (for further details, see Chapter 5). Although action planning components foster the translation from intentions to action, most behaviors require sub-actions to be completed prior to the execution of those action plans,

such as ensuring one has the appropriate equipment (e.g., desk, computer, books for study behavior, athletic apparel for exercise, etc.). Preparatory planning specifies the when and where components formulated to the specified preparatory action (e.g., placing the curb recycling schedule near the front door) required to complete the specific action (e.g., planning when and how to curbside recycle). The theory behind action planning can be traced to early work from Ach (1905) and Lewin (1951) but stems primarily from models that have sought to augment traditional social cognition theories that tend to focus on the deliberative processes of intention formation (Leventhal et al., 1965). These models suggest that mere intention is likely insufficient to enact complex behaviors and thus selfregulatory skills, such as action planning, are needed to augment intentions in either a mediation or a moderation capacity. They include, but are not limited to, the health action process approach (Schwarzer, 2008; see also Chapter 7, this volume), the integrated-change model (I-change model; (de Vries et al., 2005), the Motivational Volitional concept model (Gӧhner, Seelig, & Fuchs, 2009), the integrated behavior change model (Hagger & Chatzisarantis, 2014), and the

576

RYAN E. RHODES, STINA GRANT, AND GERT-JAN DE BRUIJN

multi-process action control framework (Rhodes, 2017; Rhodes & de Bruijn, 2013b; for a full discussion of integrated theoretical approaches to behavior change, see Chapter 15, this volume). Of key importance, action planning includes multiple cues to action for both simple (e.g., turning off a light switch to be power-smart) and complex (e.g., selecting, preparing, and eating nutritionally balanced meals for long-term health) behaviors (Hagger & Luszczynska, 2014). Thus, action planning is designed to impact behavior through mindful self-regulation (Bagozzi, 1992) and possibly through assisting in cognitive shortcuts that could assist in habit formation or automatic forms of action (de Bruijn et al., 2014; Fleig et al., 2013; Rhodes, 2017; also see Chapter 41, this volume). Furthermore, the technique of action planning is conceived as being an intuitive outcome for many people that naturally follows an intention (Carraro & Gaudreau, 2013; Schwarzer, 2008) but also a skill that can be acquired (Allan, Sniehotta, & Johnston, 2013) and thus a behavior change technique (Michie et al., 2013). Specifically, this means that action plans can be formed through prompts and exercises provided in written communications or by practitioners. It is this format of action planning that is the focus of interest in this chapter.

39.2.1 Evidence Base An appraisal of the efficacy of action planning is difficult at present because it has often been aggregated with implementation intentions (see Section 39.3 on implementation intentions) or as part of a multicomponent intervention. The best examinations of this technique come from interventions on health behaviors. For example, Carraro and Gaudreau (2013) showed a small effect of action plans on changes in physical activity behavior compared to control groups among nineteen eligible studies, suggesting the technique is effective in changing behavior above no treatment.

39.2.2 A Step-by-Step Guide The central premise of an action planning intervention is to induce individuals to develop a detailed and concrete plan of how to enact a behavior for which they have an intention to achieve (see Chapter 7, this volume). Delivery formats traditionally include either interviewer-assisted or nonassisted print or online communication, whereby individuals are asked to record their detailed plans. Typically, scripts outlining the concept of action planning are presented to the target audience, those whose behavior needs to change, from the outset (for scripts, see Appendix 39.1, supplemental materials). Individuals are then prompted to specify the details involved in carrying out their behavioral intention. In practice, this may be presented as requesting the individual to first note or record their behavioral intention and then detail the “when,” “where,” and “how” elements of how to carry out that desired behavior. This could be presented as a form requiring individuals to complete their goal and plan – such as completing sentences or filling in blank entries on a timetable or calendar (for examples, see Appendixes 39.1 and 39.2, supplemental materials). For example, in an action planning intervention to increase physical activity, a practitioner could first note the target behavior of achieving 150 minutes of moderate to vigorous physical activity per week. Next, the practitioner could ask participants to fill in the detailed plans of how to enact this behavior, namely by indicating when, where, and how they will accomplish the desired behavior. A suitable action plan would be: “On Monday, Wednesday, and Friday at 5.00 p.m. (when), I will go for a 50-minute jog each day (how) around my neighborhood (where)” (for a complete example of an action planning intervention, see Appendix 39.3, supplemental materials). Preparatory planning interventions follow a similar logic and format but differ in that they focus on the “when” and “where” of preparatory actions needed to achieve the target behavior

Planning and Implementation Intention Interventions

rather than the behavior itself. In particular, people could be asked to think of the simple sub-actions leading up to the target goal behavior and plan specifically for those sub-actions (see Appendix 39.4, supplemental materials). For instance, if one has formulated a goal to achieve 150 minutes of physical activity in a week, one may need to pack their workout gear at home in the morning in order to get their workout in after work. Alternatively, one may place this workout gear in a visible location at home, so that the moment they arrive at home they are immediately able to change into their running gear (for a full example of a preparatory action planning intervention, see Appendix 39.5, supplemental materials).

39.2.3 Implementation Considerations Despite their conceptual differences, the means of delivery for action and preparatory planning interventions are similar. In general, participants are requested to think of plans that are not too difficult to achieve and that they are certain of obtaining, consistent with some of the key principles of goal setting (see Chapter 38, this volume). Occasionally, participants are additionally requested to visualize enacting these plans. Plans are often written down using paper-andpencil formats or in website textboxes. This aids specific formulation of the plan, which is important for the quality and appropriateness of the plan and intervention fidelity. A meta-analysis of action planning interventions to promote physical activity indicated a number of conditions that enhance the effectiveness of action planning interventions (Carraro & Gaudreau, 2013). Specifically, action planning interventions produce stronger effects when people (1) already hold strong intentions; (2) are older; (3) were previously less active; (4) formulate plans that contain four components (as opposed to fewer components); and (5) are in rehabilitation

577

(Carraro & Gaudreau, 2013). The mode of delivery, such as solitary pen-and-paper or interviewer-assisted studies, or when plans are set by the study participants or by the researcher, does not seem to affect action planning effectiveness for physical activity. There is also evidence that preexisting planning skills make action planning more effective (Allan et al., 2013). For example, in a study on diet behaviors, Allan et al. (2013) demonstrated that participants who were more skilled in planning (as assessed by a planning task; Wilson et al., 1996) were more likely to follow up on their intention to log on to a website and complete an online diary on snack intake than those with poor planning skills. In contrast, research on physical activity indicates that planning seems to become less relevant once physical activity, or at least the decision to instigate physical activity, has become habitual and people feel more efficacious to engage in exercise behavior (de Bruijn et al., 2014). To date, there are relatively few experimental studies examining the proposed benefits of preparatory planning in changing behavior. Given the lack of experimental work on the added benefits of preparatory plans, caution is needed to understand the boundary limits of these instructions. However, there is preliminary evidence that these instructions are as effective as action planning in decreasing snack intake (de Bruijn et al., 2017).

39.3 Implementation Intentions Implementation intentions have considerable conceptual overlap with action plans and the two terms are often used interchangeably (see Table 39.1; Hagger & Luszczynska, 2014; Hagger et al., 2016; Michie et al., 2013; see also Chapter 6, this volume). Just like action plans, implementation intentions extend from intentions to specify the “when,” “where,” and “how” aspects of a plan (Gollwitzer & Brandstatter, 1997; Gollwitzer & Sheeran, 2006). Implementation intentions, however, are also termed “if-then” plans, whereby a

578

RYAN E. RHODES, STINA GRANT, AND GERT-JAN DE BRUIJN

link is forged between a cue and a subsequent behavioral response (Gollwitzer, 2014; Gollwitzer & Sheeran, 2006; Hagger & Luszczynska, 2014). Implementation intentions arose from the development of the model of action phases (Gollwitzer, 1990; Heckhausen & Gollwitzer, 1987). In the model, two separate “phases of action” are proposed: a motivational or “pre-decisional” phase and a subsequent volitional or “implementation” phase (see Chapters 6 and 7, this volume). Implementation intentions represent a key strategy within the volitional phase that is hypothesized to bind motivation to behavior (Gollwitzer, 1999). Given the distinction between motivation and volition, implementation intentions are particularly suitable for behaviors where the motivation to enact the desired behavior is stable, but inaction may occur due to forgetting to recall the intention, failing to seize an appropriate moment to act, or inadequate shielding from other more desirable courses of action (Gollwitzer & Sheeran, 2006). The theory behind implementation intentions suggests that the process of specifying contingency cues to a behavioral response helps build a link that is below conscious awareness (Gollwitzer & Sheeran, 2006). Thus, implementation intentions may be considered akin to the process of habit formation, without the prolonged period of behavioral repetition typically required to form a habit (Gollwitzer, 1999; Wood & Runger, 2016). This occurs by highlighting the appropriate cues aligned to the intended behavior and making the appropriate behavioral response more accessible when the cue is presented. Interestingly, research on using implementation intentions to counter an existing habit has shown that implementation intentions can also eliminate the advantage of an established habitual response and thus offer an individual flexibility to choose which behavior to perform in the situation as long as the motivation to perform a behavior is strong (Adriaanse, Gollwitzer et al., 2011). Thus, implementation intentions may be useful to break “bad

habits” or undesired behaviors and replace these with the desired behavior, by providing more flexible choice options between the healthy and unhealthy options. Hagger and Luszczynska (2014) make a critical point that implementation intentions are focused on developing a link between a cue and a highly specific behavioral response (e.g., “when I enter the grocery store, I will go to the vegetable aisle”), instead of a general behavioral response that may involve a complex sequence of actions (“regular healthy eating”). Research in complex health behaviors has tended to disregard this important conceptual aspect (de Bruijn et al., 2017; Sniehotta, 2009) and contemporary research on the psychology of habits has made the same distinction (see Gardner, Phillips, & Judah, 2016; Rhodes & Rebar, 2018). These conceptual boundaries of implementation intentions also represent how they can vary from action planning, because action planning can be both a conscious deliberative process and a nonconscious process and apply to both complex behaviors and simple behaviors (Hagger & Luszczynska, 2014).

39.3.1 Evidence Base Several reviews and meta-analyses have explored the effectiveness of implementation intentions in changing behavior for multiple behaviors (see also Chapter 6, this volume). For example, Gollwitzer and Sheeran (2006) conducted a meta-analysis of ninety-four studies across diverse behaviors and found a medium-sized effect of implementation intentions on behavior. Meta-analyses of implementation intention effects on behavior change in certain health behaviors such as healthy eating (Adriaanse, Vinkers et al., 2011; Vilà, Carrero, & Redondo, 2017) and physical activity (Bélanger-Gravel, Godin, & Amireault, 2013) have also shown mediumsized and small-sized effects from implementation intentions, respectively.

Planning and Implementation Intention Interventions

39.3.2 Step-by-Step Guide The objective of an implementation intention intervention is to have an individual identify a critical situation or cue and link it with the enactment of the appropriate behavior. This process facilitates efficient recall between the cue and the desired action. In most applications, interventions utilizing implementation intentions explicitly name the target behavior, explain why using the technique can be effective, provide an illustration of an implementation intention, and have participants work through examples in the “if-then” format. There are two main approaches to developing implementation intentions, namely self-generated and prespecified. In the self-generated form, implementation intentions are specific to the individual, and it follows that individuals generate their own personal implementation intentions in the “if-then” format (for scripts, see Appendix 39.6, supplemental materials). Conversely, in the prespecified style, participants are presented with a list of “if-then” statements by which they select the instances that apply to them and choose a correspondingly appropriate behavioral response for each instance. This prespecified approach arose from the development of volitional help sheets, which are a standard tool designed to facilitate the formation of personalized implementation intentions by providing a list of critical situations that could be encountered and potentially useful responses for behavior change (Armitage, 2008; for volitional help sheets, see Appendix 39.7, supplemental materials). For instance, in a self-generated implementation intention intervention aimed at promoting the use of reusable bags to reduce waste caused by single-use plastic bags when visiting the grocery store, the following might be adopted (adapted from Chapman & Armitage, 2010; see Appendix 39.8, supplemental materials): We would like you to begin bringing your reusable bags to the grocery store. Research has shown that if we identify a situation, then decide what to do in that situation, we may be more likely to turn our

579

intentions into actions. For example, statements such as the following can be useful: “If I am walking out the door to the grocery store then I will grab a reusable bag.” Please write your plans to use your reusable bags in the space provided, following the format in the previous example (“if … then …”).

In contrast, an implementation intention intervention aimed at promoting the use of reusable bags in the prespecified style would present a table with two columns. The first column would contain situations or temptations and the second column would contain responses or solutions. Next, individuals would be instructed to link situations with appropriate responses. For example, one may link a prespecified situation from the first column, such as “If I am walking out the door to the grocery store,” to an appropriate response from the second column, such as “then I will ensure I have a reusable bag with me.”

39.3.3 Implementation Considerations Although theory would suggest that implementation intentions are effective for all populations and behaviors, there is evidence that the effectiveness of these strategies varies across behaviors and individuals (Gollwitzer & Sheeran, 2006). For instance, people who have a strong desire to (re)act impulsively tend to benefit less from implementation intentions (Churchill & Jessop, 2010). Moreover, evidence suggests that implementation intentions tend to produce stronger effects on behavior when that behavior concerns the uptake of a behavior (e.g., increasing fruit and vegetable intake) than when it concerns reducing an existing habitual behavior (e.g., limiting snack intake) (Adriaanse, Vinkers, et al., 2011). Furthermore, effects of implementation intentions also tend to be more pronounced when higher-quality measures are used to assess behavior and there is a shorter time period over which behavior is measured. Finally, implementation intention instructions have stronger effects on behavior when people

580

RYAN E. RHODES, STINA GRANT, AND GERT-JAN DE BRUIJN

already have strong preexisting planning skills (Allan et al., 2013), when implementation intention instructions are combined with self-affirmation manipulations (Harris et al., 2014), and when implementation intentions are more specific (van Osch et al., 2010) and mentally imagined (Knäuper et al., 2011). In summary, although implementation intentions have been found to be effective for behavior change and that these effects are in the small-tomedium effect size range, practitioners employing implementation intention instructions should be aware of specific caveats, such as (1) that specific individuals may be more or less responsive to implementation intention instructions; (2) effects may vary for specific behaviors and may diminish over longer time periods; and (3) some individuals may require additional instructions, such as mentally imagining the sequence of acts.

39.4 Coping Planning Coping planning refers to plans developed to account for contingencies when barriers may arise that impede the performance of a behavior (see Table 39.1; Sniehotta et al., 2005; Chapter 7, this volume). They are ostensibly “back-up” plans to an action plan in order to help increase the chances of success of the behavioral performance in the face of expected barriers and difficulties (Napolitano & Freund, 2016; Schwarzer, 2016). Schwarzer (2016) describes coping plans as a variation of the initial action plan in terms of the target behavior, the time, the social situation, and other situational factors. Thus, coping plans may complement an action plan by having the planner consider the potential barriers that could thwart the “when,” “where,” and “how” aspects and provide alternatives to ensure that the action plan is implemented. Coping plans were developed initially for a cardiac rehabilitation study in which the authors sought to conceptualize different aspects of planning (Sniehotta et al., 2005), and the strategy has

gained traction in the scientific literature since that time (for an overview, see Schwarzer, 2016). They feature prominently in the health action process approach (Schwarzer, 2008; Chapter 7, this volume) but coping planning is also sometimes included in variants of the theory of planned behavior (Ajzen, 1991) and the multiprocess action control framework (Rhodes, 2017; Rhodes & de Bruijn, 2013b). Further, coping planning shares considerable overlap with overcoming performance barriers, which is a hallmark strategy in other social cognitive approaches such as the health belief model (Rosenstock, 1974; see also Chapter 4, this volume), the transtheoretical model (Prochaska & DiClemente, 1982; see also Chapter 10, this volume), and social cognitive theory (Bandura, 1986; see also Chapter 3, this volume). Coping planning is also linked to action planning because one must have an initial plan to use as a template in order to develop alternative plans. Thus, coping plans are more elaborate action plans (Hagger & Luszczynska, 2014).

39.4.1 Evidence Base Like action planning, evidence for the efficacy of coping planning is relatively limited at present with most of the work focused on health behaviors, given the physical activity and health origins of the technique (Sniehotta et al., 2005). Kwasnicka et al. (2013) conducted a systematic review of coping planning applied to health behaviors that identified eleven studies testing coping planning effects in interventions across multiple populations and behaviors. The authors concluded that coping planning was an effective technique to increase health behaviors and particularly effective in augmenting action planning. However, not all studies found intervention effects – this included three large sample studies with strong experimental designs. Carraro and Gaudreau’s (2013) meta-analysis of coping planning in physical activity studies identified a small effect on behavior change when compared to

Planning and Implementation Intention Interventions

no-treatment controls and a medium-sized effect when coping planning was combined with action planning.

39.4.2 Step-by-Step Guide A common approach to a coping planning intervention includes four basic steps: (1) delivering a message about coping planning and its utility; (2) providing an example of a coping plan; (3) prompting participants to consider potential barriers that may impede their performance of a target behavior; and, finally, (4) asking participants to develop a coping plan to overcome each anticipated barrier or difficulty (for coping planning interventions employed across various health behaviors, see Appendix 39.9, supplemental materials). For example, an intervention to promote studying behavior to improve academic achievement in students would initially provide students with a rationale and description of coping plans and their utility: “Coping plans are reserve plans that can be used to increase the chances of studying in the face of expected difficulties. They require a clear anticipation of barriers that may interfere with your studying behavior. By identifying these potential barriers that could impede your studying, you can develop strategies to overcome these obstacles.” This introduction might be followed by an example of a coping plan: “For example, if you identified friends asking you to hang out together as a barrier to your studying, a possible coping plan might look like this: ‘If my friends ask to get together, I will say no, I have to study and head to straight to the library.’” Next, students would be prompted to identify the salient barriers: “Which obstacles or difficulties might occur that would interfere with your studying behavior? Please list them.” The final step would be to prompt students to form strategies to overcome each anticipated barrier: “Think of one strategy to overcome each potential barrier to studying and list them” (for a complete example intervention, see Appendix 39.10, supplemental materials).

581

39.4.3 Implementation Considerations Kwasnicka and colleagues’ review of the effects of coping plans suggested that coping plans were effective in changing behaviors, particularly when used in conjunction with action plans and strategies to enhance self-efficacy (Kwasnicka et al., 2013; Chapter 32, this volume). Most studies provided participants with a list of preidentified barriers in the context of hypothetical scenarios and required participants to self-select personally relevant barriers from that list (see the examples in Appendix 39.10, supplemental materials). Such a strategy may be workable from a practical perspective, as most of the barriers can be collapsed into tempting situations (e.g., in the evening at home, watching television on the couch) and their precursors (e.g., fatigue after a long day at work) and physical boundaries (e.g., being stuck in traffic so one arrives at the gym when it has closed). Furthermore, most of the identified studies reported that coping planning instructions were filled in using a pen and paper under the supervision of a health practitioner, while only a small part were done over the internet or telephone. The benefit of monitoring how and which coping plans are formulated in a pen-andpaper setting is that these can then be checked for adequate formulation and adapted when needed (Kwasnicka et al., 2013). This is relevant for practitioners, as the majority of pen-and-paper interventions (five out of six) were more effective as compared to control conditions. Likewise, coping planning interventions supplemented with action planning interventions also appear to be more effective when they are conducted using pen-and-paper designs. Finally, there is presently limited information on effective moderators of coping planning instructions. In summary, practitioners should consider employing coping planning interventions when they have the opportunity to monitor and adapt the coping planning instructions that their clients

582

RYAN E. RHODES, STINA GRANT, AND GERT-JAN DE BRUIJN

are formulating. The evidence to date does not confirm for which individuals and behaviors coping planning instructions are more effective. There is also little evidence of the time periods over which coping planning instructions are most effective. However, the evidence to date does suggest that supplementing coping planning instructions with action planning instructions enhances the likelihood for behavioral change.

39.5 Conclusions Planning and implementation intention interventions have shown considerable promise in fostering behavior change in many contexts (e.g., prosocial, business, academic, and health). This chapter has detailed three types of planning interventions (action and preparatory planning, implementation intentions, and coping planning), outlined the theoretical and evidence base for how they work, and provided evidence-based guides on how these should be applied in behavior change interventions in practice. These planning instructions relate to the preparation of action, the identification of important cues to action, and shielding action initiation and maintenance for important barriers. The application of these three strategies in that specific order allows practitioners to develop and implement successful behavior change interventions. Practitioners should also consider important elements that may either enhance or decrease the effectiveness of these interventions, such as the type of behavior they wish to promote, the time period over which they want to see behavioral change, and the psychological makeup of their clients.

References Ach, N. (1905). Uber die Willenstatigkeit und das Denken. Göttingen: Vandenhoeck & Ruprecht. Adriaanse, M. A., Gollwitzer, P. M., De Ridder, D. T. D., de Wit, J. B. F., & Kroese, F. M. (2011). Breaking habits with implementation intentions:

A test of underlying processes. Personality and Social Psychology Bulletin, 37, 502–513. https:// doi.org/10.1177/0146167211399102 Adriaanse, M. A., Vinkers, C. D. W., De Ridder, D. T. D., Hox, J. J., & De Wit, J. B. F. (2011). Do implementation intentions help to eat a healthy diet? A systematic review and meta-analysis of the empirical evidence. Appetite, 56, 183–193. https:// doi.org/10.1016/j.appet.2010.10.012 Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50, 179–211. https://doi.org/10.1016/ 0749-5978(91)90020-T Allan, J. L., Sniehotta, F. F., & Johnston, M. (2013). The best laid plans: Planning skill determines the effectiveness of action plans and implementation intentions. Annals of Behavioral Medicine. https://doi.org/10.1007/ s12160-013-9483-9 Armitage, C. J. (2008). A volitional help sheet to encourage smoking cessation: A randomized exploratory trial. Health Psychology, 27, 557– 566. https://doi.org/10.1037/0278-6133.27.5.557 Armitage, C. J., & Conner, M. (2001). Efficacy of the theory of planned behavior: A meta-analytic review. British Journal of Social Psychology, 40, 471–499. https://doi.org/10.1348/014466601164939 Bagozzi, R. P. (1992). The self-regulation of attitudes, intentions, and behavior. Social Psychology Quarterly, 55, 178–204. https://doi.org/10.2307/ 2786945 Bandura, A. (1986). Social Foundations of Thought and Action: A Social-Cognitive Theory. Englewood Cliffs, NJ: Prentice Hall. Bélanger-Gravel, A., Godin, G., & Amireault, S. (2013). A meta-analytic review of the effect of implementation intentions on physical activity. Health Psychology Review, 7, 1–32. https://doi .org/10.1080/17437199.2011.560095 Bryan, A., Fisher, J. D., & Fisher, W. A. (2002). Tests of the mediational role of preparatory safer sexual behavior in the context of the theory of planned behavior. Health Psychology, 21, 71–80. https:// doi.org/10.1037//0278-6133.21.1.71 Carraro, N., & Gaudreau, P. (2013). Spontaneous and experimentally induced action planning and coping planning for physical activity: A meta-analysis.

Planning and Implementation Intention Interventions

Psychology of Sport and Exercise, 14, 228–248. https://doi.org/10.1016/j.psychsport.2012.10.004 Chapman, J., & Armitage, C. J. (2010). Evidence that boosters augment the long-term impact of implementation intentions on fruit and vegetable intake. Psychology and Health, 25, 365–381. https://doi.org/10.1080/08870440802642148 Churchill, S., & Jessop, D. C. (2010). Too impulsive for implementation intentions? Evidence that impulsivity moderates the effectiveness of an implementation intention intervention. Psychology and Health, 26, 1–14. https://doi.org/ 10.1080/08870441003611536 Conner, M., & Norman, P. (2015). Predicting Health Behaviour: Research and Practice with Social Cognition Models. Maidenhead: Open University Press. de Bruijn, G. J., Gardner, B., van Osch, L., & Sniehotta, F. F. (2014). Predicting automaticity in exercise behaviour: The role of perceived behavioural control, affect, intention, action planning, and behaviour. International Journal of Behavioral Medicine, 21, 767–774. https://doi.org/10.1007/ s12529-013-9348-4 de Bruijn, G. J., Nguyen, M. H., Rhodes, R. E., & van Osch, L. (2017). Effects of preparatory and action planning instructions on situation-specific and general fruit and snack intake. Appetite, 108, 161– 170. https://doi.org/10.1016/j.appet.2016.09.016 de Vries, H., Mesters, I., van de Steeg, H., & Honing, C. (2005). The general public’s information needs and perceptions regarding hereditary cancer: An application of the integrated change model. Patient Education and Counseling, 56, 154–165. https://doi.org/10.1016/j.pec.2004.01.002 Fishbein, M., & Ajzen, I. (1975). Belief, Attitude, Intention, and Behavior. Don Mills, NY: AddisonWesley. Fleig, L., Pomp, S., Parschau, L. et al. (2013). From intentions via planning and behavior to physical exercise habits. Psychology of Sport and Exercise, 14, 632–649. https://doi.org/10.1016/j.psychsport .2013.03.006 Gardner, B., Phillips, L. A., & Judah, G. (2016). Habitual instigation and habitual execution: Definition, measurement, and effects on behaviour frequency. British Journal of Health

583

Psychology, 21, 613–630. https://doi.org/10.1111/ bjhp.12189 Gollwitzer, P. M. (1990). Action phases and mind-sets. In E. T. Higgins & R. M. Sorrentino (Eds.), Handbook of Motivation and Cognition: Foundations of Social Behavior, Vol. 2 (pp. 53– 92). New York: Guilford Press. Gollwitzer, P. M. (1999). Implementation intentions: Strong effects of simple plans. American Psychologist, 54, 493–503. https://doi.org/ 10.1037/0003-066X.54.7.493 Gollwitzer, P. M. (2014). Weakness of the will: Is a quick fix possible? Motivation and Emotion, 38, 305–322. https://doi.org/10.1007/s11031-0149416-3 Gollwitzer, P. M., & Brandstatter, V. (1997). Implementation intentions and effective goal pursuit. Journal of Personality and Social Psychology, 73, 186–199. https://doi.org/10.1037/ 0022-3514.73.1.186 Gollwitzer, P. M., & Sheeran, P. (2006). Implementation intentions and goal achievement: A meta-analysis of effects and processes. Advances in Experimental Social Psychology, 38, 69–119. https://doi.org/10.1016/S0065-2601(06) 38002–1 Gӧhner, w., Seelig, H., & Fuchs, R. (2009). Intervention effects on cognitive antecedents of physical exercise: A 1-year follow-up study. Applied Psychology: Health and Well-Being, 1, 233–256. https://doi.org/10.1111/j.1758-0854 .2009.01014.x Hagger, M. S., & Chatzisarantis, N. L. D. (2014). An integrated behavior-change model for physical activity. Exercise and Sport Sciences Reviews, 42, 62–69. https://doi.org/10.1249/JES.0000000000 000008 Hagger, M. S., & Luszczynska, A. (2014). Implementation intention and action planning interventions in health contexts: State of the research and proposals for the way forward. Applied Psychology: Health and Well-Being, 6, 1–47. https://doi.org/10.1111/aphw.12017 Hagger, M. S., Luszczynska, A., de Wit, J. et al. (2016). Implementation intention and planning interventions in health psychology: Recommendations from the Synergy Expert

584

RYAN E. RHODES, STINA GRANT, AND GERT-JAN DE BRUIJN

Group for research and practice. Psychology and Health, 31, 814–839. https://doi.org/10.1080/ 08870446.2016.1146719 Hall, P. A., & Fong, G. T. (2007). Temporal selfregulation theory: A model for individual health behavior. Health Psychology Review, 1, 6–52. https://doi.org/10.1080/17437190701492437 Harris, P. R., Brearley, I., Sheeran, P. et al. (2014). Combining self-affirmation with implementation intentions to promote fruit and vegetable consumption. Health Psychology, 33, 729–736. https://doi.org/10.1037/hea0000065 Heckhausen, H., & Gollwitzer, P. M. (1987). Thought contents and cognitive functioning in motivational and volitional states of mind. Motivation and Emotion, 11, 101–120. https://doi.org/10.1007/ BF00992338 Knäuper, B., McCollam, A., Rosen-Brown, A., Lacaille, J., Kelso, E., & Roseman, M. (2011). Fruitful plans: Adding targeted mental imagery to implementation intentions increases fruit consumption. Psychology and Health, 26, 601– 617. https://doi.org/10.1080/08870441003703218 Kuhl, J. (1984). Motivational aspects of achievement motivation and learned helplessness: Towards a comprehensive theory of action control. In B. A. Maher & W. B. Maher (Eds.), Progress in Experimental Personality Research, Vol. 13 (pp. 99–171). New York: Academic Press. Kwasnicka, D., Presseau, J., White, M., & Sniehotta, F. F. (2013). Does planning how to cope with anticipated barriers facilitate health-related behaviour change? A systematic review. Health Psychology Review, 7, 129–145. https://doi.org/ 10.1080/17437199.2013.766832 Leventhal, H., Singer, R., & Jones, S. (1965). Effects of fear and specificity of recommendation upon attitudes and behavior. Journal of Personality and Social Psychology, 2, 20–29. https://doi.org/ 10.1037/h0022089 Lewin, K. (1951). Intention, will and need. In D. Rapaport (Ed.), Organization and Pathology of Thought: Selected Sources (pp. 95–153). New York: Columbia University Press. Locke, E. A., & Latham, G. P. (2002). Building a practically useful theory of goal setting and task motivation. American Psychologist, 57, 705–717. https://doi.org/10.1037/0003-066X.57.9.705

Locke, E. A., & Latham, G. P. (2006). New directions in goal setting theory. Current Directions in Psychological Science, 15, 265–268. https://doi .org/10.1111/j.1467-8721.2006.00449.x Martin, S. B., Morrow, J. R. J., Jackson, A. W., & Dunn, A. L. (2000). Variables related to meeting the CDC/ACSM physical activity guidelines. Medicine and Science in Sports and Exercise, 32, 2087–2092. https://doi.org/10.1097/00005768200012000-00019 McEachan, R., Taylor, N., Harrison, R., Lawton, R., Gardner, P., & Conner, M. (2016). Meta-analysis of the reasoned action approach (RAA) to understanding health behaviors. Annals of Behavioral Medicine, 50, 592–612. https://doi .org/10.1007/s12160-016-9798-4 Michie, S., Richardson, M., Johnston, M. et al. (2013). The behavior change technique taxonomy (v1) of 93 hierarchically clustered techniques: Building an international consensus for the reporting of behavior change interventions. Annals of Behavioral Medicine, 46, 81–95. https://doi.org/ 10.1007/s12160-013-9486-6 Napolitano, C. M., & Freund, A. M. (2016). On the use and usefulness of backup plans. Perspectives on Psychological Science, 11, 56–73. https://doi.org/ 10.1177/1745691615596991 O’Donovan, G., & Shave, R. (2007). British adults’ views on the health benefits of moderate and vigorous activity. Preventive Medicine, 45, 432– 435. https://doi.org/10.1016/j.ypmed.2007.07 .026 Orbell, S. (2004). Intention-behaviour relations: A selfregulation perspective. In G. Haddock & G. R. O. Maio (Eds.), Contemporary Perspectives on the Psychology of Attitudes (pp. 145–168). London: Psychology Press. Prochaska, J. O., & DiClemente, C. C. (1982). Transtheoretical therapy: Toward a more integrative model of change. Psychotherapy: Theory, Research and Practice, 19, 276–288. https://doi.org/10.1037/h0088437 https://doi.org/ 10.1037/h0088437 Rhodes, R. E. (2015). Building from the cracks of our foundational physical activity theories. Paper presented at the Canadian Society for Psychomotor Learning and Sport Psychology Conference, Edmonton, Alberta, October 16–18.

Planning and Implementation Intention Interventions

Rhodes, R. E. (2017). The evolving understanding of physical activity behavior: A multi-process action control approach. In A. J. Elliot (Ed.), Advances in Motivation Science, Vol. 4 (pp. 171–205). Cambridge, MA: Elsevier Academic Press. https://doi.org/10.1016/bs.adms.2016.11.001 Rhodes, R. E., & de Bruijn, G. J. (2013a). How big is the physical activity intention-behaviour gap? A meta-analysis using the action control framework. British Journal of Health Psychology, 18, 296–309. https://doi.org/10.1111/bjhp.12032 Rhodes, R. E., & de Bruijn, G. J. (2013b). What predicts intention-behavior discordance? A review of the action control framework. Exercise and Sports Sciences Reviews, 41, 201–207. https://doi.org/10.1097/JES.0b013e3182a4e6ed Rhodes, R. E., & Rebar, A. (2017). Conceptualizing and defining the intention construct for future physical activity research. Exercise and Sports Sciences Reviews, 45, 209–216. https://doi.org/ 10.1249/JES.0000000000000127 Rhodes, R. E., & Rebar, A. (2018). Physical activity habit: Complexities and controversies. In B. Verplanken (Ed.), The Psychology of Habit (pp. 91–109). Cham: Springer. Rhodes, R. E., & Yao, C. (2015). Models accounting for intention-behavior discordance in the physical activity domain: A user’s guide, content overview, and review of current evidence. International Journal of Behavioral Nutrition and Physical Activity, 12, 1–15. https://doi.org/10.1186/ s12966-015-0168-6 Rosenstock, I. M. (1974). Historical origins of the health belief model. Health Education Monographs, 2, 1–9. https://doi.org/10.1177/ 109019817400200403 Schwarzer, R. (2008). Modeling health behavior change: How to predict and modify the adoption and maintenance of health behaviors. Applied Psychology, 57, 1–29. https://doi.org/10.1111/ j.1464-0597.2007.00325.x Schwarzer, R. (2016). Coping planning as an intervention component: A commentary. Psychology and Health, 31, 903–906. https://doi .org/10.1080/08870446.2016.1158260 Sheeran, P. (2002). Intention-behaviour relations: A conceptual and empirical review. In M. Hewstone &

585

W. Stroebe (Eds.), European Review of Social Psychology, Vol. 12 (pp. 1–36). Chichester: John Wiley & Sons. https://doi.org/10.1080/14792772 143000003 Sheeran, P., & Webb, T. L. (2016). The intentionbehavior gap. Social and Personality Psychology Compass, 10, 503–518. https://doi.org/10.1111/ spc3.12265 Sniehotta, F. F. (2009). Towards a theory of intentional behaviour change: Plans, planning, and selfregulation. British Journal of Health Psychology, 14, 261–273. https://doi.org/10.1348/135910708 X389042 Sniehotta, F. F., Schwarzer, R., Scholz, U., & Schüz, B. (2005). Action planning and coping planning for long-term lifestyle change: Theory and assessment. European Journal of Social Psychology, 35, 565–576. https://doi.org/10.1002/ ejsp.258 The Telegraph. (2017). The most common New Year’s resolutions – and how to stick to them. The Telegraph, January 5. www.telegraph.co.uk/healthfitness/body/common-new-years-resolutions-stick/ van Osch, L., Lechner, L., Reubsaet, A., & De Vries, H. (2010). From theory to practice: An explorative study into the instrumentality and specificity of implementation intentions. Psychology and Health, 25, 351–364. https://doi.org/10.1080/ 08870440802642155 Vilà, I., Carrero, I., & Redondo, R. (2017). Reducing fat intake using implementation intentions: A metaanalytic review. British Journal of Health Psychology, 22, 281–294. https://doi.org/10.1111/ bjhp.12230 Webb, T. L., & Sheeran, P. (2006). Does changing behavioral intentions engender behavior change? A meta-analysis of the experimental evidence. Psychological Bulletin, 132, 249–268. https://doi .org/10.1037/0033-2909.132.2.249 Wilson, B. A., Alderman, N., Burgess, P. W., Emslie, H., & Evans, J. J. (1996). Behavioural Assessment of the Dysexecutive Syndrome. Bury St Edmunds: Thames Valley Test Company. Wood, W., & Runger, D. (2016). Psychology of habit. Annual Review of Psychology, 67, 1–11. https://doi.org/10.1146/annurev-psych122414-033417

40 Self-Control Interventions Denise de Ridder, Marleen Gillebaart, and Malte Friese

Practical Summary Pursuing long-term goals such as studying for an exam, saving enough money for a deposit to buy a house, and maintaining a harmonious interpersonal relationship requires persistence on important behaviors that lead to goal progress and avoidance of the myriad distractions and other available courses of action, particularly those that are more immediately appealing and rewarding. Such endeavors require “good” selfcontrol. Individuals with good self-control demonstrate persistence on tasks toward a particular goal and are adept at avoiding temptations that may derail those efforts. Such individuals usually develop good habits for desirable behaviors and strategically avoid situations where they might be tempted. Although research suggests that good self-control is relatively stable, there is evidence that self-control ability can be improved through training. A typical approach to training self-control involves individuals regularly engaging in tasks that require them to inhibit their innate desires and “dominant” responses. These tasks can be computerized tasks that require inhibition of a well-learned response or even everyday tasks with a similar requirement (e.g., using the nondominant hand or avoiding colloquial speech). Over time, such tasks have been shown to improve self-control for other behaviors. Despite the promising early findings of self-control training studies in improving behavioral outcomes, more studies are needed to examine the persistent effects of self-control training on meaningful, adaptive outcomes in the long term and in “real-world” contexts.

40.1 Introduction In many areas of life, people encounter selfcontrol dilemmas on a daily basis, if not multiple times a day. In the health domain, people may experience a conflict between an active lifestyle and the short-term gratification of curling up on the couch or between an appropriate bedtime and a series cliffhanger that tempts them to stay up. When it concerns financial matters, people have to balance long-term benefits of saving up for pensions or buying a home while being

bombarded with marketing strategies and opportunities for short-term gratifying consumer behavior. Relating to interpersonal affairs, controlling one’s behavior when interacting with loved ones can be challenging after a taxing work day or when fatigue sets in. These examples illustrate the potential of self-control as a means to prioritize long-term goals (such as staying fit) over https://doi.org/10.1017/9781108677318.040

Self-Control Interventions

gratification of immediate concerns (such as watching television all night). Many studies have demonstrated that self-control is an important contributing factor in determining behavior and outcomes in multiple life domains. For example, high self-control is associated with better health and well-being, adjustment, and satisfying social relationships, whereas low self-control is associated with poor health, addiction, financial debt, poor academic achievement, and lower job performance (De Ridder et al., 2012; Moffitt et al., 2011; Tangney, Baumeister, & Boone, 2004). Not only are people with high self-control more successful in various ways; they are also happier. For example, research has demonstrated that individuals with high self-control report greater happiness and psychological well-being (Cheung et al., 2014; Hofmann et al., 2014). From this broad evidence base, it follows that interventions that are effective in promoting high self-control have the potential to promote behaviors that will lead to better wellbeing and adaptive outcomes. However, despite accumulating insights into the impact of self-control on people’s lives, the development of interventions that improve self-control is still in a relatively early stage. Consistent, high-quality evidence for the implementation and long-term effectiveness of interventions to improve self-control is lacking. This chapter focuses on the current body of knowledge that has informed self-control interventions and provides an overview of the state of the research on the interventions as well as suggestions for future research that is required to gain a more comprehensive knowledge of selfcontrol interventions for behavior change in a variety of contexts and populations.

40.2 Definitions In order to discuss self-control interventions, it is first important to define two core concepts: selfcontrol and self-control training. Self-control can be defined as the process of giving precedence to distal, long-term motives over proximal, short-

587

term motives when these motives conflict. Prioritization may occur by resisting the proximal, short-term motives (such as temptations, impulses, desires) or by using self-control strategies that give way to long-term goals (De Ridder et al., 2012; Fujita, 2011; Gillebaart & De Ridder, 2015; Hofmann, Friese, & Strack, 2009; see also Chapter 11, this volume). It is important to distinguish trait from state self-control: Trait self-control is the dispositional propensity to exert self-control that differs between individuals and is relatively stable within individuals over time (Tangney et al., 2004). State self-control refers to the capacity for self-control that differs within individuals across contexts and situations. Existing self-control interventions primarily address state self-control but new interventions aimed at trait self-control are emerging (De Ridder et al., 2019). Self-control training refers to a variety of interventions aimed at enhancing the probability that individuals prioritize their longterm motives over their short-term ones when they are in conflict (Berkman, 2018).

40.3 Theory and Mechanisms of Change Traditionally, trait self-control research has focused on effortful inhibition as the core (and sometimes sole) component of self-control (e.g., Baumeister, 2014; Tangney et al., 2004). Recent theory and research has expanded this approach and proposed that people with higher trait self-control may also benefit from directly prioritizing their long-term goals, which may be achieved by strategies that take less effort than inhibiting impulses (De Ridder et al., 2012; Duckworth, Matthews, & Kelly, 2007; Fujita, 2011; Gillebaart & De Ridder, 2015; Trope & Fishbach, 2000). Specifically, a meta-analysis on trait self-control showed robust associations with outcomes that were stronger for behaviors that were rated as “automatic” as compared to behaviors that were rated as “controlled” (De Ridder et al., 2012). Recent research corroborated these findings

588

DENISE DE RIDDER, MARLEEN GILLEBAART, AND MALTE FRIESE

by demonstrating that habits (i.e., behaviors that may occur with little effort or conscious intention; see Chapter 13, this volume) consistently mediate the association between trait self-control and behavior in the areas of eating, exercising, and study behavior (Adriaanse et al., 2014; Galla & Duckworth, 2015; Gillebaart & Adriaanse, 2017). One implication of these findings is that automatization of goal-conducive behavior may become less effortful over time. Another implication is that prioritization of long-term motives does not need to happen in the heat of the moment of a self-control conflict but may also happen long before and in fact help individuals experience fewer conflicts in the first place (Hofmann, Baumeister et al., 2012). Besides the automatization of self-control as a way of successfully prioritizing long-term motives over short-term ones, scholars have also started to examine “strategic self-control,” that is, the employment of self-control skills as a function of situational requirements (Ayduk et al., 2000; Gillebaart & De Ridder, 2015; Fujita, 2011; Friese, Hofmann, & Wiers, 2011). For example, a set of studies on academic study behavior showed that students with a higher level of trait self-control more often used an avoidant strategy by opting for a room without distractions to study in instead of a room with more opportunity to socialize (Ent, Baumeister, & Tice, 2015). In addition, it has been proposed that people may rely on a set of strategies that could either be more “antecedent” to the self-control dilemma (and may potentially avoid the dilemma to fully unfold) or be more “reactive” to an already unfolded dilemma. Reactive strategies take more effort to resolve the dilemma (Duckworth, Gendler, & Gross, 2016). This is reminiscent of classic work by Mischel and colleagues, who mapped a number of similar strategies from their delay-ofgratification studies in children (Mischel, Shoda, & Rodriguez, 1989). However, a more solid evidence base to serve as a foundation for self-control interventions is still needed to support these ideas on strengthening strategies for self-control.

The contemporary view on trait self-control as going beyond effortful inhibition also holds implications for mechanisms of change that feed into self-control interventions. Trait self-control is considered to be relatively stable over time within an individual. However, this does not mean that there is no potential for improvement. Self-control interventions in this area have mainly focused on enhancing trait self-control by making self-control exertion more habitual and training the inhibition of impulsive tendencies toward short-term gratification. Improving the inhibitory part of trait self-control can potentially be achieved by direct training of “inhibitory control” as a core component of executive functioning (Diamond & Lee, 2011; Hofmann, Schmeichel, & Baddeley, 2012). Classic paradigms that have been used to gauge and train inhibitory self-control are go/no-go tasks (Donders, 1969) and stop-signal tasks (Lappin & Eriksen, 1966). Go/no-go tasks typically consist of a neutral set of stimuli and a set of target stimuli that are related to the undesired behavior that would need to be inhibited. Participants are instructed to respond as fast as possible (e.g., by pressing a button) to the neutral stimuli (go). Participants are also instructed to not respond to target stimuli (no-go). By repeating this for a number of trials, an association between inhibition and the target behavior is built that can carry over to actual behavior outside the task (e.g., Houben et al., 2011; Jones & Field, 2013; Veling, Aarts, & Papies, 2011). Stop-signal tasks typically also consist of neutral stimuli and target stimuli related to the undesired behavior. Participants need to categorize both types of stimuli as quickly as possible, except when there is a so-called stop signal (e.g., visually or auditory) presented right after the stimulus. A stop signal thus means that participants need to inhibit their already initiated response to the stimulus. This would lead to an association between “target” and “stop” that may again carry over into subsequent target behavior (Lawrence et al., 2015). However,

Self-Control Interventions

in spite of initial promising findings, recent research has convincingly demonstrated that this kind of self-control training has little or no transfer effects, making it a poor candidate for implementation into self-control interventions (Jones et al., 2018). A different type of inhibition training has been inspired by the strength or limited resource model of self-control (Baumeister et al., 1998; Baumeister & Vohs, 2016). This model understands self-control as the ability to control dominant responses such as impulses and emotions and thus heavily focuses on inhibition as the key characteristic of self-control (Baumeister, 2014). According to the model, exerting self-control draws on a domain-general and limited resource, which leads to depletion of the resource and an increased risk of subsequent failure after initial attempts at self-control. The bulk of work on the strength model of self-control is concerned with within-person fluctuations in self-control performance (Friese et al., 2019; Inzlicht & Friese, 2019). A smaller number of studies has examined the idea that self-control could be improved with practice. Following the frequently used metaphor that self-control works like a muscle or a kind of strength, the model posits that temporary demands may “weaken the muscle,” whereas repeated training would lead to a “stronger muscle” (Muraven, Baumeister, & Tice, 1999). Training tasks may involve a variety of exercises aimed at repeatedly overriding dominant responses or inhibiting impulses. For example, participants have been instructed to use their nondominant hand for everyday tasks (Miles et al., 2016), control their language use (Finkel et al., 2009), or utilize a handgrip several times a day until exhaustion of hand muscle strength (Job, Friese, & Bernecker, 2015). After repeatedly performing these self-control acts over a period of time, self-control should improve. Training success is primarily inferred in two ways: either from greater overall self-control strength as indicated by better performance in

589

self-control tasks or from reduced susceptibility to ego depletion effects, reflecting greater stamina in the face of high self-control demands. Importantly, because self-control is believed to be a domain-general construct, the strength model suggests that practicing self-control in one domain should lead to broad improvements in other domains requiring self-control. Whereas the strength model has inspired the majority of self-control interventions, it should be noted that it has been questioned on many occasions along with the implied notion of a “limited resource” (De Ridder, Kroese, & Gillebaart, 2018; Friese et al., 2019). An alternative explanation for the mechanism behind training effects based on the strength model may be that people create adaptive routines as a result of repeated engagements in tasks that originally required effortful inhibition (De Ridder et al., 2019).

40.4 Evidence Base In order to create a good overview of the current state of self-control training literature, it is useful to discuss the evidence for self-control training on two distinct levels (De Houwer, 2011; Gieseler, Loschelder, & Friese, 2019). The behavioral level refers to improvement in self-control behavior as a result of an intervention, such as those mentioned in Section 40.3. In contrast, the process level focuses on the psychological processes that mediate or explain the behavioral effect of self-control training on self-control outcomes. Even though the literature on selfcontrol interventions is growing, evidence on both levels of analyses is still relatively scarce. Looking first at habits as a way of improving trait self-control, there is a limited number of studies that have tested whether changes in habitual exertion of self-control lead to changes in self-control capacity over time (De Ridder et al., 2019; Gillebaart et al., 2019). Results from these studies are promising. At the behavioral level, an increase in self-control capacity has been

590

DENISE DE RIDDER, MARLEEN GILLEBAART, AND MALTE FRIESE

observed as a result of a self-control interventions focusing on regular practice of self-chosen selfcontrol behavior in different domains (e.g., health, interpersonal behavior; De Ridder et al., 2019). At the process level, repetition, a core component of habit formation (Verplanken & Orbell, 2003; Chapters 13 and 41, this volume), of both self-control behaviors (De Ridder, et al., 2019) and improved responding to self-control dilemmas (Gillebaart et al., 2019) seems to drive these improvements in general capacity for self-control. However, these studies are a starting point rather than a solid evidence base and more research is needed to corroborate and extend them. A second direction for improving trait selfcontrol is inhibitory control training tasks. As mentioned, go/no-go and stop-signal tasks have been demonstrated to affect subsequent target behavior such as drinking alcohol or eating unhealthy food at the behavioral level (e.g., Jones & Field, 2013; Veling et al., 2011). Meta-analyses have revealed a small but significant effect of these types of training (Allom, Mullan, & Hagger, 2016; Jones et al., 2016), with stronger effects for go/nogo compared to stop-signal–based training programs (Allom et al., 2016). However, beneficial effects were only demonstrated immediately after the training – questioning their longevity (Allom et al., 2016) – and primarily for participants who were motivated for behavior change (Jones et al., 2016). At the process level, there is a need for further research on the underlying processes of these types of interventions. For example, there is debate about the mechanisms underlying effects of training on go/no-go tasks on self-control improvements, which could be attributable to decreased affective associations (Veling et al., 2017) rather than increased inhibitory control (Jones et al., 2016). A recent preregistered randomized controlled trial went beyond the proof-of-concept studies summarized in existing meta-analyses (Jones et al., 2018). In this study, heavy drinkers were

assigned to one of three training conditions (alcohol-specific go/no-go, alcohol-specific stop-signal, general stop-signal) or an active control condition (categorizing alcohol and stationery pictures without requirement for inhibition). Participants attended up to fourteen training sessions over a four-week period and recorded their alcohol consumption. Results revealed that all groups reduced their alcohol consumption but there was no specific effect of any training condition on either alcohol consumption, inhibitory control, or affective associations with alcohol. These findings highlight that the empirical evidence and theoretical understanding of inhibitory control training do not yet warrant recommendations for interventions in applied contexts. A similar picture emerges for self-control training interventions based on the strength model of self-control. Two recent meta-analyses have summarized the current evidence for this kind of interventions that seek to train the self-control “muscle” (Beames, Schofield, & Denson, 2018; Friese et al., 2017). Both metaanalyses included only studies that employed a self-control training entailing repeated control over dominant responses, measured self-control outcomes in a different domain than the training, and assessed outcomes at least one day after the final training session. Average effect sizes were g = 0.30 (Friese et al., 2017) and g = 0.36 (Beames et al., 2018), falling within the smallto-medium range and a bit smaller than the average effect in social psychology (Richard, Bond, & Stokes-Zoota, 2003). One particularly regrettable shortcoming of this literature from a practical perspective is that only a minority of studies included a follow-up and the ones that did employed only a few days after the intervention ended. As a result, current evidence is not sufficient to provide a conclusive evaluation to estimate the longevity of the effects of this kind of self-control training. For details on the metaanalytic results of this type of intervention, see Sidebar 40.1.

Self-Control Interventions

591

Sidebar 40.1 Meta-analytic evidence on self-control training

Two meta-analyses have explored the effects of self-control training (i.e., overriding a predominant response) on self-control outcomes (Beames et al., 2018; Friese et al., 2017). In these meta-analyses, several aspects of self-control training studies were considered. For example, various types of training were defined (computerized inhibitory control training, nondominant hand tasks, squeezing a hand strength training device, posture training, and dieting instructions) but these different types did not differ significantly in their effectiveness. Similarly, length of training did not change effectiveness and no differences were found for outcomes in affect and wellbeing, inhibitory control (before or after a laboratory self-control task intended to weaken momentary self-control capacity), physical persistence, and health behavior. Furthermore, outcomes that were measured in the lab were similar to those measured in real-life settings and effectiveness did not differ between behavioral, self-reported, or cognitive outcomes. The type of incentive for participation also did not moderate effects. Note, however, that some of these nonsignificant moderator analyses may well be due to a lack of statistical power: less than thirty-five studies were eligible for inclusion in the meta-analyses. Some other factors did impact the training results: Published studies reported bigger effect sizes compared to unpublished studies; studies with active control conditions showed smaller effects than studies with inactive control conditions; bigger participant samples demonstrated smaller effects than smaller participant samples; and research groups that included proponents of the strength model reported larger effect sizes than other research groups. Finally, the distinction between self-control strength (self-control exertion without depletion) and self-control stamina (self-control exertion after depletion) proved meaningful: Self-control training affected self-control stamina to a significantly larger extent than it affected self-control strength.

In summary, evidence for the effectiveness of self-control training is promising but many questions remain unanswered with respect to the robustness and longevity of the effects on behavior change. In particular, evidence at the process level is largely absent. It is unclear as to which mediating variables and processes explain observed training effects. A multitude of processes that might be involved have been suggested, such as improved goal setting or greater self-efficacy (Inzlicht, Legault, & Teper, 2014), increased motivation for self-control (Beames et al., 2018), and enhanced self-control beliefs (Berkman, 2018). To date, none of these proposed mechanisms has been conclusively tested

empirically. More research is needed that (1) focuses on plausible boundary conditions of the training effect (e.g., change motivation in participants); (2) examines potential mechanisms, including a closer examination of expectancy effects; and (3) goes beyond small-scale proofof-concept studies and takes a more comprehensive approach in larger nonstudent samples.

40.5 Preliminary Guidelines for SelfControl Interventions In terms of establishing a guideline that describes typical means of delivery, target audience and behavior, enabling and inhibiting factors, training

592

DENISE DE RIDDER, MARLEEN GILLEBAART, AND MALTE FRIESE

Sidebar 40.2 Example of a self-control intervention protocol

Studies on self-control interventions have demonstrated that practicing self-control for a specified period of time (often two weeks) in one domain – for example, by using one’s nondominant hand for daily routines, avoiding sweets, or squeezing a spring-loaded handgrip trainer – can lead to improvement in another unrelated domain, as assessed by performance on laboratory tasks under ego depletion or by self-control behavior in or outside the lab (for meta-analytic evidence, see Friese et al., 2017). Very few studies have employed more meaningful training tasks that speak to members of the community and that are engaging for a long period of time. One example is a study on improving the capacity for self-control (De Ridder et al., 2019) that required participants to regularly practice a behavior they considered personally important but thus far had not managed to perform on a regular basis. They were provided with a choice of behaviors relating to either health, interpersonal, financial, or environmental issues and were asked to indicate in which particular contexts they wanted to practice (e.g., eating fruit when having breakfast, being patient when talking to a friend, saving money when in the supermarket, or recycling when tidying up). It was emphasized that they should choose a behavior and context that allowed them to practice on a daily basis (e.g., when they chose exercise, it was explained that a ten-minute walk was more feasible than an hour at the gym). The findings from this study show that practicing these tasks for about four months led to a considerable improvement in self-control capacity.

and skills required, intensiveness, and evaluation of fidelity, more robust evidence is required than the present state of the research literature can provide. Before recommendations can be made for translating self-control interventions into a comprehensive protocol that can be applied in real-life settings, more systematic research is needed on the different potential pathways to self-control improvement. A handful of preliminary guidelines based on the currently available evidence on self-control interventions are described in Sidebar 40.2.

40.5.1 Typical Means of Delivery There are several means to deliver self-control interventions. All training programs described in this chapter either involve tasks that are provided in person by a trainer or facilitator or involve

tasks that people can perform at home after instruction by a trainer or facilitator. Internet interventions or regular practice with the help of a smartphone or portable device “app” (or registration of completed tasks by “app”) have also been employed (see Chapter 29, this volume). These tasks involve either training of inhibitory behavior or prioritization of goal-directed behavior by frequent practice, as described in previous sections. However, to date, most self-control interventions have been delivered as part of a scientific study, primarily in student populations, and have therefore not been employed in the context of a regular intervention for the general public. Moreover, these studies typically employ lab tasks (such as solving anagrams), which might not be very involving for people outside academia (De Ridder et al., 2019), with the exception of studies that aim to engage people

Self-Control Interventions

in practicing tasks that most people would recognize as tedious but need-to-be-done jobs (such as folding one’s clothes before going to sleep, doing the dishes, or keeping a record of personal finances; De Ridder et al., 2019; Muraven et al., 1999). Future research on self-control interventions should spend more attention on these kinds of meaningful tasks that speak to motivation for self-control. Different from existing interventions (e.g., inhibition training or strengthening the selfcontrol resource) that posit that task relevance is not important, this new perspective suggests that meaningfulness of training tasks does matter.

40.5.2 Target Audience and Behaviors The majority of intervention research on self-control interventions has employed student samples. Only a minority has involved community samples and many of those interventions did not prove effective in these broader samples (Beames et al., 2018), with the exception of a few interventions that demonstrated improvement in self-control capacity in at-risk populations in the general community (De Ridder et al., 2018; Wang, Raoa, & Houserb, 2017). Training tasks may target a broad variety of behaviors but many interventions are aimed at appetitive behaviors such as food and alcohol consumption (e.g., Allom et al., 2016; Friese et al., 2011). The focus on these latter types of behaviors is unsurprising, given that many people experience problems with regulating their appetites (Hofmann, Vohs, & Baumeister, 2012). However, self-control interventions in other behavioral domains such as personal finances, social behavior, media use, or procrastination should be considered in more detail, as many people experience self-control problems in these areas of life as well.

40.5.3 Enabling or Inhibiting Factors As discussed in this chapter, a clear overview of factors that may moderate training effectiveness

593

is difficult to identify at this stage, as most evaluations of intervention effectiveness focus on establishing behavioral effects without considering personal characteristics (e.g., motivation for self-control, capacity for self-control; De Ridder et al., 2019) or contextual aspects (e.g., poor living circumstances, personal circumstances; Watts, Duncan, & Quan, 2018) that may promote training adherence or effectiveness. Also, cultural factors may be relevant, as recent research has suggested that, for example, in Asian cultures exercising self-control is regarded as energizing rather than as requiring effort (Savani & Job, 2017). These factors deserve more attention in future research.

40.5.4 Training and Skills Required Ideally, self-control interventions should be aimed at people with poor self-control skills or a lack of personal resources as they may benefit most from improving self-control. However, at this stage, it is unclear to what extent people who are struggling with self-control issues, such as those with problems with alcohol abuse, excessive food intake, or financial matters, are even interested in training self-control or capable of performing self-control training tasks. An experience sampling study on self-regulation in a community sample revealed that people with low self-control dropped out of the study at an early stage (Prinsen et al., 2018), demonstrating that it may not be so easy to involve participants who may benefit the most.

40.5.5 Intensiveness and Fidelity Formats for self-control interventions are, generally speaking, not very intensive and comprise a limited number of training sessions in a limited time frame, mostly lasting for two weeks or less, in a lab setting or online with easy-to-perform tasks. Notwithstanding this low-intense format, adhering to training tasks and/or attending

594

DENISE DE RIDDER, MARLEEN GILLEBAART, AND MALTE FRIESE

sessions may be problematic for participants with initially low levels of self-control (Prinsen et al., 2018), suggesting that even low-intensity formats of self-control interventions should pay attention to making training involvement an engaging activity. A recent study on self-control practice that involved personally relevant training tasks revealed that practicing self-chosen tasks was a reason for participants to persist with the intervention for a period of almost four months (De Ridder et al., 2019).

40.6 Further Requirements for a Step-by-Step Guide Considering the overall limited and inconclusive evidence for behavioral effects of self-control training, and the lack of insight into its working mechanisms, the compilation of a detailed stepby-step guide is largely dependent on the findings of future research. This section includes a list of issues that need to be clarified before such a stepby-step guide can be construed.

40.6.2 Sustainability of Effects Self-control is required for achieving one’s longterm goals in the face of temptation. This means that developing self-control interventions for those who struggle with long-term goal pursuit is only viable when their effects are sustained over a longer period of time. In turn, this implies that more longitudinal studies should be done to test self-control training. Training studies based on the strength model of self-control typically do not go beyond two weeks (Friese et al., 2017), with the exception of a six-week training reported in Miles et al. (2016), which may be too short for determining effects in terms of goal accomplishment (e.g., health goals usually need more than a few weeks to be achieved). Initial longitudinal studies into trait self-control improvement have employed a longer time frame, with a duration up to 110 days (75 days on average), and suggest that positive effects do seem to hold over this period of time (De Ridder et al., 2019; Gillebaart et al., 2019).

40.6.3 Near vs. Far Transfer 40.6.1 Working Mechanisms Studies on self-control training have mainly focused on documenting the behavioral effects of these interventions, precluding insights into the processes that may mediate these effects. Trait self-control improvement studies have given a first inkling as to the nature of potential mediating processes (i.e., behavioral repetition leading to self-control routines, De Ridder et al., 2018; responsivity to self-control dilemmas, Gillebaart et al., 2019) but the remainder of studies shows no or inconclusive findings on underlying processes. Similarly, in the area of state self-control training, the underlying processes remain unclear. Before using self-control intervention programs based on these studies, there is first a need to focus not only on further establishing the effectiveness but also on unravelling potential mediating and moderating processes.

Ideally, self-control interventions should improve a general self-control capacity, skill, or resource and not self-control performance that is specific to one domain. Improved documentation of near and far transfer is therefore a crucial and urgent research direction. Research on trait self-control improvement has investigated the effect of specific acts of self-control on the general capacity for selfcontrol (De Ridder et al., 2019; Gillebaart et al., 2019) but more often is focused on the effects of training a very specific behavior in one domain without considering the effects on related domains (e.g., Veling et al., 2011; Wiers et al., 2010). The strength model of self-control assumes that selfcontrol relies on a domain-general resource, suggesting that improving self-control by practice should lead to broad improvements in behaviors that require self-control across various domains (Baumeister, Vohs, & Tice, 2007). These kinds of

Self-Control Interventions

interventions therefore tend to focus on training self-control in one specific domain (e.g., using one’s nondominant hand for everyday tasks) and testing self-control in another, usually equally specific domain (e.g., performance on a depletion task in the lab or self-reports on specific behaviors). Ideally, a more systematic study of near transfer (i.e., determine whether the effect of training one specific self-control behavior holds over time) and far transfer (i.e., determine whether the effect of training one specific self-control behavior generalizes to unrelated self-control behaviors) would give a better overview of how self-control interventions perform on a generalizability dimension and provide some clues for determining the specificity of training tasks.

40.7 Summary and Conclusion Summarizing, self-control is essential in many behavioral domains, and improving self-control through interventions holds the potential for significant impact, specifically for those who struggle with low levels of self-control. However, although the current evidence base on self-control training is growing, it is, in its current form, insufficient to inform wide-scale implementation of interventions. Several identified gaps could be addressed in future research. For example, current findings do not provide conclusive evidence on near versus far transfer of training effects. Studying these effects in participant samples that resemble the target audience for self-control interventions (i.e., individuals that struggle with low self-control, are motivated to improve, and willing to partake in a self-control training program) would provide useful information. The latter aspect should not be underestimated: Many self-control studies are carried out with relatively privileged groups who may already possess high levels of self-control (e.g., students) and studies have shown early dropout from those with low self-control. In fact, the field needs to resolve and cultivate theoretical and empirical

595

debate on several aspects of self-control in addition to studying self-control training. For example, the introduction of habits as a component or process of self-control and the notion of situational strategies for self-control have provided a perspective on self-control that is broader than it has been for a long time. Integrating these different aspects in a self-control training regimen is a necessary step to take before conclusive practical advice can be given.

References Adriaanse, M. A., Kroese, F. M., Gillebaart, M., & De Ridder, D. T. D. (2014). Effortless inhibition: Habit mediates the relation between self-control and unhealthy snack consumption. Frontiers in Psychology, 5, 444. https://doi.org/10.3389/ fpsyg.2014.00444 Allom, V., Mullan, B., & Hagger, M. (2016). Does inhibitory control training improve health behaviour? A meta-analysis. Health Psychology Review, 10, 168–186. https://doi.org/10.1080/ 17437199.2015.1051078 Ayduk, O., Mendoza-Denton, R., Mischel, W., Downey, G., Peake, P. K., & Rodriguez, M. (2000). Regulating the interpersonal self: Strategic self-regulation for coping with rejection sensitivity. Journal of Personality and Social Psychology, 79, 776–792. https://doi.org/10.1037/ 0022-3514.79.5.776 Baumeister, R. F. (2014). Self-regulation, ego depletion, and inhibition. Neuropsychologia, 65, 313–319. https://doi.org/10.1016/j.neuro psychologia.2014.08.012 Baumeister, R. F., Bratslavsky, E., Muraven, M., & Tice, D. M. (1998). Ego depletion: Is the active self a limited resource? Journal of Personality and Social Psychology, 74, 1252–1265. https://doi .org/10.1037//0022-3514.74.5.1252 Baumeister, R. F. & Vohs, K. D. (2016). Strength model of self-regulation as limited resource: Assessment, controversies, update. In M. Zanna & J. Olson (Eds.), Advances in Experimental Social Psychology, Vol. 54 (pp. 67–127). Cambridge, MA: Academic Press.

596

DENISE DE RIDDER, MARLEEN GILLEBAART, AND MALTE FRIESE

Baumeister, R. F., Vohs, K. D., & Tice, D. M. (2007). The strength model of self-control. Current Directions in Psychological Science, 16, 351–355. https://doi.org/10.1111/j.1467-8721.2007.00534.x Beames, J. R., Schofield, T. P., & Denson, T. F. (2018). A meta-analysis of improving selfcontrol with practice. In D. T. D. de Ridder, M. A. Adriaanse, & K. Fujita (Eds.), Routledge International Handbook of Self-Control in Health and Well-Being (pp. 405–417). Abingdon: Routledge. https://doi.org/10.4324/ 9781315648576-32 Berkman, E. T. (2018). Self-regulation training. In K. D. Vohs & R. F. Baumeister (Eds.), Handbook of Self-Regulation: Research, Theory and Applications (3rd ed., pp. 440–457). New York: Guilford Press. Cheung, T. T., Gillebaart, M., Kroese, F. M., & De Ridder, D. T. D. (2014). Why are people with high self-control happier? The effect of trait selfcontrol on happiness as mediated by regulatory focus. Frontiers in Psychology, 5, 722. https://doi .org/10.3389/fpsyg.2014.00722 De Houwer, J. (2011). Why the cognitive approach in psychology would profit from a functional approach and vice versa. Perspectives on Psychological Science, 6, 202–209. https://doi .org/10.1177/1745691611400238 De Ridder, D. T. D., Kroese, F. M., & Gillebaart, M. (2018). Whatever happened to self-control? A proposal for integrating notions from trait selfcontrol studies into state self-control research. Motivation Science, 4, 39–49. https://doi.org/ 10.1037/mot0000062 De Ridder, D. T. D., Lensvelt-Mulders, G., Finkenauer, C., Stok, F. M., & Baumeister, R. F. (2012). Taking stock of self-control: A meta-analysis of how trait self-control relates to a wide range of behaviors. Personality and Social Psychology Review, 16, 76– 99. https://doi.org/10.1177/1088868311418749 De Ridder, D. T. D., Van der Weiden, A., Gillebaart, M., Benjamins, J. S., & Ybema, J. F. (2019). Just do it: Engaging in self-control on a daily basis improves the capacity for self-control. Manuscript under review. Diamond, A., & Lee, K. (2011). Interventions shown to aid executive function development in children 4

to 12 years old. Science, 333, 959–964. https://doi .org/10.1126/science.1204529 Donders, F. C. (1969). On the speed of mental processes. Acta Psychologica, 30, 412–431. https://doi.org/10.1016/0001-6918(69)90065–1 Duckworth, A. L., Gendler, T. S., & Gross, J. J. (2016). Situational strategies for self-control. Perspectives on Psychological Science, 11, 35–55. https://doi.org/10.1177/1745691615623247 Duckworth, A. L., Matthews, M. D., & Kelly, D. R. (2007). Grit: Perseverance and passion for longterm goals. Journal of Personality and Social Psychology, 92, 1087–1101. https://doi.org/ 10.1037/0022-3514.92.6.1087 Ent, M. R., Baumeister, R. F., & Tice, D. M. (2015). Trait self-control and the avoidance of temptation. Personality and Individual Differences, 74, 12– 15. https://doi.org/10.1016/j.paid.2014.09.031 Finkel, E. J., DeWall, C. N., Slotter, E. B., Oaten, M., & Foshee, V. A. (2009). Self-regulatory failure and intimate partner violence perpetration. Journal of Personality and Social Psychology, 97, 483–499. https://doi.org/10.1037/a0015433 Friese, M., Frankenbach, J., Job, V., & Loschelder, D. D. (2017). Does self-control training improve selfcontrol? A meta-analysis. Perspectives on Psychological Science, 12, 1077–1099. https:// doi.org/10.1177/1745691617697076 Friese, M., Hofmann, W., & Wiers, R. W. (2011). On taming horses and strengthening riders: Recent developments in research on interventions to improve self-control in health behaviors. Self and Identity, 10, 336–351. https://doi.org/10.1080/ 15298868.2010.536417 Friese, M., Loschelder, D. D., Gieseler, K., Frankenbach, J., & Inzlicht, M. (2019). Is ego depletion real? An analysis of arguments. Personality and Social Psychology Review, 23, 107–131. https://doi.org/ 10.1177/1088868318762183 Fujita, K. (2011). On conceptualizing self-control as more than the effortful inhibition of impulses. Personality and Social Psychology Review, 15, 352–366. https://doi.org/10.1177/ 1088868311411165 Galla, B. M., & Duckworth, A. L. (2015). More than resisting temptation: Beneficial habits mediate the relationship between self-control and positive life

Self-Control Interventions

outcomes. Journal of Personality and Social Psychology, 109, 508–525. https://doi.org/ 10.1037/pspp0000026 Gieseler, K., Loschelder, D. D., & Friese, M. (2019). What makes for a good theory? How to evaluate a theory using the Strength Model of Self-Control as an example. In K. Sassenberg & M. Vliek (Eds.), Social Psychology in Action: Evidence-Based Interventions from Theory to Practice (pp. 3–22). Heidelberg: Springer. Gillebaart, M., & Adriaanse, M.A. (2017). Self-control predicts exercise behavior by force of habit, a conceptual replication of Adriaanse et al. (2014). Frontiers in Psychology, 8, 190. https://doi.org/ 10.3389/fpsyg.2017.00190 Gillebaart, M., Benjamins, J. S. Van der Weiden, A., Ybema, J. F., & De Ridder, D. T. D. (2019). Practice makes perfect: Repeatedly dealing with response conflicts facilitates its identification and speed of resolution. Manuscript under review. Gillebaart, M., & De Ridder, D. T. D. (2015). Effortless self-control: A novel perspective on response conflict strategies in trait self-control. Social and Personality Psychology Compass, 9, 88–99. https://doi.org/10.1111/spc3.12160 Hofmann, W., Baumeister, R. F., Förster, G., & Vohs, K. D. (2012). Everyday temptations: An experience sampling study of desire, conflict, and self-control. Journal of Personality and Social Psychology, 102, 1318–1335. https://doi.org/ 10.1037/a0026545 Hofmann, W., Friese, M., & Strack, F. (2009). Impulse and self-control from a dual-systems perspective. Perspectives on Psychological Science, 4, 162–176. https://doi.org/10.1111/j.1745-6924.2009.01116.x Hofmann, W., Luhmann, M., Fisher, R. R., Vohs, K. D., & Baumeister, R. F. (2014). Yes, but are they happy? Effects of trait self-control on affective well-being and life satisfaction. Journal of Personality, 82, 265–277. https://doi.org/10.1111/ jopy.12050 Hofmann, W., Schmeichel, B. J., & Baddeley, A. D. (2012). Executive functions and self-regulation. Trends in Cognitive Sciences, 16, 174–180. https://doi.org/10.1016/j.tics.2012.01.006 Hofmann, W., Vohs, K. D., & Baumeister, R. F. (2012). What people desire, feel conflicted about, and try

597

to resist in everyday life. Psychological Science, 23, 582–588. https://doi.org/10.1177/09567976 12437426 Houben, K., Nederkoorn, C., Wiers, R. W., & Jansen, A. (2011). Resisting temptation: Decreasing alcohol-related affect and drinking behavior by training response inhibition. Drug and Alcohol Dependency, 116, 132–136. https://doi.org/ 10.1016/j.drugalcdep.2010.12.011 Inzlicht, M., & Friese, M. (2019). The past, present, and future of ego depletion. Social Psychology, 50, 370–378. https://doi.org/10.1027/1864-9335/ a000398 Inzlicht, M., Legault, L., & Teper, R. (2014). Exploring the mechanisms of self-control improvement. Current Directions in Psychological Science, 23, 302–307. https://doi.org/10.1177/096372141453 4256 Job, V., Friese, M., & Bernecker, K. (2015). Effects of practicing self-control on academic performance. Motivation Science, 1, 219–232. https://doi.org/ 10.1037/mot0000024 Jones, A., & Field, M. (2013). The effects of cuespecific inhibition training on alcohol consumption in heavy social drinkers. Experimental Clinical Psychopharmacology, 21, 8–16. https://doi.org/10.1037/a0030683 Jones,A., Di Lemma, L. C., Robinson, E. et al. (2016). Inhibitory control training for appetitive behavior change: A meta-analytic investigation of mechanisms of action and moderators of effectiveness. Appetite, 97, 16–28. https://doi.org/ 10.1016/j.appet.2015.11.013 Jones, A., McGrath, E., Robinson, E., Houben, K., Nederkoorn, C., & Field, M. (2018). A randomized controlled trial of inhibitory control training for the reduction of alcohol consumption in problem drinkers. Journal of Consulting and Clinical Psychology, 86, 991–1004. https://doi .org/10.1037/ccp0000312 Lappin, J. S., & Eriksen, C. W. (1966). Use of a delayed signal to stop a visual reaction-time response. Journal of Experimental Psychology, 72, 805– 811. https://doi.org/10.1037/h0021266 Lawrence, N. S., Verbruggen, F., Morrison, S., Adams, R. C., & Chambers, C. D. (2015). Stopping to food can reduce intake: Effects of stimulus-specificity

598

DENISE DE RIDDER, MARLEEN GILLEBAART, AND MALTE FRIESE

and individual differences in dietary restraint. Appetite, 85, 91–103. https://doi.org/10.1016/j .appet.2014.11.006 Miles, E., Sheeran, P., Baird, H., Macdonald, I., Webb, T. L., & Harris, P. R. (2016). Does self-control improve with practice? Evidence from a six-week training program. Journal of Experimental Psychology: General, 145, 1075–1091. https:// doi.org/10.1037/xge0000185 Mischel, W., Shoda, Y., & Rodriguez, M. I. (1989). Delay of gratification in children. Science, 244, 933–938. https://doi.org/10.1126/science.2658056 Moffitt, T. E., Arseneault, L., Belsky, D. et al. (2011). A gradient of childhood self-control predicts health, wealth, and public safety. Proceedings of the National Academy of Sciences, 108, 2693–2698. https://doi.org/10.1073/pnas.1010076108 Muraven, M., Baumeister, R. F., & Tice, D. M. (1999). Longitudinal improvement of self-regulation through practice: Building self-control strength through repeated exercise. The Journal of Social Psychology, 139, 446–457. https://doi.org/ 10.1080/00224549909598404 Prinsen, S., Evers, C., Wijngaards, L., Van Vliet, R., & De Ridder, D. T. D. (2018). Does self-licensing benefit self-regulation over time? An ecological momentary assessment study of food temptations. Personality and Social Psychology Bulletin, 44, 914–927. https://doi.org/10.1177/0146167218754509 Richard, F. D., Bond, C. F., & Stokes-Zoota, J. J. (2003). One hundred years of social psychology quantitatively described. Review of General Psychology, 7, 331–363. https://doi.org/10.1037/ 1089-2680.7.4.331 Savani, K., & Job, V. (2017). Reverse ego-depletion: Acts of self-control can improve subsequent performance in Indian cultural contexts. Journal of Personality and Social Psychology, 113, 589– 607. https://doi.org/10.1037/pspi0000099 Tangney, J. P., Baumeister, R. F., & Boone, A. L. (2004). High self-control predicts good adjustment, less pathology, better grades, and

interpersonal success. Journal of Personality, 72, 271–324. https://doi.org/10.1111/j.0022-3506 .2004.00263.x Trope, Y., & Fishbach, A. (2000). Counteractive self-control in overcoming temptation. Journal of Personality and Social Psychology, 79, 493– 506. https://doi.org/10.1037//0022-3514.79.4 .493 Veling, H., Aarts, H., & Papies, E. K. (2011). Using stop signals to inhibit chronic dieters’ responses toward palatable foods. Behaviour Research and Therapy, 49, 771–780. https://doi.org/10.1016/j .brat.2011.08.005 Veling, H., Lawrence, N. S., Chen, Z., Van Koningsbruggen, G. M., & Holland, R. W. (2017). What is trained during food go/no-go training? A review focusing on mechanisms and a research agenda. Current Addiction Reports, 4, 35–41. https://doi.org/10.1007/ s40429-017-0131-5 Verplanken, B., & Orbell, S. (2003). Reflections on past behavior: A self-report index of habit strength. Journal of Applied Social Psychology, 33, 1313– 1330. https://doi.org/10.1111/j.1559-1816.2003 .tb01951.x Wang, J., Raoa, Y., & Houserb, D. E. (2017). An experimental analysis of acquired impulse control among adult humans intolerant to alcohol. Proceedings of the National Academy of Sciences, 114, 1299–1304. https://doi.org/10.1073/pnas .1610902114 Watts, T. W., Duncan, G. J., & Quan, H. (2018). Revisiting the Marshmallow Test: A conceptual replication investigating links between early gratification delay and later outcomes. Psychological Science, 29, 1159–1177. https:// doi.org/10.1177/0956797618761661 Wiers, R. W., Rinck, M., Kordts, R., Houben, K., & Strack, F. (2010). Retraining automatic actiontendencies to approach alcohol in hazardous drinkers. Addiction, 105, 279–287. https://doi.org/ 10.1111/j.1360-0443.2009.02775.x

41 Habit Interventions Benjamin Gardner, Amanda L. Rebar, and Phillippa Lally

Practical Summary Making a behavior habitual, such that it occurs automatically in specific settings, may help to sustain that behavior even when people lose motivation. Conversely, disrupting habitual performances can help people give up existing unwanted actions. Habit forms when an action is repeated in a specific setting; this creates associations that trigger the action on subsequent occasions without deliberation. Strategies commonly used to form habit among people who want to change focus on planning and executing new actions. Unwanted habits can be disrupted by avoiding triggers, willfully restraining urges to respond to those triggers, or enacting new actions in place of old responses. Habit interventions show promise for changing behavior and can be practitioner-delivered or selfadministered, though it can be important to ensure that practitioners and recipients alike are aware of the difference between habitual behavior, which is automatically triggered, and frequently performed behavior, which need not be automatic.

41.1 Introduction

41.2 Definitions

Intervention developers are generally good at promoting short-term behavior change but, over the longer term, people tend to abandon their change attempts, lapsing back into old patterns of behavior (Tsai & Wadden, 2005). Habit theory (see also Chapter 13, this volume) can help to understand and aid behavior change in such situations. Forming a new habit for a desired action may sustain that action even where motivation dips. Conversely, habitual behaviors can obstruct change, overriding conscious motivation to perform new, desired actions. This chapter explores strategies for making and breaking habits for lasting change.

Habit is viewed as a process that generates impulses to act in a given situation, based on situation-action associations learned through repetition (Gardner, 2015; see also Chapter 13, this volume). Impulses are mental representations of action that energize enactment of those actions (West & Brown, 2013). Habitual behavior denotes the action generated by the habit process and can take one of two forms. A behavior is “habitually instigated” where a person automatically commits to performing it and so initiates the first sub-action in a behavioral sequence (Gardner, Phillips, & Judah, 2016); for example, on https://doi.org/10.1017/9781108677318.041

600

BENJAMIN GARDNER, AMANDA L. REBAR, AND PHILLIPPA LALLY

leaving work at 5:30 p.m., a habitual gym-goer is triggered to begin walking to the gym, with no consideration of alternatives. A behavior is “habitually executed” where progression through the sequence of sub-actions is facilitated by habit; for example, successfully arriving in the gym habitually cues use of the treadmill, completion of which habitually cues use of the dumbbells, and so on (Gardner et al., 2016). The characteristic effects of habit on action can be attributed to habitual instigation rather than execution (Gardner et al., 2016). This chapter focuses on habits of instigation and uses “habitual behavior” synonymously with “habitually instigated behavior.” Habit strength is most accurately portrayed on a continuum (Moors & De Houwer, 2006), making it difficult to identify a dichotomous cut point that distinguishes “habitual” and “non-habitual” responding. It is important to note that the term “habit formation” is solely used to denote the strengthening of habit (Lally et al., 2010). “Habit disruption” is an umbrella term for strategies that prevent habitual performance: “habit discontinuity” (discontinuing exposure to cues that elicit habitual responses; Walker, Thomas, & Verplanken, 2015); “habit inhibition” (willful frustration of a habit impulse prior to behavioral enactment); “habit degradation” (erosion of learned habit associations); and “habit substitution” (direct displacement of one habitual response to a cue with another; Gardner & Lally, 2018).

41.3 Theory and Mechanisms of Change Three key features distinguish habitual from nonhabitual behaviors. First, habitual responses are acquired through consistent repetition. Through associative learning and operant conditioning processes (Lally & Gardner, 2013), people acquire mental associations between cues and behavioral responses that yield satisfactory, reinforcing outcomes. For example, people may develop habits for eating high-calorie snacks (behavior) when

bored (cue) because of the pleasant taste (satisfactory outcome). Second, once formed, habit elicits behavior via automatic activation of impulses. Thus, habitual action can be initiated without conscious deliberation, motivation, or awareness. Cues elicit nonconscious impulses to perform the associated action (e.g., eating high-calorie snacks), which, unless opposed by stronger momentary conflicting influences (e.g., the desire to eat healthily), will automatically prompt the action. Third, habitual behaviors are context-dependent; only cues that have come to be associated with the behavior activate habit impulses. Habitual behaviors in one context can occur non-habitually in another; for example, the habitual “afternoon snacker” may mindfully snack at other times. Habit formation involves “locking in” new behaviors by making them automatic and so resistant to motivation lapses. The main “ingredient” of habit formation – context-dependent repetition, through which people develop associations – is in practice rarely used as a standalone behavior change technique (BCT). Additional techniques are often required to initiate and sustain repetition. Realistically, planned habit formation requires intention and selfcontrol to initiate and maintain context-dependent repetition in opportune settings (Gardner & Lally, 2018; see Figure 41.1). Interventions can thereby promote habit formation by targeting motivation (see, e.g., Chapters 2, 3, and 8, this volume), selfregulatory processes (see, e.g., Chapter 11, this volume), context-dependent performance, or factors that facilitate the translation of repetition into habit. Habitual behaviors may be “broken” via multiple mechanisms (Figure 41.2). “Downstream” strategies (habit discontinuity and inhibition) disrupt habitual behavior but do not tackle the underlying habit (Gardner, 2015). “Upstream” approaches target either the “unlearning” (habit degradation) or the overwriting of associations with new responses that will engender new, stronger associations with the same cue (habit substitution). Habit degradation refers to a natural decaying of associations, not a purposive

Habit Interventions

601

Figure 41.1 A model of the habit formation process

Figure 41.2 Four forms of “habit disruption”

standalone intervention strategy, so is not discussed further. Habit substitution concurrently degrades existing habit associations and develops new ones, so draws on both formation and disruption mechanisms.

41.4 Evidence Base Illustrative habit intervention examples can be identified across behavioral domains (see Gardner & Rebar, 2019).

41.4.1 Habit Formation The “Ten Top Tips” (TTT) weight-loss intervention focused on a leaflet promoting habitual performance

of simple dietary and physical activity behaviors (Beeken et al., 2012, 2017; Lally, Chipperfield, & Wardle, 2008; Lally, Wardle, & Gardner, 2011). The leaflet included ten habit-based recommendations, advice about the importance of weight management, and a “tick-sheet” for monitoring adherence to each tip. In a pilot study, TTT intervention recipients, who were also regularly weighed, lost more weight than a no-treatment control group at eight weeks and maintained it to thirty-two weeks. A large-scale randomized trial found that primary care patients with obesity who received the TTT lost more weight over three months than did those receiving usual care (Beeken et al., 2017). At twenty-four months, intervention recipients had maintained their weight, though the control group

602

BENJAMIN GARDNER, AMANDA L. REBAR, AND PHILLIPPA LALLY

had lost a similar amount of weight, so no betweengroup differences were found. Effects on weight loss were mediated by habit gains, pointing to habit formation as a key mechanism (Kliemann et al., 2017). For an additional example of the application of the TTT intervention see Hamilton et al. (2019).

41.4.2 Habit Discontinuity Habit discontinuity interventions have only been studied within field experiments in pro-environmental behaviors, which capitalize on residential relocation as a naturally occurring context change that offers a “window of opportunity” for intervention (Verplanken, Roy, & Whitmarsh, 2018). For example, a real-world information-based campaign promoting sustainable energy use was more effective among residents who had recently relocated than those who had not (Verplanken & Roy, 2016). In these circumstances, old habits erode alongside a strengthening of new habits (Walker et al., 2015).

Langendam, 2006). Employees who were given a personal recycling box to place near their desks, or who formed plans of when, where, and how they would recycle their waste (i.e., implementation intentions; Gollwitzer, 1999), recycled more paper and plastic up to two months after the intervention than those who did not. Although habit was not measured, the authors inferred that observed declines in waste dustbin use and increased recycling reflected direct substitution of a dustbin use habit for recycling habits.

41.4.5 Limitations and Future Directions

To our knowledge, interventions have adopted inhibition techniques only to aid habit substitution. Observational evidence suggests that vigilant monitoring – that is, watching for habit cues and mindfully inhibiting responses to them – offers a potent method for blocking habitual responding. A diary study of students’ attempts to disrupt mundane unwanted habits (e.g., procrastination, screen time) suggested that vigilant monitoring was more effective than temporarily avoiding cues or self-distraction (Quinn et al., 2010). Vigilant monitoring requires awareness of the unwanted behavior, the cues to that behavior, and the contingencies between these.

Habit-formation interventions have typically been evaluated in comparison to no-treatment or usual care (Beeken et al., 2017; Lally et al., 2008). This limits conclusions around whether techniques especially chosen to target habit are effective. It remains unclear whether interventions must promote context-dependent repetition for habit to form or whether they should target habit associations, or the expression of habitual responses, to disrupt habit. Recipients of non-habit advice could spontaneously adopt context-dependent repetition as a change method (White et al., 2017). Trials of habit-formation interventions should assess whether the strength of existing “bad” habits that conflict with new “good” habits will determine the strength or longevity of new habits and the probability of relapse. This would help to identify whether, within real-world contexts, focusing on underlying associations is indeed more effective than targeting the manifestation of habit in behavior as theory predicts. More fundamentally, intervention trials must adopt longer-term follow-ups to test predictions that habit-based change persists over time (Gardner & Rebar, 2019).

41.4.4 Habit Substitution

41.5 Step-by-Step Guide

One intervention sought to replace workplace dustbin use with recycling (Holland, Aarts, &

Figure 41.3 sets out the different forms of habitbased change, with guidance on which of these to

41.4.3 Habit Inhibition

Habit Interventions

603

Identify Target Behavior Uptake of wanted behavior, or decrease of unwanted behavior

Wanted Behavior (Habit Formation)

Unwanted Behavior (Habit Disruption)

Tips: Should be – Intrinsically valued – Simple – Realistic/Achievable If not, consider revising

Identify Context/Cues - e.g., when, where, with whom, prior action

Can the context/cues be avoided? Can the context/cues be avoided?

Yes Repeat Target Behavioral Context (Habit Formation)

No

If problems, consider revising target behavior or context

Formation of New Habit

Pursue Alternative Behavior (Habit Substitution)

Avoid Cues (Habit Discontinuity)

Inhibit Unwanted Response (Habit Inhibition)

Discontinuation of Unwanted Behavior

Overwriting of Old Habit Associations

Figure 41.3 A decision tree to aid selection of habit change strategies (Gardner, 2019)

use and how to use them according to the purpose of the behavior change attempt and the likely outcome of adopting each strategy.

41.5.1 Typical Means of Delivery Although most habit-based interventions to date have been delivered by practitioners (Gardner & Rebar, 2019), they need not be delivered by a particular authority or via a certain modality. It is possible to offer self-administrable habit interventions –

for example, via text- or video-based communication, interactive online tools or mental association training regimens in which cue-behavior associations are developed, strengthened, or weakened (De Houwer, Thomas, & Baeyens, 2001).

41.5.2 Target Audience and Behaviors People do not need to be motivated to change to benefit from habit-based interventions that curtail opportunities to act. For example,

604

BENJAMIN GARDNER, AMANDA L. REBAR, AND PHILLIPPA LALLY

choice-architecture manipulations – that is, “nudges” (Thaler & Sunstein, 2008), such as painting staircases to resemble piano keys – can make or break habits if they increase or decrease the frequency with which a behavior is repeated in a given context, regardless of individuals’ awareness or desire to change. At the population-level, marketers seek to change consumer behaviors through repeatedly presenting cue-behavior pairings in concert with their brand, training nonconscious mental associations (Rebar, 2017).

Promoting voluntary habit-based change, however, requires that people are motivated to want to change. If people are unwilling to make changes, they will not repeat a behavior sufficiently for habit to form, nor will they deploy the willpower required to disrupt existing habit impulses (see also Chapter 40, this volume). Therefore, either voluntary habit-based interventions should be delivered to audiences who are already motivated and able to change, or habit-specific techniques (see Sidebar 41.1) should be deployed as an

Sidebar 41.1 Integrating habit-based techniques into the behavior change technique taxonomy version 1 (BCTTv1)

Habit formation. The behavior change technique taxonomy version 1 (BCTTv1) defines “habit formation” as a discrete technique involving repetition in a consistent context (Michie et al., 2013). This technique is referred to here as “context-dependent repetition,” which allows for the possibility that additional techniques may strengthen habit (Gardner & Rebar, 2019). Motivation, and the ability to initiate and repeat a behavior, are prerequisites for context-dependent repetition (Figure 41.1), and so any technique that prompts behavior can contribute to habit formation. Some techniques may nonetheless be particularly conducive to developing cue-responses. The central BCT involved in habit formation is context-dependent repetition, which reinforces the mental cue-behavior associations that regulate habitual behavior. Techniques drawn from the natural consequences (e.g., information about consequences), comparison of behavior (e.g., social comparison, information about others’ approval), identity (e.g. incompatible beliefs), reward and threat (e.g., social rewards), and self-belief (e.g., verbal persuasion about capability) clusters may foster the motivation required to initiate and maintain behavior as habit forms. Techniques relating to goals and planning (e.g., goal setting, action planning, reviewing goals), feedback and monitoring (e.g., self-monitoring behavior), shaping knowledge (e.g., instruction on how to perform the behavior), associations (e.g., using prompts and cues), repetition and substitution (e.g., behavioral practice), and antecedents (e.g., restructuring the physical environment) are relevant to the action control processes implicated in repeating a behavior so habit may form. Habit discontinuity and inhibition. The techniques information about antecedents, feedback on behavior, or self-monitoring behavior can support people in identifying cues that trigger their own habitual responses. Avoiding or reducing exposure to cues for the behavior, restructuring the physical or social environments (to reduce cue exposure), and distraction can all obstruct enactment of habitual cue-responses. Action planning and conserving mental resources respectively aid habit disruption by reminding people to activate the willpower and self-control to inhibit their actions at the appropriate moment.

Habit Interventions

605

Habit substitution. Within the BCTTv1, the technique habit reversal refers to the rehearsal and repetition of a new behavior to replace an old habit and so is core to habit substitution. Substitution is, by definition, facilitated by a combination of techniques that support the formation of new habits and those that disrupt old habits. Action planning may also be particularly important for specifying new responses to perform in place of old ones. The techniques identified in Table 41.1 are based on the BCTTv1 only, which applies across behavioral domains and supersedes previous behavior-specific taxonomies (e.g., CALO-RE; Michie et al., 2011). One technique identified in the BCTTv1 is labeled “habit formation”; the term “context-dependent repetition” is, however, more descriptive (Gardner & Rebar, 2019; see techniques labeled “a” in Table 41.1). It is important to note that habit formation, and so substitution, requires (context-consistent) performance of a behavior. All BCTs that focus on uptake of a behavior, as outlined in other chapters in this volume, are thus potentially conducive to habit formation and substitution and an extensive list is not feasible within this chapter (e.g., see techniques subscripted “b” in Table 41.1). Table 41.1 Techniques for changing habits from the behavior change technique taxonomy v1 (BCTTv1; Michie et al., 2013) with related techniques and descriptions Cluster and Technique

Primary/ Closely Related Description Habit formation

Primary Cluster: 8. Repetition and substitution Technique: 8.3. Contextdependent repetition/ habit formationa Cluster/Technique: Closely Related Various

Prompt rehearsal and repetition of the behavior in the same context repeatedly so that the context elicits the behavior. Any BCT that prompts enactment of a wanted behavior can facilitate habit formation.b

Habit disruption: habit discontinuity

Cluster: 12. Antecedents Primary Technique: 12.3. Avoiding exposure to cues for the behavior Cluster: 4. Shaping Closely Related knowledge Technique: 4.2. Information about antecedents

Avoid exposure to specific contextual cues for the behavior.

Provide information about antecedents that reliably predict performance of the behavior. Where discontinuity relies on voluntarily avoiding cues to unwanted actions, identifying these cues is a prerequisite. Continued

606

BENJAMIN GARDNER, AMANDA L. REBAR, AND PHILLIPPA LALLY

Table 41.1 (cont.)

Cluster and Technique

Primary/ Closely Related Description

Cluster: 12. Antecedents Closely Related Technique: 12.1. Restructuring the physical environment

Cluster: 12. Antecedents Closely Related Technique: 12.2. Restructuring the social environment

Change or advise to change the physical environment in order to create barriers to the unwanted behavior. This is important for voluntary discontinuation of habitual actions prompted by physical environmental cues. Change or advise to change the social environment in order to create barriers to the unwanted behavior. This is important for voluntary discontinuation of habitual actions prompted by social cues.

Habit disruption: Habit inhibition Cluster: 12. Antecedents Primary Technique: 12.4. Distraction Cluster: 1. Goals & planning Technique: 1.4. Action planning Cluster 11. Regulation. Technique: Conserving mental resources

Closely Related

Closely Related

Arrange to use an alternative focus for attention to avoid performing the unwanted behavior when the cue is encountered. Prompt detailed planning of an inhibitory response to be performed in response to anticipated cues to the unwanted behavior. Minimize demands on mental resources to allow for attention to be paid to identifying the trigger and inhibiting the unwanted behavior in response to it.

Habit formation and disruption: Habit substitution Cluster: 8. Repetition Primary and substitution. Technique: 8.4. Habit reversal Primary Cluster: 8. Repetition and substitution Technique: 8.3. Contextdependent repetition/ habit formationa

Prompt rehearsal and repetition of an alternative behavior to replace an unwanted habitual behavior. Prompt rehearsal and repetition of the behavior in the same context repeatedly so that the context elicits the behavior. Continued

Habit Interventions

607

Table 41.1 (cont.)

Cluster and Technique

Primary/ Closely Related Description

Cluster: 4. Shaping knowledge Technique: 4.2. Information about antecedents

Closely Related

Cluster: 1. Goals & planning Technique: 1.4. Action planning Cluster 11. Regulation. Technique: Conserving mental resources

Closely Related

Cluster/Technique: Variousb

Closely Related

Closely Related

Provide information about antecedents that reliably predict performance of the unwanted behavior. This enables performance of the new behavior in direct substitution for the existing habitual behavior. Prompt detailed planning of the desired new behavior to be performed in response to cues to the unwanted behavior. Minimize demands on mental resources to allow for attention to be paid to performing the new behavior, and inhibiting the unwanted behavior, in response to cues to the unwanted behavior. Any BCT that prompts enactment of a wanted behavior can facilitate habit substitution.b

Note. aHabit formation is also described as context-dependent repetition; b Techniques that focus on enactment of a behavior are potentially conducive to habit formation and substitution.

adjunct to strategies that seek to motivate or foster the action control required to enact change (Gardner & Lally, 2018). In theory, any behavior repeated frequently in a consistent context can become habitual and, conversely, avoiding situational cues or performing new actions can lead to the disruption of any habitual behavior. In practice, however, all behaviors have properties that influence the likelihood that they will become, or cease to be, habitual. Tentative evidence, for example, suggests that behaviors that are easier to perform, or that include fewer “steps” to be completed, may

become habitual more quickly than more complex actions (Lally et al., 2010; Mullan & Novoradovskaya, 2018). Practitioners should consider whether skill training might be needed to supplement or precede habit-formation efforts. Alternatively, interventions might promote simpler behaviors or a graded approach to change – for example, encouraging more light activity rather than 150 minutes of moderate physical activity per week for inactive people (White et al., 2017). Growing evidence suggests that behaviors that are more inherently rewarding may become habitual more quickly or strongly

608

BENJAMIN GARDNER, AMANDA L. REBAR, AND PHILLIPPA LALLY

(Radel et al., 2017). Conversely, although discontinuity attempts are unlikely to be affected by characteristics of a behavior, habit inhibition or substitution efforts may be more challenging when the existing habitual action is easier, or more intrinsically rewarding, than not acting or adopting an alternative. The habitual nature of unwanted behaviors can pose the greatest threat to change because well-learned actions are inherently easier to perform than alternatives (Danner, Aarts, & de Vries, 2008).

41.5.3 Enabling Factors Properties of the cue, the behavior, and the person forming or disrupting the habit can determine the extent of habit change (Gardner & Lally, 2018). Habit is most likely to form when people set specific plans of what will be done in what context (Fleig et al., 2013). Plans should ideally specify cues that are frequently and reliably encountered and most readily perceived. Event-based cues, such as preceding actions (e.g., “after a shower”), may be more salient than time-based cues (e.g., “8:00 a.m.”) which require conscious monitoring (McDaniel & Einstein, 2000). Performance of the desired behavior must follow exposure to the cue (e.g., “after brushing my teeth”), not precede it (e.g., “before I go to bed”). Plans should specify behaviors that can realistically be enacted frequently, are of minimal complexity, and are of inherent value to the person pursuing them. Intervention recipients should be supported to identify everyday contexts and actions that meet such requirements for them. Similarly, habit change may be facilitated in people who are driven to pursue an action by interest or enjoyment (i.e., intrinsic motivation) rather than to satisfy external demands (extrinsic motivation; Ryan & Deci, 2000; see also Chapter 8, this volume). Change attempts seen by the individual as arising more to meet the wishes of an external source (e.g., the practitioner) may be less likely to succeed. Intervention recipients should be

encouraged to identify a behavior of prior value to them or that they will enjoy. People who are better able to regulate their behavior to achieve their goals may be more able to enact planned actions at opportune moments and potentially to overcome unwanted conflicting habitual responses (De Ridder et al., 2012). Alternatively, people might be trained in the skills required to identify contexts conducive to wanted behaviors, or those that support unwanted behaviors, and summon the willpower needed to act in those contexts (see also Chapters 11 and 40, this volume). Habit discontinuity depends on naturally occurring contextual changes that disrupt exposure to cues to existing habits. When such changes occur, people become more receptive to persuasionbased strategies designed to encourage new actions (Verplanken et al., 2018). Habit discontinuity itself therefore enables formation of new habits. An example intervention demonstrating discontinuity is presented in Sidebar 41.2.

41.5.4 Training and Skills Required Where involving the provision of advice to intervention recipients, delivering a habit-based intervention requires minimal training beyond that required to effectively support any behavior change. Educating people about how habit differs from frequent behavior may help them to more effectively focus on the underlying habit process rather than the behavioral outcome. This, in turn, should make behavior change efforts more effective and potentially long-lasting. Core habit concepts can be communicated in an accessible way in reference to everyday examples of habitual actions; people generally understand, for example, what it means for a behavior to become automatic, like “brushing your teeth” or “saying your prayers” (Matei et al., 2015).

41.5.5 Intensiveness Behaviors need not be repeated daily; consistency of cue-dependent performance, rather than absolute

Habit Interventions

609

Sidebar 41.2 Discontinuing habitual car use

Walker et al.’s (2015) field study observed the effects of a workplace relocation on workers’ travel mode choices. Measures of habitual use of several travel modes were taken nineteen months prior to the relocation of offices from one town to another and one and four weeks after the move. Employees most commonly switched from driving to commuting by train. Moreover, habit scores revealed that, for the mode most commonly used prior to the move, habit decreased rapidly immediately after the move. This was mirrored by a sharp increase in habit strength for the new mode and both changes were sustained four weeks after the move. This suggests that the intervention operated via not only discontinuation but also habit substitution mechanisms. This study attests to the power of context change as a habit disruption strategy. It also suggests that discontinuity may have the potential to not only inhibit habitual behavior but also degrade old habits, though this requires rigorous testing in more tightly controlled settings. Walker and colleagues’ study demonstrates how intervention developers might capitalize on naturally occurring context changes afforded by major life events (e.g., moving house or workplace). Although less wellstudied, discontinuity effects should in theory be achievable at the individual level, if people can identify and consistently avoid triggers to unwanted habits (e.g., not buying unhealthy snacks to prevent habitual snacking while watching TV).

frequency, generates habit associations (Aunger, 2007; Klöckner & Oppedal, 2011). Habit forms asymptotically, with initial growth slowing until a plateau is reached (Figure 41.4; see Sidebar 41.3). It is particularly important that people are supported during the early stages of habit-formation attempts, as this is both the point at which quickest gains can be made in habit strength and the point at which, should initial experiences of an action be deemed unsatisfactory, people are most likely to disengage from a change attempt. At these early stages, habit support should not only be most intensive but also be responsive to initial experiences. People should ideally be supported to refine any goals they have set and methods they have chosen to pursue those goals, if such goals or strategies appear to be unfeasible or ineffective (see also Chapter 38, this volume). It is important that the behavior becomes self-sustaining rather than an automatic response to forms of external support, such as text message reminders or monetary rewards (see also Chapter

36, this volume). As habit forms and the behavior becomes routinized and easier to perform, external support can be gradually withdrawn. Little is known about how long is required for habits to break, or how intensive habit-disruption efforts must be, but it will likely vary according to the habit-disruption strategy used. Habit discontinuity, for example, should result in immediate change because ceasing exposure to cues that sustain habitual behaviors should instantaneously discontinue associated habitual responses. Further support may be necessary where discontinuity is only temporary because later reexposure to cues may activate “dormant” habitual responses, prompting relapse into old behaviors (Gardner, 2015). For example, stress-cued habitual nail-biting may cease when relaxing on holiday but reemerge on returning home to everyday stressors (Lally et al., 2011). Where permanent discontinuation is unfeasible, habit inhibition or substitution may provide more sustainable options (Quinn et al., 2010).

610

BENJAMIN GARDNER, AMANDA L. REBAR, AND PHILLIPPA LALLY

Figure 41.4 The habit growth curve

Sidebar 41.3 How are habits formed?

Lally et al.’s (2010) study explored the relationship between repetition of a new action and habit development but has important intervention implications. Ninety-six participants developed a plan to enact a self-chosen behavior in response to a cue that occurs once per day. Each day, over a twelve-week period, they reported whether they had performed the action on the previous day and how automatic (i.e., habitual) it felt. Results suggested that habit formation is best represented by an asymptotic growth curve, whereby early repetitions cause an acceleration in habit strengthening, which dissipates as a plateau is reached (but see Fournier, d’Arripe-Longueville, & Radel, 2017). This suggests there are discrete stages in the habit-formation process, which may require different intervention approaches (Figure 41.3). At the early stages, characterized by rapid growth, people may require most external support to initiate and repeat action so that habit may form. As habit gains decelerate and plateau, the behavior should show characteristics of a habitual response – for example, starting to feel familiar, easier to perform, and self-sustaining – rendering external support less important (e.g., Matei et al., 2015; McGowan et al., 2013).

More intensive and ongoing efforts are likely to be required to support people who are willfully inhibiting habitual responses or, in the case of

substitution, seeking to both inhibit old habitual responses and enact new habits. It seems prudent to provide especially intensive support at the

Habit Interventions

early stages of a substitution attempt, to train and support the individual in not only summoning the motivation and self-control needed to enact desired responses but also suppressing existing, powerful habit impulses.

41.5.6 Evaluation of Fidelity Fidelity is not a concern for habit discontinuity, nor indeed any strategy based on restricting opportunities to act. Where habit change depends on adherence to a prescribed regimen of context-dependent repetition, inhibition, or substitution, perhaps the most important fidelity check is that both those delivering the intervention and intervention recipients understand the distinction between habitual and frequent action. Failing to comprehend this difference could lead to divergence from the core habit-focused components of an intervention, which center on promoting or discouraging context-consistent performance rather than repeated performance per se.

41.5.7 Evaluation of Effectiveness Evaluations should focus on changes in behavior and habit strength. Measuring changes in habit strength provides the opportunity to assess whether habit is a likely mechanism for behavior change (Kliemann et al., 2017). The Self-Report Habit Index (Verplanken & Orbell, 2003), and its various adaptations (Gardner et al., 2012, 2016), are the most commonly used measures, but the validity of self-report measures has been hotly debated (Hagger et al., 2015; Orbell & Verplanken, 2015; Sniehotta & Presseau, 2012). Objective or indirect measures – such as association-based tests, which quantify the speed or accuracy with which habitual responses are elicited – are uniformly more preferable to self-reported measures, which are open to various recall and presentational biases. Non–selfreport measures are, however, impractical in some research settings.

611

Qualitative interviews can assess, if not quantify, habit. People can reflect on the experience of behaviors becoming more or less familiar and routine over time and spontaneously use the language of habit (e.g., routine, automaticity) when doing so (Gardner, Sheals et al., 2014; Lally et al., 2011; Matei et al., 2015).

41.5.8 Typical Materials Needed No special materials are required for habit interventions. However, materials commonly used in habitformation and substitution interventions have focused on reinforcing motivation and self-control (McGowan et al., 2013; see Sidebar 41.4). These have typically taken the form of tips recommending “small changes” to facilitate habit change and “worksheets” on which intervention recipients can detail their habit goals and the contexts in which they will execute them, describe plans to cope with setbacks, and self-monitor progress (Lally et al., 2008; McGowan et al., 2013; White et al., 2017). Habit inhibition or substitution efforts will depend on people modifying their responses to cues, but people may lack awareness of cues or their responsiveness to them. It can therefore be useful to preface inhibition or substitution attempts with a period during which intervention recipients monitor their unwanted actions (i.e., self-monitoring) and potential cues to these (i.e., cue-monitoring; Verhoeven et al., 2014). This may take the form of a diary that records each instance of action and the circumstances in which it occurred, such as preceding behaviors or the presence of others. Reflecting on monitoring diaries may generate insights into triggers to unwanted habitual responses.

41.6 Typical Example of Implementation Here, the stages involved in developing and evaluating a voluntary habit-based intervention are described, as illustrated by an intervention to

612

BENJAMIN GARDNER, AMANDA L. REBAR, AND PHILLIPPA LALLY

Sidebar 41.4 Developing healthy child-feeding habits

One study trialed a habit-formation and substitution intervention promoting healthy child-feeding practices among parents of young children (Gardner, Sheals et al., 2014; McGowan et al., 2013). Parents received a booklet outlining the importance of child dietary quality and habitual feeding practices, meal suggestions, and training in selfregulatory skills (e.g., setting goals, action and coping planning, monitoring feeding and child intake). The intervention promoted three behaviors: serving more fruit and vegetables, healthy rather than unhealthy snacks, and unsweetened rather than sweetened drinks. Habit-formation attempts were staggered: parents pursued one of the three behaviors before also pursuing a second behavior two weeks later and a third two weeks after that. Child intake of vegetables, healthy snacks, and water increased more in the intervention group (N = 58) than in a no-treatment control condition (N = 68) and changes in intake correlated with parental automaticity change for the focal behaviors. Inspection of habit growth curves suggested that parents were able to form and maintain habits for multiple behaviors concurrently and, in interviews, parents reported that the habit-based approach was easy to follow. Inspection of parents’ written goals, however, indicated that many deviated from recommendations by, for example, failing to specify a clear behavior or cue. This points to the potential for loss of fidelity when habit interventions are implemented in the real world.

encourage light activity and discourage sitting among adults aged sixty plus (“On Your Feet to Earn Your Seat” [OYF]; Gardner, Thune-Boyle, et al., 2014). OYF centered on a leaflet emphasizing the importance of movement and limiting sitting in older adulthood, with fifteen tips for integrating activity into normally seated routines. The intervention is comprehensively detailed in Appendix 41.1 (supplemental materials).

41.6.1 Identifying Target Behaviors The fundamental consideration for any habit intervention is whether to build a new habit, disrupt an existing habit, or both. If breaking a “bad” habit, the next consideration should be whether it is feasible and sustainable to frustrate the habit impulse – by purposefully discontinuing cue exposure indefinitely or capitalizing on naturally occurring discontinuation (habit discontinuity) or adopting self-control strategies to suppress the

habit impulse (habit inhibition) – or whether the underlying habit association must be overwritten (habit substitution). In all instances, properties of the target behavior that may or may not lead it to become ingrained or render it difficult to shift must be considered. Behaviors of little appeal or benefit, for example, are unlikely to be adopted. Behaviors were selected for promotion in OYF because they could feasibly be integrated into everyday routines and were minimally disruptive. Explanations of its benefits, and “handy hints” to simplify performance, were provided for each behavior. For example, one tip (“stand up or walk around during breaks between TV programs”) was justified as helping to “stop your joints seizing up.” Elsewhere, it was suggested that, “to increase bone density and reduce the likelihood of falls,” people perform heel rises while washing up, though it should be emphasized that this “can be done anywhere when you are standing and waiting.”

Habit Interventions

41.6.2 Identifying Cues A balance must be struck between identifying specific cues that are highly relevant to individual intervention recipients and less potent cues of relevance to a broader range of recipients. Habitdisruption interventions might alternatively support people to identify their own personal cues, for example by using a monitoring diary. In OYF, everyday contexts in which people typically sit but could feasibly stand or be active were targeted – for example, while watching TV, washing dishes, or boiling the kettle. However, some cues lacked acceptability; some people were unwilling to perform heel rises while waiting in a public queue, for fear of embarrassment (“you’ll get strange looks; they’ll think, ‘what’s the matter with him?’” said one participant; Matei et al., 2015, p. 12).

41.6.3 Reinforcing Motivation and Self-Control to Persistently Respond to the Cue Habit formation, inhibition, and substitution require modification of responses to selected cues. The key challenge is to maximize the likelihood that people will retain and prioritize the motivation and self-control to act when the cue is encountered. Two strategies commonly employed for this purpose are to emphasize the importance of the new behavior and to train people in the skills needed to act in the chosen setting. Action plans specifying intended behavioral responses to cues can be coupled with “coping plans” specifying how unintended behavioral responses with the potential to derail the desired action will be overcome. The more detailed these plans, the more likely they will translate into action (Locke & Latham, 2002). It can be helpful to review habit plans prior to implementation. People tend to be overly optimistic about their behavior change capabilities but failing to achieve desired change can prompt disengagement (Carver & Scheier, 1982).

613

The OYF intervention leaflet was supplemented with a self-monitoring “tick-sheet,” on which participants could record their adherence to each of the tips; this aimed to provide positive feedback on progress to sustain the motivation required for further repetition and so habit formation. The leaflet also explained how and why to form habits, emphasizing that, as automaticity develops, the newly adopted behavior should become easier to perform.

41.6.4 Implementing, Evaluating, and Interpreting The OYF intervention was compared to a nonhabit–based activity-promotion factsheet in a small randomized trial (White et al., 2017). Intervention recipients (N = 45) reduced selfreported sitting time and sitting habit and increased self-reported activity and activity habits but so did the control group (N = 46). The contextfree habit measures (e.g., “[Sitting/physical activity] is something I do without thinking”) meant that changes could not clearly be attributed to greater habitual performance of the recommended activities in the recommended contexts. However, qualitative interviews suggested that habit formation was a causal mechanism for change (“I am [now] doing these things throughout the day, but not consciously”; White et al., 2017, p. 8). Methodological problems may have obscured any true advantage in intervention effectiveness. For example, the pilot sample was highly active at baseline, so unrepresentative of the highly sedentary population most likely to benefit from such intervention. Nonetheless, evidence pointed to the acceptability of self-administered and simple habit-based advice among intervention recipients and its potential to change behavior.

41.7 Concluding Remarks Habitual behaviors are thought to persist over time, making habit formation an attractive method

614

BENJAMIN GARDNER, AMANDA L. REBAR, AND PHILLIPPA LALLY

for promoting long-term maintenance of new behaviors and habit-disruption methods promising for lasting cessation of unwanted behaviors. Conceptually, habit-based behavior change is simple: forming a new habit requires consistently performing a wanted action in a given setting to form new associations, and discontinuing a habitual behavior is facilitated by either avoiding associated cues or forming new responses to those cues. Yet, in practice, habit-based change is often difficult, requiring continued motivation and willpower to succeed. Studies suggest that the effort that must be expended for habit to change is, however, a sound investment. Recipients of habit-based interventions typically report the approach to be an acceptable way of changing their everyday routines and make positive changes to their behavior.

References Aunger, R. (2007). Tooth brushing as routine behaviour. International Dental Journal, 57, 364– 376. https://doi.org/10.1111/j.1875-595X.2007 .tb00163.x Beeken, R. J., Croker, H., Morris, S. et al. (2012). Study protocol for the 10 Top Tips (10TT) Trial: Randomised controlled trial of habit-based advice for weight control in general practice. BMC Public Health, 12, 667. https://doi.org/10.1186/14712458-12-667 Beeken, R. J., Leurent, B., Vickerstaff, V. et al. (2017). A brief intervention for weight control based on habit-formation theory delivered through primary care: Results from a randomised controlled trial. International Journal of Obesity, 41, 246–254. https://doi.org/10.1038/ijo.2016.206 Carver, C. S., & Scheier, M. F. (1982). Control theory: A useful conceptual framework for personality– social, clinical, and health psychology. Psychological Bulletin, 92, 111–135. https://doi .org/10.1037/0033-2909.92.1.111 Danner, U. N., Aarts, H., & de Vries, N. K. (2008). Habit vs. intention in the prediction of future behaviour: The role of frequency, context stability and mental accessibility of past behaviour. British

Journal of Social Psychology, 47, 245–265. https://doi.org/10.1348/014466607X230876 De Houwer, J., Thomas, S., & Baeyens, F. (2001). Associative learning of likes and dislikes: A review of 20 years of research on human evaluative conditioning. Psychological Bulletin, 127, 853–869. http://dx.doi.org/10.1037/00332909.127.6.853 De Ridder, D. T. D., Lensvelt-Mulders, G., Finkenauer, C., Stok, F. M., & Baumeister, R. F. (2012). Taking stock of self-control: A meta-analysis of how trait self-control relates to a wide range of behaviors. Personality and Social Psychology Review, 16, 76–99. https://doi.org/10.1177/1088868311418749 Fleig, L., Pomp, S., Parschau, L. et al. (2013). From intentions via planning and behavior to physical exercise habits. Psychology of Sport and Exercise, 14, 632–639. https://doi.org/10.1016/j.psychsport .2013.03.006 Fournier, M., d’Arripe-Longueville, F., & Radel, R. (2017). Testing the effect of text messaging cues to promote physical activity habits: A worksite-based exploratory intervention. Scandinavian Journal of Medicine and Science in Sports, 27, 1157–1165. https://doi.org/ 10.1111/sms.12730 Gardner, B. (2015). A review and analysis of the use of “habit” in understanding, predicting and influencing health-related behaviour. Health Psychology Review, 9, 277–295. https://doi.org/ 10.1080/17437199.2013.876238 Gardner, B. (2019). Habit interventions: A guide to using habit to change behaviour. OSF. https://doi .org/10.17605/OSF.IO/SVHXC Gardner, B., & Lally, P. (2018). Modelling habit formation and its determinants. In B. Verplanken (Ed.), The Psychology of Habit (pp. 207–230). Cham: Springer. Gardner, B., Lally, P., & Wardle, J. (2012). Making health habitual: The psychology of “habitformation” and general practice. British Journal of General Practice, 62, 664–666. https://doi.org/ 10.3399/bjgp12X659466 Gardner, B., Phillips, L., & Judah, G. (2016). Habitual instigation and habitual execution: Definition, measurement, and effects on behaviour frequency. British Journal of Health Psychology, 21, 613– 630. https://doi.org/10.1111/bjhp.12189

Habit Interventions

Gardner, B., & Rebar, A. (2019). Habit and behaviour change. Oxford Research Encyclopedia. https://doi .org/10.1093/acrefore/9780190236557.013.129 Gardner, B., Sheals, K., Wardle, J., & McGowan, L. (2014). Putting habit into practice, and practice into habit: A process evaluation and exploration of the acceptability of a habit-based dietary behaviour change intervention. International Journal of Behavioral Nutrition and Physical Activity, 11, 135–147. https://doi.org/10.1186/ s12966-014-0135-7 Gardner, B., Thune-Boyle, I., Iliffe, S. et al. (2014). “On Your Feet to Earn Your Seat”: A habit-based intervention to reduce sedentary behaviour in older adults: Study protocol for a randomized controlled trial. Trials, 15. https://doi.org/10.1186/ 1745-6215-15-368 Gollwitzer, P. M. (1999). Implementation intentions: Strong effects of simple plans. American Psychologist, 54, 493–503. https://dx.doi.org/ 10.1037/0003-066X.54.7.493 Hagger, M. S., Rebar, A. L., Mullan, B., Lipp, O. V., & Chatzisarantis, N. L. D. (2015). The subjective experience of habit captured by self-report indexes may lead to inaccuracies in the measurement of habitual action. Health Psychology Review, 9, 296–302. https://doi.org/ 10.1080/17437199.2014.959728 Hamilton, K., Fraser, E., & Hannan, T. (2019). Habitbased workplace physical activity intervention: A pilot study. Occupational Medicine, 69, 471–474. https://doi.org/10.1093/occmed/kqz119 Holland, R. W., Aarts, H., & Langendam, D. (2006). Breaking and creating habits on the working floor: A field-experiment on the power of implementation intentions. Journal of Experimental Social Psychology, 42, 776–783. https://doi.org/10.1016/j.jesp.2005.11.006 Kliemann, N., Vickerstaff, V., Croker, H., Johnson, F., Nazareth, I., & Beeken, R. J. (2017). The role of self-regulatory skills and automaticity on the effectiveness of a brief weight loss habit-based intervention: Secondary analysis of the 10 top tips randomised trial. International Journal of Behavioral Nutrition and Physical Activity, 14, 119. https://doi.org/10.1186/s12966-017-0578-8

615

Klöckner, C. A., & Oppedal, I. O. (2011). General vs. domain specific recycling behaviour: Applying a multilevel comprehensive action determination model to recycling in Norwegian student homes. Resources, Conservation and Recycling, 55, 463–471. https://doi.org/ 10.1016/j.resconrec.2010.12.009 Lally, P., Chipperfield, A., & Wardle, J. (2008). Healthy habits: Efficacy of simple advice on weight control based on a habit-formation model. International Journal of Obesity, 32, 700–707. https://doi.org/10.1038/sj.ijo.0803771 Lally, P., & Gardner, B. (2013). Promoting habit formation. Health Psychology Review, 7(Suppl 1), S137–S158. http://dx.doi.org/10.1080/17437199 .2011.603640 Lally, P., Van Jaarsveld, C. H., Potts, H. W., & Wardle, J. (2010). How are habits formed: Modelling habit formation in the real world. European Journal of Social Psychology, 40, 998–1009. https://doi.org/ 10.1002/ejsp.674 Lally, P., Wardle, J., & Gardner, B. (2011). Experiences of habit formation: A qualitative study. Psychology, Health and Medicine, 16, 484–489. https://doi.org/10.1080/13548506.2011.555774 Locke, E., & Latham, G. (2002). Building a practically useful theory of goal setting and task motivation: A 35-year odyssey. American Psychologist, 57, 705–717. http://dx.doi.org/10.1037/0003066X.57.9.705 Matei, R., Thuné-Boyle, I., Hamer, M. et al. (2015). Acceptability of a theory-based sedentary behaviour reduction intervention for older adults (“On Your Feet to Earn Your Seat”). BMC Public Health, 15, 606. https://doi.org/10.1186/s12889015-1921-0 McDaniel, M. A., & Einstein, G. O. (2000). Strategic and automatic processes in prospective memory retrieval: A multiprocess framework. Applied Cognitive Psychology, 14, S127–S144. https://doi .org/10.1002/acp.775 McGowan, L., Cooke, L. J., Gardner, B., Beeken, R. J., Croker, H., & Wardle, J. (2013). Healthy feeding habits: Efficacy results from a cluster-randomized, controlled exploratory trial of a novel, habit-based intervention with parents. The American Journal

616

BENJAMIN GARDNER, AMANDA L. REBAR, AND PHILLIPPA LALLY

of Clinical Nutrition, 98, 769–777. https://doi.org/ 10.3945/ajcn.112.052159 Michie, S., Ashford, S., Sniehotta, F. F., Dombrowski, S. U., Bishop, A., & French, D. P. (2011). A refined taxonomy of behaviour change techniques to help people change their physical activity and healthy eating behaviours: The CALO-RE taxonomy. Psychology and Health, 26, 1479–1498. http://dx.doi.org/10.1080/08870446.2010.540664 Michie, S., Richardson, M., Johnston, M. et al. (2013). The behavior change technique taxonomy (v1) of 93 hierarchically clustered techniques: Building an international consensus for the reporting of behavior change interventions. Annals of Behavioral Medicine, 46, 81–95. http://dx.doi .org/10.1007/s12160-013-9486-6 Moors, A., & De Houwer, J. (2006). Automaticity: A theoretical and conceptual analysis. Psychological Bulletin, 132, 297–326. http://dx.doi.org/10.1037/ 0033-2909.132.2.297 Mullan, B., & Novoradovskaya, L. (2018). Habit mechanisms and behavioural complexity. In B. Verplanken (Ed.), The Psychology of Habit (pp. 70–90). Cham: Springer. Orbell, S., & Verplanken, B. (2015). The strength of habit. Health Psychology Review, 9, 311–317. https://doi.org/10.1080/17437199.2014.992031 Quinn, J. M., Pascoe, A., Wood, W., & Neal, D. T. (2010). Can’t control yourself? Monitor those bad habits. Personality and Social Psychology Bulletin, 36, 499–511. https://doi.org/10.1177/ 0146167209360665 Radel, R., Pelletier, L., Pjevac, D., & Cheval, B. (2017). The links between self-determined motivations and behavioral automaticity in a variety of real-life behaviors. Motivation and Emotion, 41, 443–454. http://dx.doi.org/10.1007/s11031-017-9618-6 Rebar, A. L. (2017). Automatic regulation used in sport and exercise research. In O. Braddick (Ed.), Oxford Research Encyclopedia of Psychology. New York: Oxford University Press. https://doi .org/10.1093/acrefore/9780190236557.013.231 Ryan, R. M., & Deci, E. L. (2000). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American Psychologist, 55, 68–78. http://dx.doi.org/ 10.1037/0003-066X.55.1.68

Sniehotta, F. F., & Presseau, J. (2012). The habitual use of the self-report habit index. Annals of Behavioral Medicine, 43, 139–140. https://doi .org/10.1007/s12160-011-9305-x. Thaler, R. H., & Sunstein, C. R. (2008). Nudge: Improving Decisions About Health, Wealth and Happiness. London: Yale University Press. Tsai, A. G., & Wadden, T. A. (2005). An evaluation of major commercial weight loss programs in the United States. Annals of Internal Medicine, 142, 56–66. https://doi.org/10.7326/0003-4819-142-1200501040-00012 Verhoeven, A. A., Adriaanse, M. A., de Vet, E., Fennis, B. M., & De Ridder, D. T. (2014). Identifying the “if” for “if-then” plans: Combining implementation intentions with cue-monitoring targeting unhealthy snacking behaviour. Psychology and Health, 29, 1476–1492. https:// doi.org/10.1080/08870446.2014.950658 Verplanken, B., & Orbell, S. (2003). Reflections on past behavior: A self-report index of habit strength. Journal of Applied Social Psychology, 33, 1313– 1330. https://doi.org/10.1111/j.1559-1816.2003 .tb01951.x Verplanken, B., Roy, D., & Whitmarsh, L. (2018). Cracks in the wall: Habit discontinuities as vehicles for behaviour change. In B. Verplanken (Ed.), The Psychology of Habit (pp. 189–206). Cham: Springer. Verplanken, B., & Roy, D. (2016). Empowering interventions to promote sustainable lifestyles: Testing the habit discontinuity hypothesis in a field experiment. Journal of Environmental Psychology, 45, 127–134. https://doi.org/10.1016/ j.jenvp.2015.11.008 Walker, I., Thomas, G. O., & Verplanken, B. (2015). Old habits die hard: Travel habit formation and decay during an office relocation. Environment and Behavior, 47, 1089–1106. https://doi.org/ 10.1177/0013916514549619 West, R., & Brown, J. (2013). Theory of Addiction (2nd ed.). Oxford: Wiley-Blackwell. White, I., Smith, L., Aggio, D. et al. (2017). On your feet to earn your seat: Pilot RCT of a theorybased sedentary behaviour reduction intervention for older adults. Pilot and Feasibility Studies, 3, 23.

42 Economic and Behavioral Economic Approaches to Behavior Change Paul M. Brown, Linda D. Cameron, Martin Wilkinson, and Denise Taylor

Practical Summary This chapter discusses economic and behavioral economic approaches to behavior change. Economics advocates four traditional ways to change behavior – provide information, change the incentives, change the prices, or impose regulations. A newer approach draws on advances in behavioral economics and considers common decision errors and resulting biases in decision-making. The chapter discusses “nudges” as an example of an approach to behavior change informed by behavioral economics, providing examples of the decision biases and the nudges that can help overcome them. Four steps to developing a nudge are described: examining the choice environment to identify decision points, ensuring the individual wants to engage in the behavior being promoted by the nudge, designing and developing the nudge, and evaluating the effectiveness and sustainability of the nudge. The chapter concludes with a discussion of the likely impact of nudges on the field of behavior change.

42.1 Introduction Economics shares with psychology and other social science fields the common goal of understanding individual decision-making in order to change behavior. While much of the early work by economists studying individual decisionmaking focused on markets and how prices influence decisions, economists since the 1960s have expanded their interests to human behavior across all contexts (Becker & Becker, 2009). As a result, economic analysis now extends to personal decisions outside traditional market settings. Whereas psychologists can draw on multiple theories to explain behavior in a given context, economists tend to use one theoretical model to explain behavior across all decision contexts. This model is called expected utility theory (Von Neumann & Morgenstern, 1953) and, while it has also been

influential in informing theory in psychology (see Takemura, 2014), it forms the basis for the majority of neoclassical (or standard) economic analyses. The central tenets of expected utility theory are that, for any given situation, individuals know and can rank their preferences over alternatives (e.g., they prefer A over B) and that these preference orderings are complete and transitive (e.g., if A is preferred over B and B is preferred over C, then A is preferred over C). Agents need not have perfect foresight or knowledge but are assumed to be able to use the available information, including probabilities of events and potential costs and benefits, to choose the action that is best for them. The primary criticism of expected utility theory is that it does not provide an accurate description of the https://doi.org/10.1017/9781108677318.042

618

PAUL M. BROWN, LINDA D. CAMERON, MARTIN WILKINSON, AND DENISE TAYLOR

way people actually make decisions (Tversky, 1975; see Sidebar 42.1). Until recently, this criticism did little to dent the reliance on expected utility theory, which remains the basis for most economic attempts to model individual decision-making (see Chapter 36, this volume). However, “behavioral economics” has integrated insights from psychology into economic models of decision-making. Economists now participate with psychologists, sociologists, neuroscientists, and others in governmental “behavioral insights” units on the development of policy to change behavior (OECD, 2019). This blossoming interest in behavioral economics can arguably be traced to the publication of Nudge (Thaler & Sunstein, 2008). The primary aim of Nudge was to apply the principles of economics and psychology to behavior change and public policy in ways that did not force people to change (“libertarian paternalism”). It advocated for the integration of psychological evidence on how people actually make decisions into the framework of rational decision-making to design interventions and public policy. The purpose of this chapter is to provide an overview of neoclassical and behavioral economic interventions aimed at behavior change. The discussion of behavioral economics focuses primarily on nudges because, while nudges are used by researchers from many disciplines, they are the most common type of behavior change intervention used by behavioral economists. The chapter provides an overview of nudges from the standpoint of economics while underscoring the limited evidence regarding the effectiveness of many types of nudges or the mechanisms through which they operate. This chapter complements other chapters in this handbook that focus on the use of incentives (Chapter 36, this volume) or changes in the decision environment (Chapter 14, this volume).

42.2 Expected Utility Theory and Behavior Change Friedman and Friedman (1953) gave one of the earliest defenses of the use of mathematical

models such as expected utility theory for decision-making. Their argument is that a theory should not be tested by the realism of its assumptions but rather by the accuracy of its predictions. That is, it does not matter if expected utility theory describes the process by which people actually make decisions, only whether it accurately describes the resulting decisions. Friedman and Friedman’s argument has validity in competitive markets. For example, a business in a perfectly competitive market that produces a good at a higher cost than its competitors will quickly fail. Market pressures force participants to find the most efficient ways of producing and selling their goods. The survivors therefore act “as if” they were adhering to expected utility theory. This claim proves to be more controversial in nonmarket environments where there are no prices or loss of profits. For example, there are no clear economic costs to telling a spouse not to read in bed with the light on; but, the argument goes, as long as individuals can receive feedback on their decisions (e.g., complaints from an unhappy spouse), learn from their mistakes, or are able to hire experts to assist in their decision-making (e.g., a marriage counselor), then their decisions will be “as if” they are adhering to the tenets of expected utility theory. Behavior change interventions guided by expected utility theory start with the assumption that individuals choose the option that maximizes their utility. Any attempt to change their behavior must change the costs and/or benefits of the options. This perspective has two major implications for developing interventions to change behavior. First, economists tend to view preferences, that is, what people do and do not like, as stable. Interventions should not try to change preferences but instead change the costs and benefits to favor the promoted behavior. This is notably different than other approaches that focus on changing preferences (e.g., Chapters 2, 3, 4, and 5, this volume). A second implication of expected utility theory is that behavior change interventions are only warranted in a limited number of cases. These

Economic and Behavioral Economic Interventions

cases usually involve externalities (i.e., actions that impact on other people, such as smoking in a crowded room or attending a meeting when one has the flu), market or coordination failures (e.g., forming a standing army that would benefit all requires cooperation), or attempts to restrict individuals from self-harm (e.g., prohibiting free use of prescription drugs). For other types of behaviors, such as motivating people to conserve energy or eat healthy foods, it is not sufficient that a third party views the desired behavior as “good for the individual.” Instead, behavior change interventions are appropriate only when there are clear impacts on broader society.

42.3 Economic Interventions to Behavior Change To illustrate the traditional economic approaches to behavior change, consider the decision of whether to eat a healthy food such as kale or an unhealthy alternative such as a doughnut. In a simplified version of consumer choice (for a more thorough presentation, see Deaton & Muellbauer, 1980), individuals have a set amount of funds to allocate to kale and doughnuts, receive a certain utility from each, and thus attempt to find the optimal mix of kale and doughnuts that maximizes their expected utility given their budget and the price of each item. There can be cases where the individual’s taste for a good (e.g., kale) is so low or the price so high that they consume none of the product, but generally the utility-maximizing behavior will involve consuming both products (e.g., kale and doughnuts). Suppose society decides that it has an interest in changing people’s behavior away from doughnuts and toward kale. Economic theory then identifies four basic options to change behavior: (1) provide information; (2) incentivize the desired behavior; (3) tax or subsidize the undesired behavior; or (4) impose regulations. Consider first the option of providing information. The decision about what to eat will

619

compel individuals to consider multiple and sometimes conflicting attributes, such as taste, nutrition, and calories. Individuals know what they like (e.g., tasty foods and nutritious foods), understand the attributes of each option, and base their demand for items on the perceived attributes and price of each alternative. For an informational intervention to be successful, it must cause the individual to realize that the value of the attributes of the promoted alternative (kale) is more than they initially believed or the value of the attributes of the other alternative (doughnuts) is worse. If this occurs, then individuals will increase consumption of kale and reduce consumption of doughnuts and there will be a new mix of doughnuts and kale that maximizes personal utility. There are cases where information campaigns have contributed to behavior change, such as in 1963 when the US surgeon general announced that smoking cigarettes causes cancer. Following this announcement and subsequent informational campaigns (along with other factors in a multipronged campaign; Mozaffarian, Hemenway, & Ludwig, 2013), the prevalence of cigarette smoking in the United States fell from 42 percent in 1963 to 18 percent in 2014 (US Department of Health and Human Services, 2014). Yet information campaigns often fail to change behavior because, from a rational decision-making perspective, lasting behavior change can only occur when informational campaigns provide new information that produces lasting changes in how people view the attributes. A second approach to behavior change involves providing financial incentives or disincentives for engaging in the desired behavior (see also Chapter 36, this volume), such as financially rewarding people for eating kale or financially punishing people for eating doughnuts. The rationale for the effectiveness of monetary incentives is straightforward: People have likes and dislikes about food but they also like money, and so the incentives increase the amount of money they have to spend when they consume more of the

620

PAUL M. BROWN, LINDA D. CAMERON, MARTIN WILKINSON, AND DENISE TAYLOR

desired item. There are two challenges with using incentives in this manner. First, providing monetary incentives raises the total amount that individuals have to spend and thus (except under special circumstances) could lead to an increase in the consumption of both desired and undesired foods (e.g., kale and doughnuts; although the consumption of kale might rise more than the consumption of doughnuts). A similar logic applies to placing a fine on eating undesired foods (e.g., doughnuts) or raising the price: These changes would lower income and thus likely lead to a reduction in the consumptions of both options (e.g., doughnuts and kale). The second problem with using incentives is that, in general, the incentives must continue to be provided in order to sustain the behavior. It is possible that the money might entice individuals to permanently change their behavior once the incentive is withdrawn (e.g., people realize how delicious kale tastes) but this would require a change in perception of the attributes of the behavior, as with an informational campaign. Taxes and subsidies operate through the pricing system. The net result, however, is the same as incentives: Individuals will shift away from both the taxed item (e.g., doughnuts) and the untaxed item (e.g., kale). However, taxes and subsidies are typically easier to introduce and sustain than incentives since they only require monitoring the actions of the seller (e.g., collecting a tax on each doughnut sold) rather than monitoring the actions of the buyer (e.g., implementing a fine for each doughnut consumed or payment for each piece of kale eaten). The downside of tax increases and subsidies, however, is that they can be unpopular with the general public and lack wide support (Lim, Slemrod, & Wilking, 2013). Taxes (or incentives) can be effective in changing behavior but they can be a “blunt instrument.” Regulations, such as bans or restrictions, can be implemented more easily and equitably than taxes or incentives. Yet regulations face the problem of needing to be enforced. That is, placing a regulation

on an activity does not change the basic internal calculus with regards to the individual’s actions. If individuals prefer doughnuts over kale, for example, then a law that bans the sale of doughnuts is not likely to be effective unless it appeals to another value (e.g., being a good member of society, adhering to social norms) or there is some enforcement mechanism to ensure adherence.

42.4 Behavioral Economics and Nudges A central thesis of the book Nudge is that small and apparently insignificant details in people’s environment can have major impacts on their behavior (Thaler & Sunstein, 2008). A nudge was defined as “any aspect of the choice architecture that alters people’s behavior in predictable ways without forbidding any options or significantly changing the economic incentives” (p. 6; see also Chapter 14, this volume). As such, nudge interventions do not alter the optimal behavior. Rather, nudge interventions start with the assumption that a desirable option (i.e., eating kale) is consistent with the individual’s preferences but, at the time of making the food choice, the individual ends up selecting the less preferred option (e.g., eating a doughnut). A nudge does not change the available options (e.g., people can still buy doughnuts if desired) and keeps a low-cost, opt-out option should the individual desire (e.g., no penalty is placed on eating doughnuts) but helps individuals achieve the decisions that maximize their utility. From a public policy perspective, part of the attraction of the nudge approach is its emphasis on libertarian paternalism, meaning that the policies try to influence behavior in ways that make people better off without restricting the available choices. For example, the Odenplan subway station in Stockholm, Sweden, transformed its stairway into a piano keyboard in an attempt to induce people to walk up the stairs rather than take the elevator (Bates, 2009). This nudge was implemented under the assumption that is better for the individual (in

Economic and Behavioral Economic Interventions

terms of exercise) and society (in terms of less energy usage) to use the stairs rather than the elevator but it does not tax people for taking the elevator nor restrict their access to the elevator. It is a nudge in that it attempts to influence behavior without limiting choice. It is difficult to explain why this intervention would be successful using expected utility theory. According to expected utility theory, for each behavioral choice (e.g., the choice of stairs versus elevator), individuals have certain and potentially contradictory preferences (e.g., wanting to get to the desired floor as quickly as possible, wanting some exercise, etc.) and they will go through their internal cost-benefit calculations when making the decision. It is not clear how transforming a stairway into a piano keyboard would change this calculation without adding additional preferences (e.g., desire for social approval or liking to walk on a keyboard) to the utility function. Thaler and Sunstein (2008) draw on an alternative explanation. Founded on the work of Kahneman (2011) and others (e.g., Chaiken & Trope, 1999), they identify the roles of two distinct systems for processing information: a fast, automatic system that is highly susceptible to environmental influences (system 1) and a slow, reflective system that considers explicit goals and intentions (system 2). System 1 decisions occur when individuals are faced with time constraints or other attentional pressures and, as a result, they rely on heuristics and simplified decision-making tactics that can lead to “errors,” or deviations of their actual decisions from the decisions that are consistent with expected utility theory. System 2 decision-making reflects more careful and reasoned decisions which, according to Thaler and Sunstein (2008), are the decisions that maximize personal utility (for a more detailed account of dual process approaches, see Chapters 12 and 14, this volume). This distinction of the two systems’ influence on decision-making provides a rationale for intervening on behalf of the individual. For instance, an individual might weigh the pros and cons and

621

sincerely decide to give up eating sugary, highfat foods such as doughnuts and increase intake of vegetables such as kale. They might be resolute in their determination and convinced it is the best option for them. However, when the time comes to actually choose – when they are hungry, under time pressure, and have low willpower – between a quick and satisfying doughnut or chewing on a wad of kale, they go with the former. Afterwards, they wish that they had selected the kale (implying that their preferences have not changed) and vow that next time they will choose the kale. Within this context, a nudge is a device that does not limit their choice (they can still eat a doughnut) but tries to encourage the behavior that they want to engage in (eat kale) but recognizing that, when they actually makes the decision, they might be in a state whereby they are inclined to choose the option that is not in their best interest (i.e., consistent with their stable preferences). The literature on heuristics and biases (e.g., Kahneman & Tversky, 2001) and on dual-process theories (e.g., Kahneman, 2011) provides the foundations for understanding the multitude of ways in which actual decisions can differ from the optimal utility-maximizing behavior. “Errors,” defined in this context as deviations from the behaviors consistent with expected utility theory, can occur in either type of processing (system 1 or 2), but more time and contemplation is available during deliberative, system 2 decision-making and thus errors are less likely to made or, if they are made, can be learned from and influence future decisions. As for environments in which the more automatic decision processes of system 1 take hold, the key to nudge interventions is that, because the heuristics that people use can be predicted, nudge interventions can be developed that overcome these known biases by aligning the automatic and fast decisions with the slow and deliberative decisions. In other words, nudges are ways of intervening to promote the decisions that are consistent with individuals’ preferences as opposed to rash decisions made in the moment.

622

PAUL M. BROWN, LINDA D. CAMERON, MARTIN WILKINSON, AND DENISE TAYLOR

In the years following the publication of Nudge, many governments started behavioral insights units whose aims are to apply the principles of fields such as behavioral economics, psychology, neuroscience, and related social sciences to translate how humans behave and make decisions in everyday life into public policy (Behavioural Insights Team, 2010; BETA, 2019; Congdon & Shankar, 2018; Lourenço et al., 2016). Nudges represent one behavior change approach commonly implemented by behavioral insights units.

42.5 Nudge Interventions Using the principles described by Thaler and Sunstein and others (e.g., Ly et al., 2013; Richburg-Hayes et al., 2014), a series of steps can be taken to develop a nudge (for a typology of nudge interventions that focus on changing behavior by altering the physical environment, see Chapter 14, this volume): Step 1: Examine the choice environment. The goal is to identify the options people have, the choices they are being asked to make, and the situations in which they are being asked to make these choices. This process includes identifying (1) whether the incentives align with the desired behavior; (2) the type and frequency of feedback that people will receive about the consequences of their decisions; (3) whether any default options or choices are currently in place; (4) the number of decision points; and (5) how individuals are currently making decisions. Step 2: Ensure that a nudge is appropriate. Nudges are only appropriate when they promote behaviors that individuals want to take (i.e., the behaviors are consistent with their preferences) but fail to choose due to specific features of the decision environment or the manner in which they are making the decision. Thus, before embarking on an intervention, it needs to be established that individuals prefer the promoted

behavior and other barriers prevent them from adopting it. This process is complicated by the fact that, for any given decision, there are likely to be competing motives and factors to consider and many decisions involve a trade-off between the preferred action and its cost. Step 3: Choose or design the nudge. This step involves considering nudge options, which can involve consulting nudge taxonomies and empirical evidence on specific nudges and then matching potential nudges with characteristics of the specific behavior and context (e.g., whether the behavior is a single choice, habitual behavior, or intermittent behavior; whether the decision is made in a market, workplace, or home environment). Researchers have proposed a variety of ways to categorize or organize nudges into taxonomies. One approach is to focus on underlying cognitive processes or mechanisms to be targeted by a nudge intervention (e.g., Datta & Mullainathan, 2014; Dolan et al., 2012). For example, Datta and Mullainathan (2014) organize nudges according to four “mental constraints”: self-control, attention, cognitive capacity, and understanding. A second approach structures the taxonomy according to nudge intervention techniques and strategies such as commitment techniques, default setting, and micro-incentives (e.g., Abrahamse et al., 2005; Szaszi et al., 2018). A third approach is to categorize nudge techniques according to decision-making categories such as decision information, decision structure, and decision assistance (Mȕnscher, Vetter, & Scheurle, 2016). Step 4: Evaluating the efficacy, effectiveness, and sustainability of the nudge intervention (see also Chapter 22, this volume). Developers of nudge-based interventions must demonstrate that they are efficacious and effective and that their behavioral effects are sustainable over time.

Economic and Behavioral Economic Interventions

623

Table 42.1 Mapping nudge approaches to cognitive biases Nudge Approaches

Biases per Type of Barrier Optimism and overconfidence Limited attention, vividness, availability heuristic, and priming Commitment consistency principle Loss and regret aversion Status quo bias Anchoring Affect heuristic Time-inconsistent preferences, hyperbolic discounting, and present bias Ego depletion and decision fatigue Social norms

Social Changing incentives Increase default Choice salience of Message Commitment and peer Microincentives architecture options information framing techniques effects X X

X

X

X

X

X

X X X X

X X X

X

X

X

X

X

X

X

X

Note. Adapted from Pinto et al. (2014) and Luoto and Carman (2014). X = expected applicability of the nudge approach to addressing the bias.

Establishing the effectiveness of a nudge is, in principle, no different from testing the effectiveness of any intervention. Numerous attempts to evaluate nudge interventions have yielded conflicting or inconclusive evidence (see, e.g., Li & Chapman, 2013; Libotte, Siegrist, & Bucher, 2014; Szaszi et al., 2018), highlighting the need for more rigorous trials testing efficacy and effectiveness. Table 42.1 presents an example of a categorization system adapted from Pinto and colleagues (2014) that maps common nudge interventions

with the cognitive biases that they target to promote the desired behavioral decision. Sidebar 42.1 provides further descriptions of heuristics and cognitive characteristics identified as causing decisions to deviate from expected utility theory and are thus targets for nudge interventions. That said, one challenge with adapting nudges to address specific biases is the paucity of evidence not only of the effectiveness of the nudge but also on whether the heuristics presumed to be targeted by the nudge (e.g., overconfidence, limited attention, and affect presumed to be targeted by the nudge of increasing salience) are present in the decision context.

624

PAUL M. BROWN, LINDA D. CAMERON, MARTIN WILKINSON, AND DENISE TAYLOR

Sidebar 42.1 Heuristics or characteristics that cause decisions to deviate from expected utility theory

Optimism and overconfidence. A person’s subjective confidence in their judgments is reliably greater than the objective accuracy of those judgments, especially when confidence is relatively high. Limited attention, vividness, availability heuristic, and priming. People make decisions based on salient events and information and are influenced by tendencies to assess the likelihood of an event based on the ease with which they can think of examples of that event. Commitment consistency principle. People feel obliged to work hard to fulfill promises they made. Loss and regret aversion. People react more strongly to perceived losses (relative to a reference point) than perceived gains. Their willingness to engage in risky behavior increases when outcomes are perceived as losses, and people can be motivated to reduce decision regret. Status quo bias. People are more likely to select the default option or the current decision than to actively choose an alternative. Anchoring. The presentation of information with high or low values as initial options can bias decisions toward those values. Affect heuristic. Information can be more impactful on decisions when it has strong emotional content. Time-inconsistent preferences, hyperbolic discounting, and present bias. People can make choices that suggest they significantly discount future benefits and costs, even though these discounts can be inconsistent with their reflective preferences (system 2 thinking). Ego depletion, decision fatigue, and too many choices. Decision-making under conditions of stress and fatigue can result in choices that give in to temptation and lead to deviations from reflective preferences. Reliance on social norms. The knowledge of social norms can motivate decisions to act in accordance with those norms. For more information, see Luoto and Carman (2014), Pinto et al. (2014), Hanoch, Barnes, and Rice (2017), and Roberto and Kawachi (2015).

Some examples of how these nudges are used include the following: Increasing the salience of information. Making information that is consistent with the reflective goals salient is expected to promote system 2 processing. Providing calorie amounts for items on menus is one example of this strategy. Another

example is the trompe l’oeil painting of a girl chasing a ball in the street aimed at prompting drivers to slow down at high-risk intersections (Calamia, 2010). Framing. Message framing strategies are designed to change the way individuals perceive the target behavior so that they are more likely to

Economic and Behavioral Economic Interventions

engage in the preferred action. Reducing plate sizes to reduce food consumption and waste (Wansink & Van Ittersum, 2007; for more examples, see Chapter 14, this volume) and providing options for charitable giving in wills that start with high dollar amounts are examples of framing strategies for changing perceptions to change behavior (Behavioural Insights Team, 2010). Commitment techniques. Commitment devices take advantage of individual motivations to be consistent in attitudes, stated intentions, and actions (Aronson, 1969). Having a person create and sign a commitment contract represents one way to harness the power of this motivation to engage in a desired behavior. Another form of commitment contracting to overcome limited willpower is to have persons commit to engaging in an undesirable action should they fail to engage in the desired action. For example, there are smartphone apps that enable individuals to set up contracts to donate their money to their least-favored political party if they do not meet an established goal, such as going to the gym four times a week.1 Social incentives and peer effects. Altering social incentives such as by providing information on social norms or enhancing peer recognition can motivate behavior change. For instance, providing information on the number of people in the neighborhood who vote can increase voting participation (Behavioural Insights Team, 2010). Micro-incentives. Incentives are a commonly used technique for changing behavior that can be very effective. In contrast to large incentives, micro-incentives involve small amounts that do not change the cost-benefit calculus; instead, they provide an additional factor for the individual to consider at the time of the decision (for a more thorough discussion of how incentives work, see Chapter 36, this volume). One example is a “smoke pole” that lights up and plays sounds when cigarette smokers place their used butts into its bin, thereby encouraging smokers to dispose of them properly

625

instead of throwing them on the ground and polluting the environment (Markham, 2014). Choice architecture. The rationale behind choice architecture is to change the decision environment by removing, or discouraging, the undesirable option in favor of a more preferred option. An example is arranging the pantry so that the first thing one sees when searching for food is a healthy option and either eliminate unhealthy options in the house or place them in less-obvious positions (see Chapter 14, this volume). This approach involves changing the physical environment to make it more likely that the preferred, system 2 decision is made, although, as with other nudge strategies, empirical evidence regarding the cognitive mechanisms through which choice architecture affects behavior remains lacking. Default options. A frequently cited example of default options is to enroll employees in a specific occupational pension plan unless they actively opt out of it. Employees are more likely to be in such plans if they have to “opt out” than if they have to “opt in” (see Sidebar 42.2). Other examples include placing standing desks in the “stand” position every day before work (Venema, Kroese, & De Ridder, 2017) and changing the default from opt in to opt out for organ donation (English et al., 2019). Part of the attractiveness of nudges is the promise that small, and presumably inexpensive, changes can lead to significant behavior changes, including reduced energy consumption, increased tax revenues, and reductions in traffic accidents. Thus, there is an implicit, if not explicit, assumption that these nudges will save money and be cost-effective. Yet economic analyses of the budgetary impact, return on investment, and related financial outcomes of nudge interventions remain scant. The need for precision in assessments of cost in behavior intervention evaluations and the importance of assessing those costs that are of interest to the parties who are responsible for sustaining the intervention have

1

For example, see StickK at www.stickk.com/.

626

PAUL M. BROWN, LINDA D. CAMERON, MARTIN WILKINSON, AND DENISE TAYLOR

Sidebar 42.2 Example of a nudge

Examples of errors in retirement and savings decisions: • • • • •

Severely underinvesting in retirement savings. Choosing companies that charge a load or surcharge, including excessive amounts of their company’s stock in their portfolio. Being overly risk-averse or not diversifying at all. Using portfolio construction criteria that are overly simplistic. Rarely varying the investment portfolio over time as one ages.

Ways to remedy these errors: •

•

•

Availability. Many employers do not make payroll deduction plans available to their employees. Providing workers with easy access to a payroll deduction–based plan would greatly increase their use. Automatic enrollment. Traditional retirement plans require participants to opt into a plan. Change the default to make opting-in the default and include choices of a savings rate and investment portfolio as part of their standard decisions. Commitment. To overcome tendencies to discount future costs, encourage employees to commit now to increase their savings rate at some later time (e.g., several months in the future) or commit to planned increases in savings rates when they get raises.

For more details, see Rice (2013) and Benartzi, Shlomo, and Thaler (2013).

been previously noted (e.g., Brown, Cameron, & Ramondt, 2015; Glasgow, Vogt, & Boles, 1999; see also Chapters 22 and 23, this volume).

42.6 Summary and Conclusion The goal of this chapter was to review neoclassical and behavioral economic approaches to behavior change. Neoclassical approaches (information, incentives, taxes, and regulations) are predicated on the assumption that preferences are stable and attempts to change behavior are only warranted when there is some greater social good. The use of information, taxes, subsidies, and regulations is not unique to economics, as other behavior change models draw on similar interventions and approaches (see Michie, van Straalen, & West, 2011), but they are the main approaches advocated by neoclassical economics. Research based on

behavioral economics and the use of nudges to change behavior has increased dramatically since Thaler and Sunstein’s (2008) Nudge was first published in 2008. Behavioral economics theory and research extend beyond nudges, with many researchers working to incorporate realistic behavioral assumptions into mathematical models of behavior. For example, the expected utility framework has been adapted to incorporate such disparate aspects as hyperbolic discounting, learning behavior, emotions, and neuroeconomics (for a review, see Dhami, 2016). Many of these attempts, however, are descriptive and thus aimed at explaining rather than changing behavior. At least for the time being, nudges remain the most prominent type of behavior change intervention in behavioral economics. When assessing the increase in the use of nudges, five questions arise. First, do nudges

Economic and Behavioral Economic Interventions

represent a different type of intervention or merely a repackaging of techniques that psychologists and others have used for decades? After all, the heuristics on which nudges are motivated were first explored more than forty years ago (see Gilovich, Griffin, & Kahneman, 2002; Kahneman & Tversky, 2001; Tversky, 1975) and subsequent research has identified an everincreasing list of ways in which decision-making deviates from expected utility theory (Dhami, 2016). Any good intervention should be designed by considering the manner in which decisions are made, the barriers that prevent behavior change, and the decision points along the way. From this standpoint, there is nothing unique about nudges that would suggest that they represent a wholesale change from prior intervention efforts. This point raises the second question of why nudges have risen to prominence. Unlike other types of new approaches to behavior change, nudges were adopted by governments via behavioral insights units around the world primarily because they promise to change behavior without restricting choice, which is consistent with “libertarian paternalism,” to be low cost and easily implemented, and to yield significant changes in behavior. Nudges are not, by definition, large or complex interventions that require significant resources nor are they heavy-handed approaches that impose or restrict behavior. Thus, they appeal because they potentially shape behavior at a low cost and without significant public resistance. The third question concerns the ease with which previous research on nudges can be applied to develop new interventions. While there are numerous guides to developing and implementing nudges, as described in Section 42.5, most appear to assume that the interventionist can identify the cognitive biases inherent in the decision process and then use nudges to address these biases. While these guides do provide an overview, there is a need to tailor the nudges to specific decision contexts, such as one-time decisions, repeated decisions with feedback, and repeated decisions

627

without feedback. Additional research is needed to identify the types of nudges that would be appropriate in a given decision environment and their impact on cognitive mechanisms. The fourth question is whether nudges are effective in changing behavior. Numerous studies have demonstrated the effectiveness of incentives in changing behavior (e.g., Frey, 1997; Giles et al., 2014); while it is difficult to say which of these effects were due to payments being sufficiently high to change the cost-benefit calculation, it appears that, in general, financial incentives do influence behavior. The evidence for other types of nudge interventions appears more inconclusive. While some studies report statistically significant changes in behavior change in response to nudges (e.g., Arshad, Anderson, & Shariff, 2019; Basu & Madsen, 2017), concerns have been expressed about the meaningful significance of these changes (Marteau et al., 2011). Although nudges are relatively new and more time might be required before their effectiveness is demonstrated, it remains possible that some claims of their effectiveness are overstated. The fifth question is, will nudges have a lasting impact as behavioral interventions going into the future? On one hand, numerous critiques about the basic assumptions underlying nudges, including the dual-processing system (systems 1 and 2), and the viability and ethics of developing interventions to overcome the observed biases in decision-making resulting from the use of heuristics, could retard future growth in the development of nudges (Hollands, Marteau, & Fletcher, 2016; Krajbich et al., 2015; Lin et al., 2017; Moseley & Stoker, 2013). However, it is also clear that nudges have already had a significant impact on the field, having been applied to areas as diverse as cancer screening (Purnell et al., 2015), the design of social programs (Richburg-Hayes et al., 2014), health behaviors (Pinto et al., 2014), water quality management (Barnes et al., 2013), combating substance abuse (Correia, 2004), nutrition/diet (e.g., Heshmat, 2006; Rozin et al., 2011), shopping

628

PAUL M. BROWN, LINDA D. CAMERON, MARTIN WILKINSON, AND DENISE TAYLOR

behavior (Thapa et al., 2014), physical activity (e.g., Zimmerman, 2009), risky behaviors (Galizzi, 2012), and lowering health care costs (e.g., King et al., 2013). One way to view the rise in behavioral economics in general, and nudges in particular, is that Thaler and Sunstein’s (2008) book, Nudge, translated developments in economics and psychology about the way that people make decisions into the policy arena. That is, while the psychological and behavioral principles on which nudges are based have been known for several decades, the systematic attempt to integrate nudges and other behavioral insights into public policy is more recent. Economics and expected utility theory have long been the basis for policy making, and thus expanding economics to include these behavioral principles informed by psychological theory has provided an avenue for these ideas to become part of the policy toolkit. Whether nudges will continue to increase in popularity or become a historical sidenote remains to be seen, but it will ultimately depend on further systematic theory development and research testing the effectiveness of nudge interventions.

References Abrahamse, W., Steg, L., Vlek, C., & Rothengatter, T. (2005). A review of intervention studies aimed at household energy conservation. Journal of Environmental Psychology, 25, 273–291. https:// doi.org/10.1016/j.jenvp.2005.08.002 Aronson, E. (1969). The theory of cognitive dissonance: A current perspective. In L. Berkowitz (Ed.), Advances in Experimental Social Psychology, Vol. 4 (pp. 1–34). New York: Academic Press. Arshad, A., Anderson, B., & Sharif, A. (2019). Comparison of organ donation and transplantation rates between opt-out and opt-in systems. Kidney International, 95, 1453–1460. https://doi.org/ 10.1016/j.kint.2019.01.036 Barnes, A. P., Toma, L., Willock, J., & Hall, C. (2013). Comparing a “budge” to a “nudge”: Farmer responses to voluntary and compulsory

compliance in a water quality management regime. Journal of Rural Studies, 32, 448–459. https://doi.org/10.1016/j.jrurstud.2012.09.006 Basu, S., & Madsen, K. (2017). Effectiveness and equity of sugar-sweetened beverage taxation. PLoS Medicine, 14, e1002327. https://doi.org/ 10.1371/journal.pmed.1002327 Bates, C. (2009). Scaling new heights: Piano stairway encourages commuters to ditch the escalators, Daily Mail, April 4. www.dailymail.co.uk/scien cetech/article-1218944/Scaling-new-heightsPiano-stairway-encourages-commuters-ditchescalators.html. Becker, G. S., & Becker, G. S. (2009). A Treatise on the Family. Cambridge, MA: Harvard University Press. Behavioural Insights Team. (2010). Applying behavioural insight to health. Discussion Paper, Cabinet Office Behavioural Insights, London. https://assets.publishing.service.gov.uk/govern ment/uploads/system/uploads/attachment_data/ file/60524/403936_BehaviouralInsight_acc.pdf Benartzi, S., & Thaler, R. H. (2013). Behavioral economics and the retirement savings crisis. Science, 339, 1152–1153. https://doi.org/10.1126/ science.1231320 BETA (Behavioural Economics Team of Australian Government). (2019). Impact Report, April. https://behaviouraleconomics.pmc.gov.au/sites/ default/files/resources/impact-report.pdf Brown, P. M., Cameron, L. D., & Ramondt, S. (2015). Sustainability of behavioral interventions: Beyond cost-effectiveness analysis. International Journal of Behavioral Medicine, 22, 425–433. https://doi.org/10.1007/s12529-014–9437-z Calamia, J. (2010). Speed bumps of the future: Creepy optical illusion children. Discovery Magazine, April 4. http://blogs.discovermagazine.com/disco blog/2010/09/07/speed-bumps-of-the-futurecreepy-optical-illusion-children/ Chaiken, S., & Trope, Y. (Eds.). (1999). Dual-Process Theories in Social Psychology. New York: Guilford Press. Congdon, W. J., & Shankar, M. (2018). The role of behavioral economics in evidence-based policymaking. The Annals of the American Academy of Political and Social Science, 678, 81–92. https://doi.org/10.1177/0002716218766268

Economic and Behavioral Economic Interventions

Correia, C. J. (2004). Behavioral economics: Basic concepts and clinical applications. In M. Cox & E. Klinger (Eds.), Handbook of Motivational Counseling: Goal-Based Approaches to Assessment and Intervention with Addiction and Other Problems (pp. 49–72). Chichester: Wiley. Datta, S., & Mullainathan, S. (2014). Behavioral design: A new approach to development policy. Review of Income and Wealth, 60, 7–35. https:// doi.org/10.1111/roiw.12093 Deaton, A., & Muellbauer, J. (1980). Economics and Consumer Behavior. Cambridge: Cambridge University Press. Dhami, S. (2016). The Foundations of Behavioral Economic Analysis. Oxford: Oxford University Press. Dolan, P., Hallsworth, M., Halpern, D., King, D., Metcalfe, R., & Vlaev, I. (2012). Influencing behaviour: The mindspace way. Journal of Economic Psychology, 33, 264–277. https://doi .org/10.1016/j.joep.2011.10.009 English, V., Johnson, E., Sadler, B. L., & Sadler, A. M. (2019). Is an opt-out system likely to increase organ donation? BMJ, 364, 1967. https://doi.org/ 10.1136/bmj.l967 Frey, B. (1997). Not Just for the Money: An Economic Theory of Personal Motivation. Cheltenham: Edward Elgar. Friedman, M., & Friedman, M. (1953). Essays in Positive Economics. Chicago: University of Chicago Press. Galizzi, M. M. (2012). Label, nudge or tax? A review of health policies for risky behaviours. Journal of Public Health Research, 1, 14. https://doi.org/ 10.4081/jphr.2012.e5 Giles, E. L., Robalino, S., McColl, E., Sniehotta, F. F., & Adams, J. (2014). The effectiveness of financial incentives for health behaviour change: Systematic review and meta-analysis. PLoS ONE, 9, e90347. https://doi.org/10.1371/ journal.pone.0090347 Gilovich, T., Griffin, D., & Kahneman, D. (Eds.). (2002). Heuristics and Biases: The Psychology of Intuitive Judgment. Cambridge: Cambridge University Press. Glasgow, R., Vogt T., & Boles S. (1999). Evaluating the public health impact of health promotion interventions: The RE-AIM framework. American

629

Journal of Public Health, 89, 1322–1327. https:// doi.org/10.2105/AJPH.89.9.1322 Hanoch, Y., Barnes, A., & Rice, T. (Eds.). (2017). Behavioral Economics and Healthy Behaviors: Key Concepts and Current Research. New York: Routledge. Heshmat, S. (2006). Applying behavioral economics to changing health behavior: The case of weight-loss management. Californian Journal of Health Promotion, 4, 21–29. Hollands, G. J., Marteau, T. M., & Fletcher, P. C. (2016). Non-conscious processes in changing health-related behaviour: A conceptual analysis and framework. Health Psychology Review, 10, 381–394. https://doi.org/10.1080/17437199.2015 .1138093 Kahneman, D. (2011). Thinking, Fast and Slow. New York: Farrar, Straus, and Giroux. Kahneman, D., & Tversky, A. (2001). Choices, Values, and Frames. Cambridge: Cambridge University Press. King, D., Greaves, F., Vlaev, I., & Darzi, A. (2013). Approaches based on behavioral economics could help nudge patients and providers toward lower health spending growth. Health Affairs, 32, 661– 668. https://doi.org/10.1377/hlthaff.2012.1348 Krajbich, I., Bartling, B., Hare, T., & Fehr, E. (2015). Rethinking fast and slow based on a critique of reaction-time reverse inference. Nature Communications, 6, 7455. https://doi.org/ 10.1038/ncomms8455 Li, M., & Chapman, G. B. (2013). Nudge to health: Harnessing decision research to promote health behavior. Social and Personality Psychology Compass, 7, 187–198. https://doi.org/10.1111/ spc3.12019 Libotte, E., Siegrist, M., & Bucher, T. (2014). The influence of plate size on meal composition: Literature review and experiment. Appetite, 82, 91– 96. https://doi.org/10.1016/j.appet.2014 .07.010 Lim, D., Slemrod, J., & Wilking, E. (2013). Expert and public attitudes towards tax policy: 2013, 1994, and 1934. National Tax Journal, 66, 775–806. https://doi.org/10.17310/ntj.2013.4.01 Lin, Y., Osman, M., & Ashcroft, R. (2017). Nudge: Concept, effectiveness, and ethics. Basic and Applied Social Psychology, 39, 293–306. https:// doi.org/10.1080/01973533.2017.1356304

630

PAUL M. BROWN, LINDA D. CAMERON, MARTIN WILKINSON, AND DENISE TAYLOR

Lourenço, J. S., Ciriolo, E., Almeida, S. R., & Dessart, F. J. (2016). Behavioural Insights Applied to Policy-Country Overviews 2016 (No. JRC100547), June 16. Joint Research Centre (Seville site), European Commission. https://pub lications.jrc.ec.europa.eu/repository/bitstream/ JRC100547/biap%20country%20overviews% 202016.pdf Luoto, J., & Carman, K. G. (2014). Behavioral economics guidelines with applications for health interventions. Technical Note No. IDB-TN-665, March, Inter-American Development Bank. https:// publications.iadb.org/en/publication/12074/beha vioral-economics-guidelines-applications-healthinterventions Ly, K., Mazar, N., Zhao, M., & Soman, D. (2013). A practitioner’s guide to nudging. Working Paper No. 2609347, Rotman School of Management. https://papers.ssrn.com/sol3/papers.cfm?abstract _id=2609347 Markham, D. (2014). Could a giant musical cigarette butt reduce cigarette litter? TreeHugger, April 4. www.treehugger.com/gadgets/could-giant-musi cal-cigarette-butt-reduce-cigarette-litter.html Marteau, T. M., Ogilvie, D., Roland, M., Suhrcke, M., & Kelly, M. P. (2011). Judging nudging: Can nudging improve population health? BMJ, 342, 228. https://doi.org/10.1136/bmj.d228 Michie, S., van Straalen, M., & West, R. (2011). The behaviour change wheel: A new method for characterizing and designing behaviour change interventions. Implementation Science, 6, 42. https://doi.org/10.1186/1748-5908 Moseley, A., & Stoker, G. (2013). Nudging citizens? Prospects and pitfalls confronting a new heuristic. Resources, Conservation and Recycling, 79, 4–10. https://doi.org/10.1016/j.resconrec.2013.04.008 Mozaffarian, D., Hemenway, D., & Ludwig, D. S. (2013). Curbing gun violence: Lessons from public health successes. Journal of the American Medical Association, 309, 551–552. https://doi .org/10.1001/jama.2013.38 OECD (Organisation for Economic Co-operation and Development). (2019). Delivering Better Policies Through Behavioural Insights: New Approaches. Paris: OECD Publishing. https://doi.org/10.1787/ 6c9291e2-en

Pinto, D. M., Ibarrarán, P., Stampini, M. et al. (2014). Applying behavioral tools to the design of health projects. Policy Brief No. IDB-PB-228, July, Inter-American Development Bank. https://publi cations.iadb.org/en/applying-behavioral-toolsdesign-health-projects Purnell, J., Thompson, T., Kreuter, M., & McBride, T. (2015). Behavioral economics: “Nudging” underserved populations to be screened for cancer. Preventing Chronic Disease, 12: 140346. https:// doi.org/10.5888/pcd12.140346 Rice, T. (2013). The behavioral economics of health and health care. Annual Review of Public Health, 34, 431–447. https://doi.org/10.1146/annurevpublhealth-031912-114353 Richburg-Hayes, L., Anzelone, C., Dechausay, N. et al. (2014). Behavioral economics and social policy: Designing innovative solutions for programs supported by the Administration for Children and Families. Office of Planning, Research and Evaluation, Administration for Children and Families, US Department of Health and Human Services. www.acf.hhs.gov/opre/resource/beha vioral-economics-and-social-policy-designinginnovative-solutions-for-programs-supported-bythe-administration-for Roberto, C. A., & Kawachi, I. (Eds.). (2015). Behavioral Economics and Public Health. Oxford: Oxford University Press. Rozin, P., Scott, S., Dingley, M., Urbanek, J. K., Jiang, H., & Kaltenbach, M. (2011). Nudge to obesity I: Minor changes in accessibility decrease food intake. Judgment and Decision Making, 6, 323–332. Szaszi, B., Palinkas, A., Palfi, B., Szollosi, A., & Aczel, B. (2018). A systematic scoping review of the choice architecture movement: Toward understanding when and why nudges work. Journal of Behavioral Decision Making, 31, 355– 366. https://doi.org/10.1002/bdm.2035 Takemura, K. (2014). Behavioral Decision Theory: Psychological and Mathematical Descriptions of Human Choice Behavior. Tokyo: Springer. https:// doi.org/10.1007/978-4-431-54580-4_5 Thaler, R. H., & Sunstein, C. R. (2008). Nudge: Improving Decisions About Health, Wealth, and Happiness. New Haven, CT: Yale University Press.

Economic and Behavioral Economic Interventions

Thapa, J., Lyford, C., Belasco, E. J. et al. (2014). Nudges in the supermarket: Experience from point of sale signs. Paper presented at the 2014 Annual Meeting of the Agricultural and Applied Economics Association, Minneapolis, July 27–29. https://ideas.repec.org/p/ags/aaea14/171442.html Tversky, A. (1975). A critique of expected utility theory: Descriptive and normative considerations. Erkenntnis, 9, 163–173. https://doi.org/10.1007/ BF00226380 US Department of Health and Human Services. (2014). The Health Consequences of Smoking – 50 Years of Progress: A Report of the Surgeon General. Atlanta, GA: US Department of Health and Human Services. www.ncbi.nlm.nih.gov/books/ NBK179276/pdf/Bookshelf_NBK179276.pdf

631

Venema, T. A. G., Kroese, F. M., & De Ridder, D. T. D. (2017). I’m still standing: A longitudinal study on the effect of a default nudge. Psychology and Health, 33, 669–681. https://doi.org/10.1080/ 08870446.2017.1385786) Von Neumann, J., & Morgenstern, O. (1953). Theory of Games and Economic Behavior. Princeton: Princeton University Press. Wansink, B., & Van Ittersum, K. (2007). Portion size me: Downsizing our consumption norms. Journal of the Academy of Nutrition and Dietetics, 107, 1103– 1106. https://doi.org/10.1016/j.jada.2007.05.019 Zimmerman, F. J. (2009). Using behavioral economics to promote physical activity. Preventive Medicine, 49, 289–291. https://doi.org/10.1016/j.ypmed.2009 .07.008

43 Dyadic Behavior Change Interventions Urte Scholz, Corina Berli, Janina Lüscher, and Nina Knoll

Practical Summary Including another person in a behavior change intervention for a target person seems to be a promising strategy for behavior change, given that most behaviors happen in a social context. The scientific evidence base for the ideal content of dyadic interventions and their effectiveness, however, is relatively sparse and more evidence is needed. This chapter reviews existing theories and research on dyadic behavior change interventions and presents a continuum of individual to dyadic behavior change techniques that allow choosing from different degrees of involvement of one or both dyad members in behavior change treatment. Owing to the heterogeneity in dyadic behavior change research, the step-by-step guide mainly serves the function of sensitizing practitioners to the complexity of dyadic behavior change and of calling for more research in this area.

43.1 Introduction Members of individuals’ social networks are often involved in behavior change attempts. Consistent associations between social integration and survival benefits (e.g., Holt-Lunstad, Smith, & Layton, 2010), together with the broader effects of social interaction and the real and implied influence of groups on behavior (see Chapter 16, this volume), indicate the potentially facilitative role that close members of an individual’s social network may have in efforts to change behavior. This chapter classifies and reviews existing dyadic behavior change research in different dyadic constellations (e.g., romantic couples, peers). Given the heterogeneity of the techniques used within and across different constellations of dyads, this chapter will not provide a comprehensive review of dyadic interventions for all kinds of dyads. Most examples for dyadic interventions will be from the

literature of dyadic behavior change in romantic couples, given that this is the closest and most important relationship during the adult lifespan (e.g., Antonucci, Akiyama, & Takahashi, 2004). The chapter focuses on techniques that promote behavior change as the endpoint, in contrast to behavior change as a means (e.g., as in couple therapy that focuses on relationship outcomes). The chapter first introduces a continuum of individual to dyadic behavior change techniques (DBCTs) that address different degrees of involvement of a dyad partner. The DBCTs can target the behavior change in one or both partners of the dyad. Next, a theoretical framework and evidence on dyadic behavior change interventions are outlined. Given the relatively early developmental stage of this literature, the subsequent step-bystep guide focuses on promising approaches to https://doi.org/10.1017/9781108677318.043

Dyadic Behavior Change Interventions

behavior change in dyads and provides suggestions for future research.

43.2 Definitions Recently, there has been considerable effort to identify “active ingredients” involved in behavior change interventions (see Chapter 20, this volume). The resulting taxonomies of behavior change techniques (BCTs; Michie et al., 2013) have mostly identified techniques addressing the individual (e.g., goal setting), with very few involving nonprofessional social interactions to change individual behavior (e.g., social support). A BCT taxonomy based on the intervention mapping approach (Kok et al., 2016; see also Chapter 19, this volume) addresses interpersonal influences to a greater extent, but there are still many DBCTs that have not been listed in existing taxonomies. Owing to the heterogeneity of DBCTs and the fairly small existing evidence on such approaches in the behavior change literature, examples of such techniques are provided instead of a comprehensive list and they are aligned along a continuum of individual to dyadic techniques (see Figure 43.1). Note that the continuum excludes professional interaction partners (e.g., interventionists, teachers, instructors, health professionals, coaches,

633

social workers) that could be present for all techniques alike (e.g., instructing the target person to set goals). All DBCTs can target the behavior of either one or both dyad members. “Individual” BCTs at the left of the continuum are well covered in existing taxonomies (e.g., Kok et al., 2016; Michie et al., 2013). Guidelines for delivery mode do not explicitly specify the presence of a social network member as a category (Dombrowski, O’Carroll, & Williams, 2016; Hagger & Hardcastle, 2014; Hoffmann et al., 2014; Schulz et al., 2010). However, the mere presence of a social network member might account for differential intervention effectiveness due to implicit mechanisms of action. The first dyadic category, “parallel” techniques, can be any individual technique directed at both dyad members but without any instructed interaction between them. The “crossover” category comprises techniques that involve an interaction between dyad members but do not have to be administered with both. Finally, the “joint” category covers fully dyadic techniques in which both dyad members are actively involved and considered a unit instead of two individuals. Accordingly, these joint DBCTs cannot be delivered if only one dyad member is present (for examples, definitions, and assumed/implicit mechanisms of action, see Table 43.1 in Sidebar 43.1).

Figure 43.1 Continuum of dyadic behavior change techniques

634

¨ SCHER, AND NINA KNOLL URTE SCHOLZ, CORINA BERLI, JANINA LU

Sidebar 43.1 Dyadic behavior change techniques and relations with existing taxonomies

Table 43.1 displays examples of (D)BCTs for each category along the continuum, cited from existing taxonomies or the literature when they had not been identified as BCTs yet. Note that this table is not a comprehensive list of all possible DBCTs. Table 43.1 Examples of dyadic behavior change techniques along the continuum of dyadic behavior change techniques displayed in Figure 43.1

Dyadic BCT Category Label

Definition

Any individual BCT See individual BCTs in existing addressed at target person and taxonomies dyad member in parallel, e.g., both dyad members receive information but are not instructed to discuss with each other Crossover Social support Advise on, (practical) arrange, or provide practical help from dyad member for performance of the behavior Social support Advise on, (emotional) arrange, or provide emotional social support from dyad member for performance of the behavior Persuasive Guiding communication individuals and agents toward Parallel

Examples for Assumed/Implicit Mechanism of Action (Michie Source of BCT et al., 2018) (Kok et al., Depends on the BCT 2016; Michie administered. et al., 2013) Additionally, received or provided social support (Knoll, Scholz, & Ditzen, 2019) Demonstration of behavior/ modeling (Bandura, 1989) (Michie et al., Positive and 2013) negative affect (Seidman, Shrout, & Bolger, 2006) Self-efficacy (Bandura, 1989) (Michie et al., Positive and 2013) negative affect (Seidman et al., 2006) Self-efficacy (Bandura, 1989)

(Kok et al., 2016)

Positive social control Continued

Dyadic Behavior Change Interventions

635

Table 43.1 (cont.)

Dyadic BCT Category Label

Coercion

Definition

Examples for Assumed/Implicit Mechanism of Action (Michie Source of BCT et al., 2018) (Butterfield & Lewis, 2002) Positive and negative affect (Seidman et al., 2006) Negative social control (Butterfield & Lewis, 2002) Negative affect (Seidman et al., 2006) (Michie et al., Social support 2013); (Knoll et al., (Kok et al., 2019) 2016) Social control (Butterfield & Lewis, 2002)

the adoption of an idea, attitude, or action by using arguments or other means Attempting to (Kok et al., control others 2016) against their will

Feedback on beha- Dyad member vior or outcome monitors and provides informative or evaluative feedback on target person’s performance Support plan Create a specific (Voils et al., behavior plan to 2013) Not yet support target included in person’s any behavior taxonomy

Dyadic action control

Dyad member prompting target person’s awareness of behavior

Received or provided social support (Knoll et al., 2019) Positive and negative affect (Seidman et al., 2006) Action control (Berli et al., 2016; Scholz (Sniehotta et al., 2006) & Berli, 2014) Not yet included Continued

636

¨ SCHER, AND NINA KNOLL URTE SCHOLZ, CORINA BERLI, JANINA LU

Table 43.1 (cont.)

Dyadic BCT Category Label

Definition

Social support (Knoll et al., 2019) Positive social control (Butterfield & Lewis, 2002) (Michie et al., Commitment Behavioral contract Create written 2013) (Inauen, Tobias, specification of & Mosler, 2014) behavior to be Social control performed (Butterfield agreed by target & Lewis, 2002) person and dyad member, witnessed by another person Responsiveness Empathy training Stimulating dyad (Kok et al., 2016) (Reis & Gable, member to 2015) empathize with Social support another person, (Knoll et al., i.e., imagine 2019) how the other person would feel (Kok et al., Social support Cooperative Engineering 2016) (Knoll et al., learning lessons in a way 2019) that dyad Companionship members must (Rook, 1990) learn from one another Dyadic planning Planning together (Burkert et al., Social support with a dyad (Knoll et al., 2011; Knoll member but 2019) et al., 2017) carrying plan Social control Not yet out alone (Butterfield included in & Lewis, 2002) any taxonomy standards, monitoring of behavior, and regulation efforts

Joint

Examples for Assumed/Implicit Mechanism of Action (Michie Source of BCT et al., 2018) in any taxonomy

Continued

Dyadic Behavior Change Interventions

637

Table 43.1 (cont.)

Dyadic BCT Category Label

Definition

Examples for Assumed/Implicit Mechanism of Action (Michie Source of BCT et al., 2018)

Action control (Sniehotta et al., 2006) Plan quality (Keller et al., 2017) Self-efficacy (Bandura, 1989) Collaborative plan- Planning and car- (Prestwich et Social support (Knoll et al., ning/implementa- rying plan out al., 2005; 2019) tion intentions together with a Radtke et Plan quality (Keller dyad member al., 2018) et al., 2017) Not yet included in Modeling (Bandura, 1989) any taxonomy Self-efficacy (Bandura, 1989) Companionship (Rook, 1990) Dyadic efficacy Strengthening a (Lewis et al., Self-efficacy (Bandura, 1989) 2006) Not dyad’s percepyet included Social support tions of confi(Knoll et al., in any dence that 2019) taxonomy together they Companionship can manage (Rook, 1990) problems

43.3 Theory and Mechanisms of Change Several behavior change theories address the role of the (social) environment (e.g., Bandura, 1989; Bronfenbrenner, 1986; see also Chapters 16 and 18, this volume). For example, Bandura’s (1989; Chapter 3, this volume) social cognitive theory

builds on a triadic model of reciprocal determinism comprising behavior, person, and environment. The environment is assumed to produce the individual’s behavior, as well as cognition and affect, and also to be a product of the individual’s behavior, cognition, and affect. Bandura’s triadic model of reciprocal determinism is well suited to describe the interplay between social

638

¨ SCHER, AND NINA KNOLL URTE SCHOLZ, CORINA BERLI, JANINA LU

interactions, cognitions, affect, and behavior. The related mechanisms in terms of social factors and person factors can have multiple knock-on effects on other processes. For example, dyadic or collaborative planning might facilitate social support or social control (Knoll et al., 2017; Prestwich et al., 2005). Additionally, the effectiveness of planning collaboratively or dyadically might be enhanced by the joint generation of higher-quality plans compared to individual planning in the applied context of behavior change (see Chapters 6 and 39, this volume).

43.4 Evidence Base The literature on dyadic interventions for behavior change is very heterogeneous for several reasons. Dyadic interventions differ with regard to study populations (e.g., women only, people with chronic health conditions, minorities), behaviors targeted (e.g., different health behaviors, delinquency), dyadic constellations (e.g., couples, peers, parent-child), goal constellations (e. g., shared goals within the dyad, or individual goals of one dyad member; see Carr et al., 2019; Fitzsimons, Finkel, & vanDellen, 2015), and DBCTs used (e.g., crossover, joint, see Figure 43.1). For example, different dyadic constellations focusing on specific behaviors in specific study populations require different DBCTs. For instance, due to differences in power symmetry, a DBCT used in parent-child interventions might not be appropriate in couple interventions. Given this heterogeneity, this chapter will provide examples rather than a comprehensive overview of the evidence base of dyadic interventions in behavior change. Peer-based interventions can take different forms, such as group-based approaches that go beyond dyadic interventions, dyadic approaches with peers and target persons being coupled, and mixed approaches (Webel et al., 2010). A basic definition of peers is that they “share a common culture, language, and knowledge about the

problems that their community experiences” (Webel et al., 2010, p. 247). Recent systematic reviews on peer interventions targeting health behavior change indicate mixed support (Ramchand et al., 2017; Webel et al., 2010). For example, twelve RCTs reported positive, while ten RCTs reported no effects of dyadic peer-based interventions compared to usual care, education control, or professional-led interventions (Ramchand et al., 2017). Interventions targeting parent-child dyads with minors differ with regard to either focusing on the dyad or using a mix of dyadic and individual interventions by including additional sessions for youth and/or parents only (Sutton et al., 2014). With regard to behaviors related to sexual health in minority youth as another example, dyadic parent-child interventions tended to result in reduced sexual risk behaviors compared to active/passive dyadic control groups (Sutton et al., 2014). For weight loss, involvement of mothers seemed beneficial for children but not for adolescents. For the latter, individual interventions seemed more beneficial (McLean et al., 2003). Couple-based dyadic interventions resulted in small negative effects on unprotected sexual intercourse and increased condom use (Burton, Darbes, & Operario, 2010; Crepaz et al., 2015) and in increased physical activity compared to individual treatments or passive control groups (Richards et al., 2017). For couples with one partner at risk for chronic illness, dyadic interventions resulted in better health behavior compared to usual care or individual intervention (see Arden-Close & McGrath, 2017). Other couple-based intervention studies in the context of chronic illness targeting dyadic illness management, including medication adherence and lifestyle changes (e.g., Baucom, Kirby, & Kelly, 2010; Burkert et al., 2011), are inconclusive. Finally, systematic reviews and meta-analyses synthesizing evidence for dyadic interventions with different dyadic constellations (e.g., couples, peers, target person–family member) again find

Dyadic Behavior Change Interventions

ambiguous effects. For smoking cessation, there is no evidence for the effectiveness of support interventions targeting various dyads to encourage smoking cessation (Faseru et al., 2018). However, these interventions failed to increase support exchange in the first place (Faseru et al., 2018). With regard to physical activity and sedentary behavior, dyadic interventions resulted in small beneficial effects (Carr et al., 2019). Overall, evidence for dyadic interventions for behavior change is inconclusive. Aside from the pronounced heterogeneity described previously that precludes comparability and synthesis by means of meta-analyses, one of the main issues preventing the field making substantial progress is that the studies either fail to report the specific DBCTs used or describe the intervention content inconsistently across studies. Generating a common language for DBCTs as available for individual-focused interventions (Kok et al., 2016; Michie et al., 2013) will be an important next step. In addition, future studies adopting DBCTs as means to change behavior should consider general guidelines on reporting behavioral interventions (Hoffmann et al., 2014; Schulz et al., 2010). This should also include providing open access to the intervention material. Related to the poor reporting of DBCTs is a frequent neglect of a solid theoretical background (Arden-Close & McGrath, 2017). Furthermore, many intervention studies are of poor methodological quality (see Faseru et al., 2018; McLean et al., 2003), including being underpowered (see Martire & Helgeson, 2017). Thus, there is a strong need for more high-quality systematic research in this field.

43.5 Step-by-Step Guide Given the heterogenous and inconclusive nature of the current evidence base for dyadic interventions, the following step-by-step approach comprises tentative recommendations based on the best currently available information.

639

43.5.1 Typical Means of Delivery Means of delivery of behavioral interventions in dyads are as heterogeneous as DBCTs themselves. A possible classification can be derived from Figure 43.1 and is largely defined by the degree of interaction between dyad members during interventions. For example, DBCTs in the “parallel” category might be delivered by a practitioner in face-to-face settings, jointly or separately for both members of the dyad. In contrast, DBCTs in the “joint” category always require the dyad members to enact the DBCT jointly. Further, in case of DBCTs that have an equivalent technique at the individual level, such as collaborative or dyadic planning with the equivalent of individual planning, the typical means of delivery of the individual-level technique might simply be extended to the dyadic level (see Section 43.5.9). So far, there is not enough evidence to draw conclusions about the effectiveness of specific means of delivery for specific populations, dyadic constellations, and target behaviors.

43.5.2 Target Audience and Behaviors There is preliminary evidence that the effectiveness of dyadic behavior change interventions might depend on several moderator variables. For example, in the domain of weight loss, the target population was found to make a difference. For adolescents, individual interventions seem more beneficial with regard to weight loss than interventions involving mothers, whereas the opposite seems to be true for children (McLean et al., 2003). These results, however, are based on a limited number of studies, and studies included in this review tended to be of poor quality and frequently underpowered (McLean et al., 2003). Also, the behavior targeted might make a difference. For example, dyadic peer interventions seem to result in better outcomes for condom use but not for breastfeeding (Webel et al., 2010). In addition, goal constellations within dyads were found to make a difference in the context of physical activity

640

¨ SCHER, AND NINA KNOLL URTE SCHOLZ, CORINA BERLI, JANINA LU

and sedentary behavior (Carr et al., 2019). In couples, stronger effects were observed in interventions manipulating a shared target-oriented goal, where both members of the dyad agree on a goal to change the behavior of one member, compared to other goal constellations, such as when only one member of the dyad sets the goal for the other member’s behavior (Carr et al., 2019; Fitzsimons et al., 2015). Finally, dyadic constellations can make a difference with regard to effectiveness. In a meta-analysis of physical activity interventions comprising different dyadic constellations, peer/friend dyads resulted in higher effect sizes than other dyadic constellations (e.g., couples, parent-child) compared to individual interventions, with a total small and heterogeneous effect across all dyadic interventions (Carr et al., 2019). When considering these potential moderators, it should be kept in mind that evidence so far is rather limited and ambiguous. Future research needs to make a stronger effort to identify ideal matches between target audiences, behaviors, dyadic constellations, goal constellations, and specific DBCTs or the alternative use of individual interventions.

43.5.3 Enabling or Inhibiting Factors Enabling or inhibiting factors of a dyadic behavior change intervention should ideally be specified in the theory informing the intervention technique. In the intervention mapping approach, these factors are called parameters under which the effectiveness of an intervention technique may vary (Kok et al., 2016). For example, for dyadic empathy training, the parameters of effectiveness are that the dyad members need to be able and willing to identify with their target partner’s feelings but should not necessarily imagine how they themselves would feel as this might induce distress (Kok et al., 2016). Depending on the underlying theory and its specified DBCTs, different parameters need to be accounted for. The challenge is that these parameters are not yet fully developed for most DBCTs.

An enabling factor discussed in the literature is the joint engagement in the behavior change (Berli et al., 2018; Best et al., 2016; Jackson, Steptoe, & Wardle, 2015). For example, an intervention study in parent-child dyads found that joint changes in dietary behavior were related to maintenance in both parents’ and children’s weight loss across two years (Best et al., 2016). The crucial enabling function of joint behavioral change could be vicarious experience and strengthened self-efficacy in dyad members (Bandura, 1989; see Chapter 3, this volume). Joint engagement in a behavior change is also assumed to strengthen social support transactions between dyad members, reduce individual self-regulation effort, increase enjoyment of the behavior change, and strengthen goal commitment, all of which might, in turn, explain the enabling function of joint behavioral change (see Berli et al., 2018). Additionally, dyad-specific enabling and inhibiting factors of behavior change interventions are likely. For example, in couple-based interventions a priori relationship quality has been theorized (Lewis et al., 2006) and empirically verified as a moderator of dyadic behavior change interventions (e.g., Knoll et al., 2017). For example, in a study comparing a dyadic planning condition with an individual planning condition with an overall null effect on physical activity, lower relationship quality reported by target persons from healthy couples was associated with a decrease in moderate and vigorous activity after the intervention in the dyadic planning condition, whereas dyadic planning condition participants with higher relationship quality activity levels held their activity levels up to six weeks post-intervention. In the individual planning condition, a different pattern was observed. Target persons with lower relationship quality showed increased post-intervention physical activity levels, while those with higher relationship quality exhibited decreased levels (Knoll et al., 2017). This moderation effect was attributed to an inhibiting effect of planning dyadically with a partner with whom the target person

Dyadic Behavior Change Interventions

shares an unhappy relationship. In this case, abandoning the dyadic intervention in favor of an individual planning intervention seems the better choice (Knoll et al., 2017). In line with these results, a note of caution seems warranted. In cases where there is low willingness of a target person to change a behavior, combined with a strong conviction of the other dyad member that this change is necessary, it might be better not to intervene dyadically to protect the relationship and increase chances for successful behavior change (Sullivan & Davila, 2014). One mediator of such inhibiting effects of dyadic interventions can be the use of negative social control strategies (e.g., nagging, coercing). Negative social control was shown to relate to increased negative affect in the target person and to backfiring behaviors, such as doing the opposite of what the dyad partner wanted (Craddock et al., 2015). Related to this, dyad members’ motivation to engage in the dyadic intervention is likely a strong enabling factor (e.g., Baucom et al., 2012). In case of a low motivation of one or both partners, interventionists might explain the benefits of a dyadic approach (Baucom et al., 2012). In this instance, a motivational interviewing intervention targeting one or both dyad members could be combined with a dyadic intervention (see Sullivan & Davila, 2014).

43.5.4 Training and Skills Required In dyadic behavior change interventions, interventionists and dyad members need specific training and skills. For instance, in multiple-session interventions, the interventionist might be well advised to take dyad-related issues (e.g., relationship quality, see Section 43.5.3), dyad-specific challenges of the behavior change (e.g., depressive symptoms in one of the dyad members), and dyad-interventionist-specific factors (e.g., working alliance) into account (Baucom et al., 2012). Regarding the latter, a good working alliance with each partner but also with the dyad as a

641

unit is required. For instance, taking sides when the behavior change is indicated in one dyad member only should be prevented. Furthermore, ideally both dyad members should receive training for their roles in the dyadic intervention. The concept of skilled support comprises the when, what, how, and who of support (Rafaeli & Gleason, 2009). The concept of skilled support was introduced in the context of couple-based interventions but might well be generalized to other dyadic interventions with adults. The when of support refers to the ideal timing. This requires the target person to be specific about when support from the partner is needed and the partner to be responsive to this need. The what of support refers to the different functions of social support – for example, emotional or instrumental support (Knoll et al., 2019). Ideally, the kind of support provided by one partner matches the needs of the other (e.g., Cutrona & Russell, 1990). The how of social support refers to the dimensions of visibility and directness. Invisible social support is defined as support that is provided by a partner but not reported as being received by the other (Bolger, Zuckerman, & Kessler, 2000). Correlational and experimental research suggests the superiority of invisible social support compared to visible social support (Lüscher et al., 2015). Invisible support is supposed to be effective because it comes with all the benefits of social support but excludes the negative side effects, such as introducing a need to reciprocate or undermining the self-esteem (Bolger et al., 2000). The who of support describes that support receipt and provision are best equally distributed among partners to avoid feelings of inequity in terms of which of the two benefits most or least (Gleason et al., 2003). Dyads with one or both dyad members undergoing a behavior change attempt should be advised to ensure that both dyad members can provide reciprocal support to each other. For example, one technique of behavior couple therapy is behavior exchanges to enhance reciprocity (Sullivan & Davila, 2014). This may include one partner avoiding snacking

642

¨ SCHER, AND NINA KNOLL URTE SCHOLZ, CORINA BERLI, JANINA LU

in front of the other who intends to lose weight and this partner in turn agrees not to skip family dinners. When partners encode these agreed-on behaviors as provision of support, supportive equity can be established. Overall, skilled support is one promising avenue for training of adult dyads undergoing a dyadic behavior change.

43.5.5 Intensiveness Systematic information on dyadic intervention intensity or dose-response relationships is lacking at this point. This again is mainly due to the heterogeneity in dyadic behavior change interventions. For example, a systematic review of parentchild dyadic interventions focusing on sex behaviors reports one study that resulted in significant reductions in sexual risk behaviors twenty-four months later when a single parent-child session was added to six youth-only sessions (Sutton et al., 2014). In contrast, a single session of dyadic planning in adult couples did not result in the hypothesized superior effects compared to individual planning or control conditions (Knoll et al., 2017). Future research is needed to evaluate effects of different intensity and dose-response relationships for different target populations, target behaviors, dyadic constellations, and DBCTs.

43.5.6 Evaluation of Fidelity Evaluation of fidelity in dyadic interventions is facilitated by the basic availability of two sources of information on the dyadic and individual engagement with the dyadic techniques. For example, both dyad members can be asked to report on their own and the other’s intervention fidelity. However, this can also create challenges as discrepancies between dyad members’ reports are not easily reconciled. Further, the dyad may agree not to confess low fidelity with the intervention. As in individual interventions, assessment of fidelity in dyadic treatments is often easier when the intervention is delivered with interventionist and dyad

members present. When DBCTs are delivered via remote digital interventions, the evaluation of the fidelity can be objectively assessed – for example, data on app use (e.g., see Lüscher et al., 2019).

43.5.7 Evaluation of Effectiveness Assessing the target behavior in both partners in a dyad is highly recommended, even if the intervention only focuses on behavior change in one individual within the dyad (e.g., Knoll et al., 2017). This allows for evaluation of intervention and transfer effects in both dyad members. Further, joint behavioral change might work as an enabling factor, strengthening dyadic- and self-efficacy, social support, and companionship in the dyad. As for individual BCTs, it is also recommended to assess the assumed mechanisms of action for DBCTs. For example, in collaborative implementation intentions (Prestwich et al., 2005; Radtke et al., 2018), these might not only be the higher quality of plans but also social support, modeling, or more frequent reminders (see Table 43.1). All potential mechanisms of action should be assessed in both dyad members. Moreover, developments in information and communication technology may allow for objective assessment of behavior outcomes in addition to self-report. For example, both dyad members can wear activity trackers or use their smartphones for tracking activity and then share this information with the interventionist.

43.5.8 Typical Materials Needed Current evidence indicates that reporting of intervention content and materials used in dyadic intervention studies is underdeveloped (see Section 43.4). Future research needs to provide more information on the materials used and ideally provide examples of these materials in supplemental material of published studies or provide open access to these materials via open science repositories.

Dyadic Behavior Change Interventions

43.5.9 Typical Examples of Implementation Owing to the heterogeneity of DBCTs (see examples in Table 43.1), there is no “typical” implementation example for dyadic interventions. This section provides two example dyadic interventions: a “crossover” DBCT, dyadic action control intervention applied to romantic couples (Berli et al., 2016; Scholz & Berli, 2014) and a “joint” DBCT, collaborative planning intervention (Prestwich et al., 2005) applied to adolescent best-friends dyads (Radtke et al., 2018).

43.5.9.1 Dyadic Action Control Dyadic action control is defined as the involvement of a dyad member in the process of action control by prompting the target person’s behavior standards, monitoring of actual behavior, and, if necessary, increased self-regulatory effort (Berli et al., 2016). In addition to increased levels of action control, this technique is expected to elicit social support and positive social control and thus to be superior in effects compared to action control at the individual level. In a dyadic action control intervention, dyad members are instructed to save personalized drafts of action control text messages alternately addressing the three action control components (awareness of standards, selfmonitoring, self-regulatory effort) on their smartphone during a face-to-face session between interventionist and partner but without the target person. Partners are then reminded via automated text messages to send the saved text messages to the target person’s smartphone during the course of the intervention. Partners should not change the order and the content of the messages. Examples of text messages can be found in Table A43.1.1 (see Appendix 43.1, supplemental materials). In addition, because action control is a volitional behavior change process that is based on behavioral goals, goal setting serves as an important prerequisite for the main action control intervention (see Chapter 38, this volume). To strengthen

643

commitment with behavioral goals and ensure sufficient goal quality, face-to-face delivery by the interventionist is ideal, with dyad partners assisting target persons in setting personal goals for the new behavior. The dyadic action control intervention was conducted and evaluated with romantic couples. However, dyads with at least one member reporting an intention to change a behavior and the other willing to assist can be considered an eligible target audience. Given that action control strategies are particularly beneficial for individuals who are already motivated to change the behavior, the target persons need to have a strong intention already (see Chapters 6 and 39, this volume). Consequently, this intervention is not suitable for dyads with insufficient motivation to change a behavior or to assist in the change. The effectiveness of the dyadic action control intervention was tested in a study on overweight couples (N = 121 couples) intending to become physically active (Berli et al., 2016). The technique “dyadic action control” was implemented as a dyadic variation of the intervention for couples and compared to an individual variation of the intervention (target persons set their behavioral goals individually and text messages were sent to target persons’ smartphones from the study staff via an automated system) in two intervention groups. Overall, the action control intervention effectively enhanced the daily adherence to physical activity recommendations in 36.5 percent of the intervention group participants compared to 23 percent of control participants, who were in a no-treatment condition that did not set any behavioral goals and received neutral text messages with no action control content. However, no additional benefit of the dyadic variation of the action control intervention compared to the individual variation emerged (Berli et al., 2016). This suggests that the dyadic action control technique as applied in this study was no more effective than the individual action control condition in which the study team sent the text messages without involvement of the partner.

644

¨ SCHER, AND NINA KNOLL URTE SCHOLZ, CORINA BERLI, JANINA LU

Various reasons for the lack of difference in effectiveness for the individual- versus dyadicfocused intervention were identified (see Berli et al., 2016): (1) the manipulation of dyadic action control may not have been “strong” enough (e.g., partner could only personalize messages); (2) the individual action control also had some dyadic components (e.g., target persons were also prompted by “others,” i.e., a member of the study team, to engage in action control); (3) the effect of the “mere presence” of the partner could not be gauged, as well as the effect of parallel techniques (partners also receiving health information) that may have increased social support in the individual intervention condition; (4) the parameters of effectiveness are unclear so far (e.g., relationship quality may play an important role for effectiveness); and (5) the comparison between the two intervention groups was underpowered (Berli et al., 2016). Thus, aside from addressing the issue of statistical power in future studies testing dyadic action control, optimized dyadic action control interventions could, for example, introduce a stronger personal involvement of partners in the intervention. In addition to the action control component, partners could, for instance, be instructed to closely tailor the content of the messages to their partner’s needs. Nevertheless, given that the dyadic action control intervention was equally effective as the individual action control intervention compared to a control group, dyadic action control is a tool that can be used by practitioners and interested dyads for promoting behavior change in one dyad member. It can be applied to a wide range of behaviors and provides a simple way of how to involve close others in behavior change efforts.

43.5.9.2 Collaborative Planning Collaborative planning is the planning and enacting of a behavior together with a dyad partner (Prestwich et al., 2005; Prestwich et al., 2012; Radtke et al., 2018). The effects of collaborative

planning are assumed to be driven by the same mechanisms as individual planning interventions (Gollwitzer, 1999; see also Chapter 39, this volume) and, additionally, by perceived social influences, dyad partner support, increased enjoyment of the new behavior, intentions, self-efficacy (Prestwich et al., 2014), and more reminders (Prestwich et al., 2005). In the studies by Prestwich and colleagues, the intervention was typically delivered as written text to individuals instructing them to find a partner (e.g., romantic partner, housemate, etc.; Prestwich et al., 2014) and engage in collaborative planning of the target behavior with this partner (i.e., planning to enact the behavior together). In the supplemental material of this chapter (see Appendix 43.2), an example of a collaborative planning sheet is provided for the example of a collaborative planning intervention in adolescent best-friend dyads (Radtke et al., 2018). So far, the target audiences of collaborative planning interventions have been adult (e.g., Prestwich et al., 2005; Prestwich et al., 2012) and adolescent (Radtke et al., 2018) romantic and peer dyads. Like action control, planning is a volitional strategy that requires a high motivation to change a behavior in dyad members. Thus, comparable to the dyadic action control intervention, collaborative planning is most suitable for people with high motivation and potentially unhelpful for people with no motivation to change the behavior. Collaborative planning resulted in positive effects on the target behavior (e.g., 100 percent achievers) compared to an individual planning condition (63 percent achievers; Prestwich et al., 2005). Another study reported moderate effects over and above an individual planning condition and a control group up to six months after the intervention (Prestwich et al., 2012). A third study reported rather inconclusive effects of collaborative planning compared to a partner-only and control conditions (Prestwich et al., 2014).

Dyadic Behavior Change Interventions

Overall, collaborative planning seems to be an effective and easy-to-administer joint dyadic intervention strategy that is suitable for different dyadic constellations. Promising evidence exists for adult dyads (Prestwich et al., 2005; Prestwich et al., 2012), with an ongoing study evaluating effectiveness in adolescent best-friend dyads (Radtke et al., 2018). Sample intervention materials of the dyadic action control intervention (Berli et al., 2016) and the collaborative planning intervention in adolescent best-friend dyads (Radtke et al., 2018) are provided in the supplemental materials (see Appendix 43.1 for the dyadic action control intervention and Appendix 43.2 for the collaborative planning intervention). In future, it is expected that providing access to intervention materials of dyadic interventions will be far more common, which will provide interventionists with a richer selection of intervention methods and implementation examples of dyadic interventions.

43.6 Conclusion Dyadic interventions for behavior change are a growing field of research taking into account that most behaviors and behavior changes happen in a social environment and are thus influenced by others. Given the omnipresence and importance of close others in people’s lives, capitalizing on dyadic influences for behavior change is a promising avenue for intervention. The current evidence for dyadic interventions, however, is inconclusive. There is a strong need for high-quality intervention studies and systematic research adequately addressing the heterogeneity of dyadic interventions with regard to the study population; the target behaviors; the dyadic constellations; the dyadic behavior change techniques (including a common language for describing the DBCTs consistently); the goal constellations in dyads; and the parameters that might increase or reduce the success of dyadic interventions.

645

References Antonucci, T. C., Akiyama, H., & Takahashi, K. (2004). Attachment and close relationships across the life span. Attachment and Human Development, 6, 353–370. https://doi.org/ 10.1080/1461673042000303136 Arden-Close, E., & McGrath, N. (2017). Health behaviour change interventions for couples: A systematic review. British Journal of Health Psychology, 22, 215–237. https://doi.org/10.1111/ bjhp.12227 Bandura, A. (1989). Human agency in social cognitive theory. American Psychologist, 44, 1175–1184. https://doi.org/10.1037/0003-066X.44.9.1175 Baucom, D. H., Kirby, J. S., & Kelly, J. T. (2010). Couple-based interventions to assist partners with psychological and medical problems. In K. Hahlweg, M. Grawe-Gerber, & D. H. Baucom (Eds.), Enhancing Couples: The Shape of Couple Therapy to Come (pp. 78–92). Cambridge, MA: Hogrefe Publishing. Baucom, D. H., Porter, L. S., Kirby, J. S., & Hudepohl, J. (2012). Couple-based interventions for medical problems. Behavior Therapy, 43, 61–76. https:// doi.org/10.1016/j.beth.2011.01.008 Berli, C., Bolger, N., Shrout, P. E., Stadler, G., & Scholz, U. (2018). Interpersonal processes of couples’ daily support for goal pursuit: The example of physical activity. Personality and Social Psychology Bulletin, 44, 332–344. https:// doi.org/10.1177/0146167217739264 Berli, C., Stadler, G., Inauen, J., & Scholz, U. (2016). Action control in dyads: A randomized controlled trial to promote physical activity in everyday life. Social Science & Medicine, 163, 89–97. https:// doi.org/10.1016/j.socscimed.2016.07.003 Best, J. R., Goldschmidt, A. B., Mockus-Valenzuela, D. S., Stein, R. I., Epstein, L. H., & Wilfley, D. E. (2016). Shared weight and dietary changes in parent-child dyads following family-based obesity treatment. Health Psychology, 35, 92–95. https:// doi.org/10.1037/hea0000247 Bolger, N., Zuckerman, A., & Kessler, R. C. (2000). Invisible support and adjustment to stress. Journal of Personality and Social Psychology, 79, 953– 961. https://doi.org/10.1037/0022-3514.79.6.953

646

¨ SCHER, AND NINA KNOLL URTE SCHOLZ, CORINA BERLI, JANINA LU

Bronfenbrenner, U. (1986). Ecology of the family as a context for human development: Research perspectives. Developmental Psychology, 22, 723–742. https://doi.org/10.1037/0012-1649 .22.6.723 Burkert, S., Scholz, U., Gralla, O., Roigas, J., & Knoll, N. (2011). Dyadic planning of health-behavior change after prostatectomy: A randomizedcontrolled planning intervention. Social Science and Medicine, 73, 783–792. https://doi.org/ 10.1016/j.socscimed.2011.06.016 Burton, J., Darbes, L. A., & Operario, D. (2010). Couples-focused behavioral interventions for prevention of HIV: Systematic review of the state of evidence. AIDS and Behavior, 14, 1–10. https:// doi.org/10.1007/s10461-008-9471-4 Butterfield, R. M., & Lewis, M. A. (2002). Healthrelated social influence: A social ecological perspective on tactic use. Journal of Social and Personal Relationships, 19, 505–526. https://doi .org/10.1177/0265407502019004050 Carr, R. M., Prestwich, A., Kwasnicka, D. et al. (2019). Dyadic interventions to promote physical activity and reduce sedentary behaviour: Systematic review and meta-analysis. Health Psychology Review, 13, 91–109. https://doi.org/10.1080/ 17437199.2018.1532312 Craddock, E., vanDellen, M. R., Novak, S. A., & Ranby, K. W. (2015). Influence in relationships: A metaanalysis on health-related social control. Basic and Applied Social Psychology, 37, 118–130. https:// doi.org/10.1080/01973533.2015.1011271 Crepaz, N., Tungol-Ashmon, M. V., Vosburgh, H. W., Baack, B. N., & Mullins, M. M. (2015). Are couple-based interventions more effective than interventions delivered to individuals in promoting HIV protective behaviors? A metaanalysis. AIDS Care, 27, 1361–1366. https://doi .org/10.1080/09540121.2015.1112353 Cutrona, C. E., & Russell, D. W. (1990). Type of social support and specific stress: Toward a theory of optimal matching. In B. R. Sarason, I. G. Sarason, & G. R. Pierce (Eds.), Wiley Series on Personality Processes. Social Support: An Interactional View (pp. 319–366). Oxford: John Wiley & Sons. Dombrowski, S. U., O’Carroll, R. E., & Williams, B. (2016). Form of delivery as a key “active

ingredient” in behaviour change interventions. British Journal of Health Psychology, 21, 733– 740. https://doi.org/10.1111/bjhp.12203 Faseru, B., Richter, K. P., Scheuermann, T. S., & Park, E. W. (2018). Enhancing partner support to improve smoking cessation. The Cochrane Database of Systematic Reviews, 8, CD002928. https://doi.org/ 10.1002/14651858.CD002928.pub4 Fitzsimons, G. M., Finkel, E. J., & vanDellen, M. R. (2015). Transactive goal dynamics. Psychological Review, 122, 648–673. https://doi.org/10.1037/ a0039654 Gleason, M. E. J., Iida, M., Bolger, N., & Shrout, P. E. (2003). Daily supportive equity in close relationships. Personality and Social Psychology Bulletin, 29, 1036–1045. https://doi.org/10.1177/ 0146167203253473 Gollwitzer, P. M. (1999). Implementation intentions: Strong effects of simple plans. American Psychologist, 54(7), 493–503. https://doi.org/ 10.1037/0003-066X.54.7.493 Hagger, M. S., & Hardcastle, S. J. (2014). Interpersonal style should be included in taxonomies of behavior change techniques. Frontiers in Psychology, 5, 254. https://doi.org/10.3389/ fpsyg.2014.00254 Hoffmann, T. C., Glasziou, P. P., Boutron, I. et al. (2014). Better reporting of interventions: Template for intervention description and replication (TIDieR) checklist and guide. BMJ, 348, g1687. https://doi.org/10.1136/bmj.g1687 Holt-Lunstad, J., Smith, T. B., & Layton, J. B. (2010). Social relationships and mortality risk: A metaanalytic review. PLoS Medicine, 7, e1000316. Inauen, J., Tobias, R., & Mosler, H.-J. (2014). The role of commitment strength in enhancing safe water consumption: Mediation analysis of a clusterrandomized trial. British Journal of Health Psychology, 19, 701–719. https://doi.org/10.1111/ bjhp.12068 Jackson, S. E., Steptoe, A., & Wardle, J. (2015). The influence of partner’s behavior on health behavior change: The English Longitudinal Study of Ageing. JAMA Internal Medicine, 175, 385–392. https://doi.org/10.1001/jamainternmed.2014.7554 Keller, J., Fleig, L., Hohl, D. H. et al. (2017). Which characteristics of planning matter? Individual and

Dyadic Behavior Change Interventions

dyadic physical activity plans and their effects on plan enactment. Social Science and Medicine, 189, 53–62. https://doi.org/10.1016/j.socscimed .2017.07.025 Knoll, N., Hohl, D. H., Keller, J., Schuez, N., Luszczynska, A., & Burkert, S. (2017). Effects of dyadic planning on physical activity in couples: A randomized controlled trial. Health Psychology, 36, 8–20. https://doi.org/10.1037/ hea0000423 Knoll, N., Scholz, U., & Ditzen, B. (2019). Social support and family processes. In T. A. Revenson & R. A. R. Gurung (Eds.), The Handbook of Health Psychology (pp. 279–289). New York: Psychology Press. Kok, G., Gottlieb, N. H., Peters, G.-J. Y. et al. (2016). A taxonomy of behaviour change methods: An intervention mapping approach. Health Psychology Review, 10, 297–312. https://doi.org/ 10.1080/17437199.2015.1077155 Lewis, M. A., McBride, C. M., Pollak, K. I., Puleo, E., Butterfield, R. M., & Emmons, K. M. (2006). Understanding health behavior change among couples: An interdependence and communal coping approach. Social Science and Medicine, 62, 1369–1380. https://doi.org/10.1016/j .socscimed.2005.08.006 Lüscher, J., Berli, C., Schwaninger, P., & Scholz, U. (2019). Smoking cessation with smartphone applications (SWAPP): A randomized controlled trial. Manuscript submitted for publication. Lüscher, J., Stadler, G., Ochsner, S. et al. (2015). Daily negative affect and smoking after a self-set quit attempt: The role of dyadic invisible social support in a daily diary study. British Journal of Health Psychology, 20, 708–723. https://doi.org/ 10.1111/bjhp.12135 Martire, L. M., & Helgeson, V. S. (2017). Close relationships and the management of chronic illness: Associations and interventions. American Psychologist, 72, 601–612. https://doi.org/ 10.1037/amp0000066 McLean, N., Griffin, S., Toney, K., & Hardeman, W. (2003). Family involvement in weight control, weight maintenance and weight-loss interventions: A systematic review of randomised trials. International Journal of Obesity, 27, 987– 1005. https://doi.org/10.1038/sj.ijo.0802383

647

Michie, S., Carey, R. N., Johnston, M. et al. (2018). From theory-inspired to theory-based interventions: A protocol for developing and testing a methodology for linking behaviour change techniques to theoretical mechanisms of action. Annals of Behavioral Medicine, 52, 501– 512. https://doi.org/10.1007/s12160-016-9816-6 Michie, S., Richardson, M., Johnston, M. et al. (2013). The behavior change technique taxonomy (v1) of 93 hierarchically clustered techniques: Building an international consensus for the reporting of behavior change interventions. Annals of Behavioral Medicine, 46, 81–95. https://doi.org/ 10.1007/s12160-013-9486-6 Prestwich, A., Conner, M., Lawton, R., Bailey, W., Litman, J., & Molyneaux, V. (2005). Individual and collaborative implementation intentions and the promotion of breast self-examination. Psychology and Health, 20, 743–760. https://doi .org/10.1080/14768320500183335 Prestwich, A., Conner, M. T., Lawton, R. J., Ward, J. K., Ayres, K., & McEachan, R. R. C. (2012). Randomized controlled trial of collaborative implementation intentions targeting working adults’ physical activity. Health Psychology, 31, 486–495. https://doi.org/ 10.1037/a0027672 Prestwich, A., Conner, M. T., Lawton, R. J., Ward, J. K., Ayres, K., & McEachan, R. R. C. (2014). Partner- and planning-based interventions to reduce fat consumption: Randomized controlled trial. British Journal of Health Psychology, 19, 132–148. https://doi.org/ 10.1111/bjhp.12047 Radtke, T., Luszczynska, A., Schenkel, K., Biddle, S., & Scholz, U. (2018). A cluster randomized controlled trial comparing the effectiveness of an individual planning intervention with collaborative planning in adolescent friendship dyads to enhance physical activity (TWOgether). BMC Public Health, 18, 911. https://doi.org/ 10.1186/s12889-018-5818-6 Rafaeli, E., & Gleason, M. E. J. (2009). Skilled support within intimate relationships. Journal of Family Theory and Review, 1, 20–37. https://doi.org/ 10.1111/j.1756-2589.2009.00003.x Ramchand, R., Ahluwalia, S. C., Xenakis, L., Apaydin, E., Raaen, L., & Grimm, G. (2017). A systematic

648

¨ SCHER, AND NINA KNOLL URTE SCHOLZ, CORINA BERLI, JANINA LU

review of peer-supported interventions for health promotion and disease prevention. Preventive Medicine, 101, 156–170. https://doi.org/10.1016/ j.ypmed.2017.06.008 Reis, H. T., & Gable, S. L. (2015). Responsiveness. Current Opinion in Psychology, 1, 67–71. https:// doi.org/10.1016/j.copsyc.2015.01.001 Richards, E. A., Franks, M. M., McDonough, M. H., & Porter, K. (2017). “Let’s move”: A systematic review of spouse-involved interventions to promote physical activity. International Journal of Health Promotion and Education, 56, 1–17. https://doi.org/10.1080/14635240.2017.1415160 Rook, K. S. (1990). Support, companionship, and control in older adults’ social networks: Implications for well-being. In M. A. Stephens, J. H. Crowther, S. E. Hobfoll, & D. L. Tennenbaum (Eds.), Stress and Coping in Later-Life Families (pp. 437–463). New York: Hemisphere. Scholz, U., & Berli, C. (2014). A dyadic action control trial in overweight and obese couples (DYACTIC). BMC Public Health, 14, 1321. https://doi.org/10.1186/1471-2458-14-1321 Schulz, R., Czaja, S. J., McKay, J. R., Ory, M. G., & Belle, S. H. (2010). Intervention taxonomy (ITAX): Describing essential features of interventions. American Journal of Health Behavior, 34, 811–821. https://doi.org/10.5993/AJHB.34.6.15 Seidman, G., Shrout, P. E., & Bolger, N. (2006). Why is enacted social support associated with increased distress? Using simulation to test two possible sources of spuriousness. Personality and Social

Psychology Bulletin, 32, 52–65. https://doi.org/ 10.1177/0146167205279582 Sniehotta, F. F., Nagy, G., Scholz, U., & Schwarzer, R. (2006). The role of action control in implementing intentions during the first weeks of behaviour change. British Journal of Social Psychology, 45, 87–106. https://doi.org/10.1348/014466605X62460 Sullivan, K. T., & Davila, J. (2014). The problem is my partner: Treating couples when one partner wants the other to change. Journal of Psychotherapy Integration, 24, 1–12. https://doi.org/10.1037/ a0035969 Sutton, M. Y., Lasswell, S. M., Lanier, Y., & Miller, K. S. (2014). Impact of parent-child communication interventions on sex behaviors and cognitive outcomes for black/African-American and Hispanic/Latino youth: A systematic review, 1988–2012. The Journal of Adolescent Health, 54, 369–384. https://doi.org/10.1016/j.jadohealth .2013.11.004 Voils, C. I., Coffman, C. J., Yancy, W. S. et al. (2013). A randomized controlled trial to evaluate the effectiveness of CouPLES: A spouse-assisted lifestyle change intervention to improve lowdensity lipoprotein cholesterol. Preventive Medicine, 56, 46–52. https://doi.org/10.1016/j .ypmed.2012.11.001 Webel, A. R., Okonsky, J., Trompeta, J., & Holzemer, W. L. (2010). A systematic review of the effectiveness of peer-based interventions on health-related behaviors in adults. American Journal of Public Health, 100, 247–253.

44 Social Identity Interventions Mark Tarrant, Catherine Haslam, Mary Carter, Raff Calitri, and S. Alexander Haslam

Practical summary The use of social groups to deliver behavior change interventions has grown markedly in recent years. However, this growth has generally not been informed by theory and research relevant to the social psychological processes that play an important role when people come together in a group. This chapter describes an evidence-based approach to behavior change interventions that are intended for delivery in group settings. A set of guidelines for practitioners is presented that specify key areas of activity important to the design and delivery of interventions informed by this approach. These guidelines describe key actions that practitioners working to change the behavior of people in group settings can apply to encourage the formation of meaningful social connections between intervention recipients so that the group becomes an effective resource for change. A worked example from a recent application of this approach is provided to supplement the guide.

44.1 Introduction The social identity approach, comprising social identity and self-categorization theories (Tajfel & Turner, 1986; Turner et al., 1987), offers a comprehensive account of the psychological processes that impact on the dynamics that occur within and between social groups. While originally developed to explain people’s behaviors and cognitions in the context of negative intergroup relations (e.g., prejudice, discrimination), the approach has evolved to also address important questions associated with wider social phenomena that include the psychology of crowds and mass response to emergency situations, leadership and organizational mergers, and, more recently, health cognitions together with associated behaviors and outcomes. Yet the approach is not limited to the explanation of the consequences of social group

membership; it also has great translational utility as a result of its capacity to inform and structure the design of group interventions focused on changing people’s behavior. This chapter presents a new model, the social identity model of behavior change, which builds on a recently developed core set of principles and hypotheses that pertain to the emergence of shared social identity and its psychological and behavioral consequences (see Haslam et al., Aspects of the work presented in this chapter were supported by a grant from the Stroke Association (QQ12 TSA 2016/14) and by the National Institute for Health Research Applied Research Collaboration South West Peninsula. Alex Haslam’s contribution was supported by an ARC Australian Laureate Fellowship (FL110100199). We thank Plymouth Music Zone (PMZ) and Anna Batson for their support with the trial, and the contributions of the research participants and stakeholder groups who supported the project. https://doi.org/10.1017/9781108677318.044

650

MARK TARRANT, CATHERINE HASLAM, MARY CARTER, RAFF CALITRI, AND S. ALEXANDER HASLAM

2018). These are relevant not only in health contexts but also in other behavioral domains (e.g., environmental and organizational behavior, education). The theoretical underpinnings of the social identity approach are described earlier in this volume (Chapter 16, this volume). In this chapter, this reasoning is applied to a new set of guidelines to help practitioners develop and manage effective group interventions for behavior change. Researchers and practitioners who use the guide will need to tailor their interventions to take account of inevitable complexity associated with a range of contextual factors, including the behavior and population targeted in intervention, and the implementation setting.1

44.2 Definitions 44.2.1 The Social Identity Approach to Group-Based Behavior Change Interventions Haslam et al. (2018) identified fifteen core hypotheses that specify relationships between group membership, social identity, and health. Seven are of particular relevance to group-based behavior change interventions and primary among these is the identification hypothesis. This states that “a person will generally experience the healthrelated benefits or costs of a given group membership only to the extent that they identify with that group” (p. 17). Among other things, this means that, if intervention recipients define themselves in terms of a shared social identity (as members of the intervention group), the group has the potential to facilitate access to a range of psychological resources that support behavior change. These 1

2

The issue of complexity in group-based intervention is considered in Appendix 44.1 (supplemental materials). Other parameters associated with the intervention design (e.g., open vs. closed groups), or target recipients (e.g., member similarity), may also shape intervention effectiveness, through their impact on social identification (see Cruwys et al., 2019; Khan et al., 2019).

resources form the “active ingredients” of groupbased behavior change interventions and are represented by the hypotheses specified in Table 44.1. These hypotheses suggest that, to the extent that individuals come to share a social identity as members of the intervention group, participation in the group will imbue in them a sense of meaning that contributes purpose and direction to their continued participation or commitment (the meaning hypothesis); encourage them to see themselves as similar to, and to trust, one another (the connection hypothesis); and allow the group to become a resource for social support (the support hypothesis). Furthermore, shared social identity encourages the enactment of group norms relating to change behaviors (e.g., motivating continued attendance at intervention sessions and pursuit of change goals – the norm enactment hypothesis) and transforms the group intervention into a basis for mutual social influence (the potential for group members to influence each other – e.g., through advice-giving and the sharing of intervention materials: the influence hypothesis). Finally, shared social identity provides a foundation for group members to build selfefficacy in pursuit of change goals – for example, by observing fellow members perform target behaviors (the agency hypothesis). Social identification is, therefore, a critical “parameter” of group intervention effectiveness (see Kok et al., 2016; see also Chapters 19 and 20, this volume) because – by enabling the group processes outlined in Table 44.1 – the group itself becomes the instantiation of these resources that enhances the motivation and capability required for intervention recipients to change behavior (see Michie, van Stralen, & West, 2011). This parameter and associated resources are represented schematically in the model depicted in Figure 44.1.2 The model proposes that the effectiveness of group interventions depends on first establishing a sense of shared and positive social identity among members of the intervention group. Rather than imagining that group processes conducive to behavior change will emerge naturally

Social Identity Interventions

651

Table 44.1 Six hypotheses from the social identity approach to health (Haslam et al., 2018) that can make group-based interventions a powerful resource for behavior change Hypothesis

Definition

Meaning

When, and to the extent that, people define themselves in terms of shared social identity, that identity will focus their energies and imbue them with a sense of meaning, purpose, and worth. When, and to the extent that, people define themselves in terms of shared social identity, they will be more likely to perceive themselves as similar and connected and to be positively oriented toward (e.g., trusting of) each other. When, and to the extent that, people define themselves in terms of shared social identity, they will (1) expect to give each other support, (2) actually give each other support, and (3) construe the support they receive more positively. When, and to the extent that, a person defines themselves in terms of a given social identity, they will enact – or at least strive to enact – the norms and values associated with that identity. When, and to the extent that, people define themselves in terms of shared social identity, they will be more likely to influence each other. When, and to the extent that, a group of people define themselves in terms of shared social identity, they will develop a sense of collective efficacy, agency and power.

Connection

Support

Norm enactment

Influence Agency

within group interventions or leaving this to chance, this observation points to key social identity–building processes that are critical to the design and delivery of group-based interventions. Indeed, it is suggested that the heterogeneity in effectiveness that has been observed across different group interventions (e.g., Borek et al., 2018) may be attributed, at least in part, to a general failure to take account of variability in social identity management (for meta-analytic evidence that supports this claim, see Steffens et al., 2019).

44.2.2 Building Shared Social Identity How do individual group members come to see themselves in terms of a shared social identity? According to self-categorization theory, a social identity becomes salient for individuals as a result of two interactive processes: perceiver “readiness” and “fit” (Oakes, Haslam & Turner, 1994). Readiness relates to the fact that a person is more likely to self-categorize in terms of a given group

membership to the extent that they are predisposed to do so – for example, because they have prior experience of acting in terms of that selfcategory and if it is meaningful to them. Fit relates to the fact that a person is more likely to self-categorize in terms of a given group membership when (1) differences between members of the group are seen to be smaller than differences between that group and other groups that are salient in a given comparative context (comparative fit) and (2) this pattern of category difference matches the perceiver’s content-related expectations (normative fit). As a hypothetical example, imagine that John, who is morbidly obese, joins a group-based weight-management program. Whether or not John sees himself (i.e., self-categorizes) as a member of the group will depend on processes of readiness and fit. This means that John is more likely to self-categorize as a member of the group if he has previous experience with the group – for example, if he knows other members or if he has been part of a weight-management group before.

652

MARK TARRANT, CATHERINE HASLAM, MARY CARTER, RAFF CALITRI, AND S. ALEXANDER HASLAM

Associated group processes

Motivation to advance the group and its goals

Group Members Opportunity to create and consolidate shared social identity

Group Facilitators (leaders)

Reflect Observe the group in order to understand it

Behavior 1. Meaning 2. Connectedness 3.Emergent norms, values and goals 4. Support 5. Infuence 6. Agency

that advances the group and its goals Capability to advance the group and its goals

Realize

Represent

Deliver outcomes that matter for the group

Work with the group to advance its values

Figure 44.1 The social identity model of behavior change

In addition, John is also more likely to self-categorize as a member of the weight-management group if he sees, and expects to see, that the people in the group have a lot in common (e.g., all seeking help with weight management) and are clearly different from the other people around them in ways that correspond with the group’s putative meaning (e.g., if they are all morbidly obese). On the other hand, John is less likely to see himself as a member of the group if he has no experience of weight-management groups like this one (or perhaps has only had “bad experiences”) and if the group members seem to have little in common to define them as different from other people (e.g., if they are no more overweight than other patient groups). Questions of social identity salience are important because group-based interventions can provide a foundation for behavior change only when recipients identify with the group in question and when they see themselves as sharing this social category membership with other group members. Accordingly, it is important for practitioners to attend to, and manage, these processes to encourage engagement with the group and with the courses of action that it prescribes. The significance of this point in intervention settings was

highlighted in a study that showed that perceived conflict between recipients of a groupbased weight-loss program was associated with poorer intervention adherence at six months, lower levels of program attendance, and smaller weight loss (Nackers et al., 2015). Nackers et al.’s (2015) study illustrates that failure to attend to the group processes that play out in group settings can lead to negative outcomes – for example, because a lack of shared identity creates conflict and schism within the group (Sani, 2005). Working to build shared social identity – in ways that promote both group harmony and group commitment – is thus critical to intervention effectiveness. In this regard, the role played by those with responsibility for facilitating the intervention group cannot be overstated (see Reicher, Haslam, & Hopkins, 2005; Steffens et al., 2014). The step-by-step guide provided in Section 44.5 elaborates the actions that group facilitators can take to achieve this objective.

44.3 Theory of Change In the absence of a shared social identity, there is no functional group – only a collection of individuals who are primarily concerned with their own

Social Identity Interventions

individual interests and goals and not those of the wider group. Such intervention groups can provide only limited opportunity for behavior change. Yet when, and to the extent that, members internalize the group as part of the self, then they and the group become cognitively interchangeable, through a process that Turner (1982) refers to as depersonalization. Depersonalization leads individuals to define themselves in terms of the group – so that the group acquires the capacity to shape their behavior and they, in turn, are able to exert influence on the group. In the social identity model of behavior change, it is this process of depersonalization that transforms the group into a psychological reality – and provides the basis for group-based motivation and capability for individual members to pursue shared behavioral goals. It should be clear from this discussion that the model offers an integrated account of behavior change in group settings by formalizing the contribution of shared social identity in shaping the delivery of wider intervention content that targets specific behavioral goals. It also follows from this that the model is not advocating for radically different intervention content (i.e., “better” educational material) when used in a group setting but, rather, specifies how – by attending to social identity processes – that content can be more effectively delivered so as to become a resource for change when intervention recipients self-categorize as group members.

44.4 Evidence Base In line with the norm enactment and meaning hypotheses, several studies have shown that, the more a person identifies as a member of a particular social group, and the more salient that identity is, the more likely it is that they will act in the interests of that group and in accordance with its norms and values – and the more they will derive a sense of worth and purpose from the group as a result. For example, Oyserman, Fryberg, and Yoder (2007) refer to an “identity-motivation” process whereby group members articulate values and normative

653

beliefs that they perceive to be aligned with those of their in-group – even when the behaviors flowing from these cognitions may have negative consequences for them as individuals (e.g., not seeking help from a health care professional when ill-health symptoms are perceived; Kearns et al., 2015). Speaking to this point, patients with obesity interviewed in Tarrant and colleagues’ (2017) study emphasized how their commitment to developing a new perspective on health was shaped by other group members (the influence hypothesis), with several also referencing the social support derived from their group (the support hypothesis; see also Haslam, Reicher, & Levine, 2012). Beyond this, patients talked about how the connections they developed with other members through the intervention (the connection hypothesis) contributed to an increased sense of belonging that supported their well-being. One way in which social support exerts these positive effects is through its capacity to act as a catalyst for individual action (as predicted by the agency hypothesis). Following Bandura’s (1994; see also Chapter 3, this volume) observation that individual self-efficacy is influenced by observing similar others perform the same action, the social identity approach to health holds that seeing others with whom one shares a social identity perform an action may have a similar impact on efficacy beliefs. Empirical research also supports this prediction. Smith and Woodworth (2012) showed how, by working alongside others with whom they share a social identity, students’ group membership promotes a sense of ability to meet curricular aims. As these studies highlight, self-efficacy experienced at the individual level is promoted by the collective efficacy of the group (see Frings et al., 2016). While numerous studies support the hypothesized effects of social identification on group processes (e.g., Greenaway et al., 2015; Oyserman et al., 2007), there is scope for more research into the specific effects that the six resources specified in the model have on behavior change outcomes (e.g., through trial process evaluations; see

654

MARK TARRANT, CATHERINE HASLAM, MARY CARTER, RAFF CALITRI, AND S. ALEXANDER HASLAM

Section 44.5.3; cf. Haslam et al., 2016). Doing so requires practitioners to develop interventions that embed social identity–building principles, and the step-by-step guide that follows details how this can be achieved when designing, delivering, and evaluating group-based interventions.

44.5 A Step-by-Step Guide to Building Social Identity Interventions Group interventions based on the social identity approach are informed by three key leadership processes of reflecting, representing, and realizing shared social identity among intervention recipients (see Haslam, Reicher, & Platow, 2011) and can be operationalized in terms of the following three practitioner points depicted in the lower part of Figure 44.1: •

•

•

Practitioner point 1 (“identity reflection”): Reflect on the things that matter to the intervention group – so as not to take its members (and their views and understandings) for granted. Practitioner point 2 (“identity representation”): Work with the group to help clarify its values and goals – in ways that align intervention recipients’ shared social identity with intervention objectives. Practitioner point 3 (“identity realization”): Take steps to make the group matter – to maximize the potential for it to become a resource that promotes behavior change.

44.5.1 Target Audience: Preparing for Change Illustrating practitioner point 1, planning for group interventions – like other complex interventions – needs to involve a thorough understanding of the “problem” and the population that is being targeted by the intervention (see Chapters 19 and 20, this volume). People often join groups with diverse past experiences and preferences (e.g., different backgrounds, different health issues)

and this diversity will shape members’ readiness to self-categorize as members of the intervention group (Cruwys et al., 2019; Khan et al., 2019). By understanding diversity, group facilitators can plan how to organize intervention content in ways that meet individual needs and expectations. A variety of research and engagement activities can inform this planning, such as syntheses of existing research (e.g., systematic reviews), secondary analyses of existing data, and primary research (see Chapter 21, this volume). Stakeholder meetings, including with members of the target population and those who commission or facilitate the proposed intervention, should also inform key decisions around intervention design (e.g., likely acceptability of intervention procedures, barriers to implementation) and contribute to facilitators’ appreciation of group heterogeneity (see also Chapter 24, this volume). Prior to starting a group program, meetings with intervention recipients (e.g., clinical review) can help facilitators further understand individual trajectories into the group. These activities should also ensure that, as far as possible, the intervention and the group itself are coproduced rather than simply imposed on recipients. Illustrating practitioner point 2, once the intervention has commenced, facilitators should aim to nurture a group environment that encourages members to see themselves as a collective entity (an “us” rather than an aggregation of “I’s”). Sidebar 44.1 outlines some techniques that can help in this regard. While techniques for establishing shared social identity should be prioritized in the early stages of an intervention, when the group is being established, it is also important for facilitators to monitor the group’s progression and to be ready to intervene – for example, to manage potential conflict between members that can influence social identity processes (as observed by Nackers et al., 2015). To this end, facilitator training sessions can be used to develop strategies for intervention – for example, helping to identify

Social Identity Interventions

655

Sidebar 44.1 Building social identity interventions

Table 44.2 provides some examples of techniques that facilitators can use to understand and shape social identity formation in new intervention groups, corresponding to practitioner points 1 and 2 (identity reflection and identity representation). The examples have been informed by the authors’ recent trial research (e.g., Haslam et al., 2016; Tarrant et al., 2018) and group intervention mapping work (Borek et al., 2019), as well as the wider social identity and group processes literature (e.g., Budman et al., 1987; Haslam et al., 2018). Note that the first technique in Table 44.2 (understanding social identity “fit”) is employed prior to group intervention in order to support early group integration, while the remaining techniques are employed once the intervention has started. Table 44.2 Techniques to promote shared social identity Social Identity Principle

Technique

Practice Example

Understanding social Mapping similarities between Prior to group intervention, reflect identity “fit” group members on individual member backgrounds, experiences, and preferences with a view to integrating members into the group Increasing social Using inclusive language “Welcome to our group”; “We’re identity salience doing really well this week” Invite group members to share Encouraging interaction Increasing member their experiences (e.g., an between members; profamiliarity and experience of previous weightmoting cooperation and trust loss success); set exercises for trust, interest, and group-based problem-solving involvement Promoting social Encourage group members to Use activities for group members to identify commonalities with connectedness see themselves in terms of one another (e.g., in aspirations) shared identity in ways that across the group, starting with help them to recognize pair-based activities similarities Promote member reflections on Increasing self-cate- Promoting member interest personal achievements that and involvement in the gorization relate to the group’s function group; emphasizing the “readiness” (e.g., increase in physical activity relevance of the group in since the last group session); helping members achieve shared decision-making (e.g., behavior change goals rules of engagement)

656

MARK TARRANT, CATHERINE HASLAM, MARY CARTER, RAFF CALITRI, AND S. ALEXANDER HASLAM

signs of disengaged or disruptive group members and how they can be managed.

awareness of others’ performance as it pertains to group norms (the norm enactment and influence hypotheses).

44.5.2 Delivery Planning By attending to social identity processes, group facilitators pave the way to deliver the specific intervention content that targets change behaviors. Illustrating practitioner point 3, this stage of the intervention is concerned with helping group members realize the group’s values in a way that transforms behavior change into “something for us” rather than “something done to us.” Practitioners need to ensure that change techniques are suitable for the planned group-based mode of delivery (see Dombrowski, O’Carroll, & Williams, 2016) so that they effectively enact group members’ social identity. Table 44.3 offers some suggestions for doing this – thereby mapping Michie et al.’s (2013) behavior change technique taxonomy version 1 to the group resources articulated in the social identity model of behavior change. Guidelines exist to help ensure precision in the reporting of group-based intervention content (e.g., Borek et al., 2015). The examples presented in Table 44.3 are by no means exhaustive. Moreover, techniques applied at the individual level (e.g., prompting individual members to give self-rewards; technique 10.7 in Michie et al., 2013) also remain appropriate in group interventions, not least because the behaviors that are the target of change are commonly performed outside the intervention group setting (e.g., keeping a dietary diary). However, facilitators can draw on individual performances and experiences in order to support the group resources articulated in the model. For instance, collectively celebrating an individual member’s achievements during a group session can bring benefits for the wider group by reinforcing the meaning of members’ social identity and their group commitment (in line with the meaning hypothesis), helping build their collective efficacy beliefs through exposure to others’ experiences (the agency hypothesis), and raising their

44.5.3 Evaluation of Effectiveness Beyond the assessment of primary outcomes (e.g., whether the intervention leads to behavior change), the evaluation of group-based interventions should consider the group processes by which such outcomes are achieved (Chapter 22, this volume). In line with the identification hypothesis, social identification is the key mechanism that accounts for (i.e., mediates) the effect of the group intervention on behavior change and thus needs to be measured – ideally, at several time points across the intervention to capture the construct’s progression. Assessing social identification early in an intervention can also help identify those recipients who are not engaged with the group and are therefore at risk of dropping out of the program. Information on appropriate measures of social identification is presented in Appendix 44.2. While social identification is the principal hypothesized mechanism, the six group resources specified in the social identity model of behavior change that flow from this can themselves be assessed. A variety of scales have been developed and validated with this in mind and interested readers are advised to consult the appendix to Haslam et al. (2018) for details.

44.6 Implementation Example from the Singing for People with Aphasia Trial Appendix 44.3 describes a clinical trial of a groupbased singing intervention for people with aphasia after a stroke that followed the step-by-step approach described in Section 44.5 (Tarrant et al., 2018). Aphasia is a language disorder that leads to deficits in gesture, reading/writing, expression, and comprehension of speech. Beyond this, though, people with aphasia often also report poor psychosocial health, including reduced social participation,

Table 44.3 Techniques from Michie et al.’s (2013) behavior change technique taxonomy version 1 interpreted through a group-based behavior change lens Social Identity Hypothesis Targeted

Technique Number/Label

Original Example

Suggested Adaptations and Examples for Group Delivery

1.1. Goal setting (behavior)

Agree a daily walking goal with the person and reach agreement about the goal Set a weight-loss goal as an outcome of changed eating patterns Arrange for a housemate to encourage continuation with the behavior change program

Encourage the group to set collective walking and weight-loss Norm enactment; Influence; Meaning goals that depend on individual members’ performance (e.g., meeting individual exercise targets); emphasize shared purpose and brainstorming to determine common goals (e.g., around daily walking)

1.3. Goal setting (outcome)

Build meaningful bonds between group members, using techniques in Sidebar 44.1, which can be used to encourage the group as a whole to support each other’s participation in the program Group celebrations of individual member successes (emotional support); group discussions to identify behavioral “solutions” for members expressing difficulty in meeting change targets (instrumental support) No adaptation required, although practitioners should use 6.3. Information about others’ Tell the staff at the hospital inclusive language (e.g., “us,” “we”) in order that the approval ward that staff at all other technique becomes a collective strategy for behavior wards approve of washing change their hands according to the guidelines Encourage the group to see that, together, they can success15.1. Verbal persuasion about Tell the person that they can fully increase their physical activity capability successfully increase their physical activity, despite their recent heart attack Encourage members to share experiences of successful behavior change with the group to enhance capability beliefs; discuss strategies for achieving goals; “trying out” strategies together (e.g., group stretches/warm-ups in preparation for physical activity) 3.1. Social support (unspecified)

Note. Corresponds to practitioner point 3.

Connectedness; Support

Agency

Norm enactment; Influence; Connectedness

Agency

Support; Agency

658

MARK TARRANT, CATHERINE HASLAM, MARY CARTER, RAFF CALITRI, AND S. ALEXANDER HASLAM

social isolation, and lower levels of overall wellbeing (Cruice, Worral, & Hickson, 2006; Dalemans et al., 2010; Hilari et al., 2010). Nevertheless, many people with aphasia retain the ability to sing after their stroke and group singing is, by definition, social, demanding coordination between participants (Pearce, Launay, & Dunbar, 2015; Schlaug et al., 2010). Accordingly, it was reasoned that the intervention of bringing people with aphasia together as members of a singing group might improve aspects of their life that are often experienced as significant barriers to psychosocial functioning. The intervention was a ten-week program and it was assessed in a two-group, assessor-blinded, randomized controlled external pilot trial involving a parallel mixed-methods process evaluation and economic evaluation (Tarrant et al., 2018; Tarrant et al., 2019). The trial was not powered to evaluate intervention effectiveness but the analysis indicated that the intervention was both feasible and acceptable to study participants. Nevertheless, it is noteworthy here that the process evaluation highlighted the clear potential for the intervention to provide the basis for the development of shared social identity among recipients.

44.7 Summary and Conclusion The social identity model of behavior change, and the social identity approach more generally, offers a set of evidence-based practical strategies pertinent to the development, delivery, and evaluation of group-based behavior change interventions. Central to the model is the observation that intervention groups can become a positive resource for behavior change to the extent that facilitators engage in social identity management in ways that help group members to develop a sense of shared social identity. By attending to key social identity processes, facilitators can ensure that the group provides a platform for delivery of intervention content targeted at specific change behaviors. The step-by-step guide

presented in this chapter outlines key actions that facilitators can take to shape group members’ social identity and manage the group processes that flow from this and underpin change.

References Bandura, A. (1994). Self-efficacy. In V. S. Ramachaudran (Ed.), Encyclopedia of Human Behavior, Vol. 4 (pp. 71–81). New York: Academic Press. Borek, A. J., Abraham, C., Greaves, C. J. et al. (2019). Identifying change processes in group-based health behaviour-change interventions: Development of the mechanisms of action in group-based interventions (MAGI) framework. Health Psychology Review, 13, 227–247. https:// doi.org/10.1080/17437199.2019.1625282 Borek, A. J., Abraham, C., Greaves, C. J., & Tarrant, M. (2018). Group-based diet and physical activity weight-loss interventions: A systematic review and meta-analysis of randomised controlled trials. Applied Psychology: Health and Wellbeing, 10, 62–86. https://doi.org/10.1111/aphw.12121 Borek, A. J., Abraham, C., Smith, J. R., Greaves, C. J., & Tarrant, M. (2015). A checklist to improve reporting of group-based behaviour-change interventions. BMC Public Health, 15, 963. https://doi.org/10.1186/s12889-015-2300-6 Budman, S. H., Demby, A., Feldstein, M. et al. (1987). Preliminary findings on a new instrument to measure cohesion in group psychotherapy. International Journal of Group Psychotherapy, 37, 75–94. https://doi.org/10.1080/ 00207284.1987.11491042 Cruice, M., Worral, L., & Hickson, L. (2006). Perspectives of quality of life by people with aphasia and their family: Suggestions for successful living. Topics in Stroke Rehabilitation, 13, 14–24. https://doi.org/10.1310/4JW5-7VG8G6X3-1QVJ Cruwys, T., Steffens, N. K., Haslam, S. A. et al. (2019). Predictors of social identification in group therapy. Psychotherapy Research, 21, 303–335. https:// doi.org/10.1177/1088868316656701 Dalemans, R. J., De Witte, L. P., Beurskens, A. J., Van Den Heuvel, W. J., & Wade, D. T. (2010). An

Social Identity Interventions

investigation into the social participation of stroke survivors with aphasia. Disability and Rehabilitation, 32, 1678–85. https://doi.org/ 10.3109/09638281003649938 Dombrowski, S. U., O’Carroll, R. E., & Williams, B. (2016). Form of delivery as a key “active ingredient” in behaviour change interventions. British Journal of Health Psychology, 21, 733– 740. https://doi.org/10.1111/bjhp. 12203s Frings, D., Collins, M., Long, G., Pinto, I. R., & Albery, I. P. (2016). A test of the social identity model of cessation maintenance: The content and role of social control. Addictive Behaviors Reports, 3, 77– 85. https://doi.org/10.1016/j.abrep.2016.02.003 Greenaway, K. H., Cruwys, T., Haslam, S. A., & Jetten, J. (2015). Social identities promote well-being because they satisfy global psychological needs. European Journal of Social Psychology, 46, 294– 307. https://doi.org/10.1002/ejsp.2169 Haslam, C., Cruwys, T., Haslam, S. A., Dingle, G., Chang, M. X. (2016). Groups 4 Health: Evidence that a social-identity intervention that builds and strengthens social group membership improves health. Journal of Affective Disorders, 194, 188– 195. https://doi.org/10.1016/j.jad.2016.01.010 Haslam, C., Jetten, J., Cruwys, T., Dingle, G., & Haslam, S. A. (2018). The New psychology of Health: Unlocking the Social Cure. London: Routledge. https://doi.org/0.4324/9781315648569 Haslam, S. A., Reicher, S. D., & Levine, M. (2012). When other people are heaven, when other people are hell: How social identity determines the nature and impact of social support. In J. Jetten, C. Haslam, & S. A. Haslam (Eds.), The Social Cure: Identity, Health and Well-Being (pp. 157–174). New York: Psychology Press. Haslam, S. A., Reicher, S. D., & Platow, M. J. (2011). The New Psychology of Leadership: Identity, Influence and Power. New York: Psychology Press. Hilari K., Northcott S., Roy P. et al. (2010). Psychological distress after stroke and aphasia: The first six months. Clinical Rehabilitation, 24, 181– 190. https://doi.org/10.1177/0269215509346090 Kearns, M., Muldoon, O. T., Msetfi, R. M., & Surgenor, P. W. (2015). Understanding help-seeking amongst university students: The role of group identity, stigma, and exposure to suicide and help-

659

seeking. Frontiers in Psychology, 6, 1462. https:// doi.org/10.3389/fpsyg.2015.01462 Khan, S. S., Tarrant, M., Kos, K., Daly, M., Gimbuta, C., & Farrow, C. V. (2019). Making connections: Social identification with new treatment groups for lifestyle management of severe obesity. Unpublished manuscript, University of Exeter. Kok, G., Gottlieb, N.H., Peters, G. J. et al. (2016). A taxonomy of behaviour change methods: An intervention mapping approach. Health Psychology Review, 10, 297–312. https://doi.org/ 10.1080/17437199.2015.1077155 Michie, S., Richardson, M., Johnston, M. et al. (2013). The behavior change technique taxonomy (v1) of 93 hierarchically clustered techniques: Building an international consensus for the reporting of behavior change interventions. Annals of Behavioral Medicine, 46, 81–95. https://doi.org/ 10.1007/s12160-013-9486-6 Michie, S., Van Stralen, M. M., & West, R. (2011). The behaviour change wheel: A new method for characterising and designing behaviour change interventions. Implementation Science, 6. https:// doi.org/10.1186/1748-5908-6-42 Nackers, L. M., Dubyak, P. J., Lu, X., Anton, S. D., Dutton, G. R., & Peri, M. G. (2015). Group dynamics are associated with weight loss in the behavioral treatment of obesity. Obesity, 23, 1563–1569. https://doi.org/10.1002/oby.21148 Oakes P. J., Haslam, S. A. & Turner, J. C. (1994). Stereotyping and Social Reality. Oxford: Blackwell. Oyserman, D., Fryberg, S. A., & Yoder, N. (2007). Identity-based motivation and health. Journal of Personality and Social Psychology, 93, 1011–1027. https://doi.org/10.1037/0022-3514.93.6.1011 Pearce, E., Launay, J., & Dunbar, R. I. M. (2015). The ice-breaker effect: Singing mediates fast social bonding. Royal Society Open Science, 2. https:// doi.org/10.1098/rsos.150221 Reicher, S. D., Haslam, S. A., & Hopkins, N. (2005). Social identity and the dynamics of leadership: Leaders and followers as collaborative agents in the transformation of social reality. The Leadership Quarterly, 16, 547–568. https://doi .org/0.1016/j.leaqua.2005.06.007 Sani, F. (2005). When subgroups secede: Extending and refining the social psychological model of

660

MARK TARRANT, CATHERINE HASLAM, MARY CARTER, RAFF CALITRI, AND S. ALEXANDER HASLAM

schism in groups. Personality and Social Psychology Bulletin, 31, 1074–1086. https://doi .org/10.1177/0146167204274092 Schlaug, G., Norton, A., Marchina, S., Zipse, L., & Wan, C. Y. (2010). From singing to speaking: Facilitating recovery from nonfluent aphasia. Future Neurology, 5, 657–665. https://doi.org/ 10.2217/fnl.10.44 Smith, I. H., & Woodworth, W. P. (2012). Developing social entrepreneurs and social innovators: A social identity and self-efficacy approach. Academy of Management Learning and Education, 11, 390–407. https://doi.org/10.5465/ amle.2011.0016 Steffens, N. K., Haslam, S. A., Reicher, S. D. et al. (2014). Leadership as social identity management: Introducing the Identity Leadership Inventory (ILI) to assess and validate a four-dimensional model. The Leadership Quarterly, 25, 1001–1024. https://doi.org/10.1016/j.leaqua.2014.05.002 Steffens, N. K., La Rue, C. J., Haslam, C. et al. (2020). Social identification-building interventions to improve health: A systematic review and meta-analysis. Health Psychology Review. https://doi.org/10.1080/17437199.2019.1669481 Tajfel, H., & Turner, J. C. (1986). The social identity theory of intergroup behavior. In S. Worchel & W. G. Austin (Eds.), Psychology of

Intergroup Relations (pp. 7–24). Chicago: Nelson-Hall. Tarrant, M., Carter, M., Dean, S. G. et al. (2018). Singing for people with aphasia (SPA): A protocol for a pilot randomised controlled trial of a group singing intervention to improve well-being. BMJ Open, 8, e025167. https://doi.org/10.1136/bmjo pen-2018-025167 Tarrant, M., Carter, M., Dean, S. G. et al. (2019). Singing for people with aphasia (SPA): Results of a pilot randomised controlled trial of a group singing intervention to improve wellbeing. Unpublished manuscript, University of Exeter. Tarrant, M., Khan, S. S., Farrow, C. V., Shah, P., Daly, M., & Kos, K. (2017). Patient experiences of a bariatric group programme for managing obesity: A qualitative interview study. British Journal of Health Psychology, 22, 77–93. https://doi.org/ 10.1111/bjhp.12218 Turner, J. C. (1982). Towards a cognitive redefinition of the social group. In H. Tajfel (Ed.), Social Identity and Intergroup Relations ( pp. 15–40). Cambridge: Cambridge University Press. Turner, J. C., Hogg, M. A., Oakes, P. J., Reicher, S. D., & Wetherell, M. S. (1987). Rediscovering the Social Group: A Self-Categorization Theory. Oxford: Blackwell.

45 Motivational Interviewing Interventions Anne H. Berman, Maria Beckman, and Helena Lindqvist

Practical Summary Motivational interviewing (MI) is used to evoke and reinforce individuals’ inner motivation to change their behavior toward better health. MI can be used by practitioners from a variety of backgrounds and in many settings including health care, criminal justice, occupational, educational, and others. To attain and maintain proficiency, practitioners need to undergo brief training and improve their skills through steady practice accompanied by coaching and fidelity measures. The strongest evidence for MI effectiveness has been shown in adults with addictive behaviors including alcohol, substance use, and smoking, as well as for increasing physical activity among people with chronic illnesses. Studies on many other behaviors and among other populations, including adolescents, have also shown positive results. Implementing MI in organizations requires systematic, persistent attention to maintaining adequate levels of adherence and competence among practitioners. For the motivated learner and practitioner of MI, web-based instruction and supporting materials are available free of charge.

45.1 Introduction Motivational interviewing (MI) is a method of eliciting individuals’ inner motivation to change their behavior, generally with a focus on better health. MI was originally developed within the addiction field, to reduce excessive alcohol consumption (Miller, 1983), and has subsequently been applied to a variety of problematic behaviors (Frost et al., 2018). The method has been widely disseminated via books in three editions, with the latest published in 2013 (Miller & Rollnick, 2013), and via training events by the Motivational Interviewing Network of Trainers (MINT, 2019) for practitioners and trainers. Recently, online applications for clinical and educational purposes, via eHealth interventions (Shingleton & Palfai,

2016), as well as web-based instructional platforms (e.g., Berman et al., 2017), have also been used to disseminate the method. MI has been implemented mainly in primary care, psychiatry, addiction care, hospital emergency care, and criminal justice settings, but also in workplace, education, and environmental contexts. The method is also used to target behavior change in multiple populations, including adults in different age groups, adolescents, and children. Since the turn of the millennium, research has begun to focus more specifically on the theoretical mechanisms behind the effectiveness of MI, complementing empirical studies (Magill et al., 2018; Miller & Rose, 2009; Pace et al., 2017). https://doi.org/10.1017/9781108677318.045

662

ANNE H. BERMAN, MARIA BECKMAN, AND HELENA LINDQVIST

Laboratories that code MI interventions, to evaluate the quality of practitioners’ MI skills and to analyze the practitioner-client verbal interaction, are seminal to promoting the maintenance of MI fidelity in clinical and research contexts. This chapter provides an evidence-based overview of applications of MI in behavior change based on the research evidence and a brief guide on how MI interventions have been designed, implemented, and evaluated.

45.2 Definitions An understanding of MI builds on an overarching definition of its purpose and description of its central concepts. MI is fundamentally “a collaborative conversation style for strengthening a person’s own motivation and commitment to

change” (Miller & Rollnick, 2013, p. 12). To establish this style of conversation, MI builds on central concepts of practice that are both relational and technical. The principal relational concepts are empathy and MI “spirit.” Empathy is the attempt to enter into the client’s perspective, to “walk in the shoes” of the client to better understand what they are feeling and thinking. A practitioner with high empathy skills displays curiosity about the client and interest in respectfully exploring the client’s experience to better understand their perspective. MI “spirit” is a general conversational approach that embodies partnership, acceptance, compassion, and evocation (see Sidebar 45.1). The technical aspects of MI consist of conversational techniques for evoking and reinforcing client change talk (Miller &

Sidebar 45.1 The motivational interviewing (MI) “spirit”

The relational, or “spirit,” component of MI emerged from the client-centered approach developed by Rogers (1956), with particular emphasis on accurate empathy, respect for client autonomy, and egalitarian collaboration in the relationship (Miller & Moyers, 2017). According to Miller and Rollnick (2013), MI spirit includes four interrelated key components: partnership, acceptance, compassion, and evocation (Table 45.1). Table 45.1 The four interrelated components of motivational interviewing (MI) “spirit” (after Miller & Rollnick, 2013) Component

Description

Partnership

Partnership in MI means working in collaboration with the client, acknowledging that practitioner and client contribute inputs of equal value in a positive atmosphere conducive to change. The practitioner focuses on supporting and exploring the client’s situation and ideas, avoiding attempts at persuasion. Acceptance in MI covers four aspects: (1) absolute worth, reflecting the inherent potential and value of every human being; (2) accurate empathy, being the practitioner’s active and genuine interest in attempting to understand the client’s experience and internal perspective; (3) autonomy support, recognizing that the true power for change rests within the client and acknowledging the individual’s own freedom to make choices; and (4) affirmation,

Acceptance

Continued

Motivational Interviewing Interventions

663

Table 45.1 (cont.) Component

Compassion Evocation

Description about expressing and intentionally acknowledging and reinforcing the client’s efforts, strengths, intentions, or worth. Compassion in MI emphasizes the practitioner’s deliberate commitment to promote the client’s genuine needs and well-being. Evocation in MI means that the practitioner elicits and emphasizes the client’s existing inner resources and motivation for change rather than imparting information or opinions.

Rose, 2009). The skillful practitioner weaves these techniques into the conversation in parallel with continual attunement to the relational quality of the interaction. In the current chapter, the definitions of relational and technical aspects of MI as formulated by Miller and Rollnick (2013) are adopted. However, it is worth noting that there has been some lack of clarity on the dividing line between relational and technical aspects of MI spirit, particularly from a research perspective, where different coding schemes have been used to define MI spirit (cf. McCambridge et al., 2011). Also, it is important to note that MI has been used in adapted versions, widely known as adapted motivational interviewing, primarily centered around brief feedback on the need to change health- or illness-related behaviors and incorporating the principles of MI as needed (Burke, Arkowitz, & Menchola, 2003). The most widely used adaptation of MI, motivational enhancement therapy, is a manual-based, four-session intervention including assessment feedback developed as part of Project MATCH (Babor & Del Boca, 2003). In addition, within the framework of brief interventions, which are quick advice-giving procedures within primary health care as well as hospital and student health care settings, MI has been included to personalize the intervention and then termed brief MI interventions (Gaume et al., 2014).

A considerable body of research, briefly summarized in Section 45.3, has evaluated MI and its adaptations and shows that the evidence varies by specific behavioral focus for MI, the target population, and the setting for MI delivery.

45.3 Theory and Mechanisms of Change The theoretical underpinnings of MI include Rogers’s (1956) nonconfrontational, accepting and supportive client-centered approach to psychotherapy, paired with Festinger’s (1962) cognitive dissonance theory and Bem’s (1972) self-perception theory (Miller & Rose, 2009). Cognitive dissonance can arise from hearing oneself speak of behavior change that has not been acted on. Hearing oneself utter thoughts, intentions, and plans to change a behavior can reinforce the need to change the behavior in order to resolve the dissonance between intentions and actual behavior. Hearing oneself speak about behavior change can also enhance a person’s self-perception that they are actually intending to change the behavior in question. In an MI session, uttering thoughts, intentions, and plans to change a behavior is thus posited to be associated with later behavior change (Miller, 1983). Decades of research and clinical practice on MI have led to an explication of the technical

664

ANNE H. BERMAN, MARIA BECKMAN, AND HELENA LINDQVIST

and relational aspects that was summarized in a proposed theoretical model for causal change (Miller & Rose, 2009). According to this model, MI training should lead to practitioner proficiency in relational skills of empathy and MI spirit as well as technical skills in MI-consistent techniques. Use of these dual-track skills is postulated to cause increased frequency of client utterances concerning adaptive behavior change, termed change talk, and reduced frequency of client utterances about sustaining problematic behaviors, termed sustain talk. These client verbal behaviors could then, in turn, lead to expression of commitment to behavior change and then to actual change of the behavior in focus (see Sidebar 45.2). In practice, client change talk functions as a signal of client motivation in the MI conversation toward adaptive behavior change. Client sustain

talk, in contrast, is a signal of stagnation. Change talk is elicited and reinforced by MI techniques such as open questions, affirmations, reflections, and summaries (OARS; see Appendix 45.1, supplemental materials). The reflection techniques, categorized as simple or complex, is at the heart of change talk evocation. The practitioner’s simple reflections repeat or reformulate client utterances and indicate that the practitioner has listened to the client as well as allowing the client to hear their own words uttered by another. In complex reflections, the practitioner juxtaposes their own perceptions of the client’s thoughts, feelings and behaviors with the client’s actual utterances, in order to deepen mutual understanding of the multiple elements influencing variations in their motivation and move the conversation toward commitment and plans for behavior change (see Appendix 45.1, supplemental materials).

Sidebar 45.2 Empirical evidence for the theoretical conceptualization of motivational interviewing (MI)

The increasing volume of research on MI has led to theoretical models of MI in which relational and technical skills are proposed to lead to client verbal behaviors that move the client toward actual behavior change (Miller & Rose, 2009). Two metaanalytic reviews (Magill et al., 2018; Pace et al., 2017), and a secondary analysis of a randomized controlled trial (RCT; Villarosa-Hurlocker et al., 2019), have provided important evidence in support of the model. The two meta-analyses showed support for the relationships between practitioner MI behaviors (empathy, MI Spirit, and MI-consistent behaviors) and increased client change talk, as well as increased client sustain talk (Magill et al., 2018; Pace et al., 2017), while the RCT showed that the relationship between practitioner MI relational behaviors and the percentage of client change talk was mediated by practitioner reflections of change and sustain talk, that is, technical skills (Villarosa-Hurlocker et al., 2019). In terms of client behavioral outcomes, both meta-analyses showed that increased client sustain talk was negatively related to improved client behavioral outcomes (i.e., reducing a risky behavior or increasing a healthy behavior) but, in conflict with the conceptual model (Miller & Rose, 2009), no significant association was found between increased client change talk and behavioral outcomes. (for an overview of theory and findings, see Figure 45.1).

Motivational Interviewing Interventions

665

Client change talk

+ Relational skills: Empathy, MI spirit

+

+ MI training with fidelity measures & coaching

+

+ Client sustain talk

–

Healthier client behavioral outcome

+ + Technical skills: MI-consistent

+

+ Percent client change talk

Figure 45.1 Theoretical conceptualization of how motivational interviewing (MI) works based on Miller and Rose (2009), with approximate summaries of effect sizes based on current empirical evidence (Magill et al., 2018; Pace et al., 2017; Schwalbe et al., 2014). Solid bold lines indicate a medium effect size, the dashed bold line represents a small-tomedium effect size, solid lines represent a small effect size, and the gray dashed line represents a null effect. All effects are positive in sign (+), with the exception of the effect of client sustain talk on healthier client behavioral outcome, which is negative in sign (−).

With empathy and MI spirit as a platform for the client-practitioner conversation and MI techniques as the “nuts and bolts” of their interaction, the conversational progression toward behavior change is formulated in four dynamic processes that simultaneously structure and permeate MI conversations. The first, termed the engagement process, builds a relational foundation for open, trustful practitionerclient communication. The focusing process follows with the aim of identifying a specific behavior as the target for the MI conversation. Third is the evoking process, where the practitioner’s explicit focus is on evoking the client’s motivation to change the target behavior in a healthier direction, and going into detail about the desire, ability, reasons, and need for change, summarized in the DARN acronym. The final, fourth process concerns planning for the behavior change, where the practitioner reinforces signs of actual commitment, action, and taking steps, summarized in the CAT acronym (for examples of DARN-CAT, see Table 45.3). In principle, the

processes occur consecutively. However, in practice, MI practitioners and their clients sometimes cycle between the four processes as necessary to promote behavior change. This can, for example, mean that engagement will require repeated attention to maintain the relational foundation for the MI conversation, and that planning may need to be revised as the target behavior is further defined or shifts to another behavior.

45.4 Evidence Base The existing research literature for MI is extensive, with numerous context- and behavior-specific meta-analyses or systematic reviews published. A systematic review of reviews and meta-analyses of MI effectiveness for a variety of behaviors in health and social care settings among adults (Frost et al., 2018) revealed small positive effects of MI in reducing or eliminating unhealthy behaviors over six months: binge

666

ANNE H. BERMAN, MARIA BECKMAN, AND HELENA LINDQVIST

drinking, frequency and quantity of alcohol consumption, problematic substance use, and smoking. In terms of promoting healthy behaviors, MI was found to be effective in increasing physical activity among persons with chronic illnesses, with small effect sizes. The quality of this evidence was found to be moderate. MI was also found to be effective in reducing problematic gambling behavior and promoting weight loss among obese or overweight adults, although the evidence was of lower quality (Frost et al., 2018). Among adolescents, combined systematic reviews and meta-analyses have been published on MI for reducing or eliminating specific behaviors such as alcohol consumption (Kohler & Hofmann, 2015; Tanner-Smith & Lipsey, 2015), illicit drug use (Li et al., 2016), smoking and obesity (Heckman, Egleston, & Hofmann, 2010; Vallabhan et al., 2018), and promoting oral health (Werner et al., 2016). A more recent scoping review has been published regarding sexual risk behaviors (Bahner & Stenqvist, 2019). Finally, a meta-analysis of MI for a variety of adolescent health behaviors excluding substance use has been conducted (Cushing et al., 2014). The results of these studies have been inconclusive. A review of six studies of MI for reducing alcohol consumption in the emergency room (ER) setting suggested that MI was at least as effective as other brief interventions in the ER, and some findings suggested MI could be even more effective than existing interventions (Kohler & Hofmann, 2015). Another review of twenty-four studies of brief interventions for reducing alcohol consumption and alcohol-related problems among adolescents showed a significant effect that persisted up to one year compared to control interventions, with greater effects for MI. Effects were consistent over diverse settings and particularly effective components included decisional balance (juxtaposition of pros and cons of change) and goal setting (Tanner-Smith & Lipsey, 2015). Regarding MI for illicit drug use among adolescents, a review of ten studies showed no effects on drug use behaviors;

however, changes in attitudes toward drug use were found, which could be translated into intentions to change behaviors (Li et al., 2016). Similar results for adolescents have been found in metaanalyses regarding MI for smoking (Heckman et al., 2010) but not for weight management and obesity (Vallabhan et al., 2018). A meta-analytic review of RCTs investigating MI among adults and adolescents with poor oral health found no RCT investigating MI for adolescents despite the urgent need for research on improving oral health in this group (Werner et al., 2016). A scoping review concerning MI for reducing sexual risk behaviors among adolescents identified twenty-nine unique studies with a wide variety of designs, conceptualizations of MI, and specific risk behaviors, making it difficult to generalize regarding outcomes but indicating the need to increase research efforts (Bahner & Stenqvist, 2019). In addition, small but significant effects of MI on a variety of adolescent health behaviors (e.g., diet, physical activity, and sexual risk behaviors) were found in a metaanalysis of fifteen studies comparing MI to control conditions (Cushing et al., 2014). Although the scientific evidence base for MI is growing, future research of higher quality is needed for most behaviors and settings. Also, more research is needed on the effectiveness of MI adaptations. Some adaptations of the method will be necessary depending on the setting, but the various differences in the adaptation of MI and the wide and rapid dissemination of this complex method contribute to the risk of MI being watered down into a simpler copy of the method, raising the issue of treatment fidelity evaluation (Miller & Rollnick, 2009, 2014).

45.5 Congruent Behavior Change Techniques from Taxonomies Sorting MI behavior change techniques according to taxonomies has three purposes: first, to identify techniques that can be tabulated in order to assess

Motivational Interviewing Interventions

intervention fidelity; second, to facilitate practitioner understanding of their own in-session behaviors and how they relate to the process of client behavior change; and, third, such categorization can facilitate synthesis of findings across behavior change techniques. Both the first and second aims have been addressed by the plethora of coding instruments developed within the MI research community, where the most widely used coding schemes are the Motivational Interviewing Treatment Integrity Code 4.0 (MITI 4.0; Moyers et al., 2016) and the Motivational Interviewing Skill Code 2.5 (MISC 2.5; Houck et al., 2010) developed to map practitioner-client utterances and transitions between the two (for an overview of coding instruments, see Appendix 45.2, supplemental materials). Addressing the third purpose, an expert consensus analysis of MI techniques from a taxonomical perspective, concerning behavior change techniques, has also shown that MI techniques to promote behavior change can be categorized into relational or contentbased categories (Hardcastle et al., 2017). The consensus analysis involved categorizing MI techniques according to the conversational processes of engaging, focusing, evoking, and planning. Of the thirty-eight MI techniques identified, twenty-two were termed “unique” and were found in higher proportions as relational components in the engaging (75 percent of four techniques) and focusing (75 percent of four techniques) processes, in comparison to the evoking (52 percent of twenty-one techniques) and planning (0 percent of seven techniques) processes (Hardcastle et al., 2017). The remaining sixteen techniques were found to coincide with techniques described in Michie et al.’s (2013) behavior change technique taxonomy version 1 (BCTTv1). Selected MI techniques that were aligned with BCTTv1 techniques in the consensus analysis are shown in Table 45.2 and compared with categorization in the MITI coding scheme.

667

45.6 Step-by-Step Guide This section provides a step-by-step guide to delivering and evaluating MI-based interventions for behavior change. The section outlines the typical content, means of delivery, audiences and behaviors, enabling and inhibiting factors, training and skills required, intervention intensiveness needed, and evaluation of fidelity and effectiveness. It is particularly important to note that the effective delivery of MI interventions is dependent on adequate training and experience in MI techniques, as MI builds on a skill set that is apparently simple but is quite difficult to master and requires considerable practitioner flexibility to appropriately adapt to client responses (see Sections 45.6.3 and 45.6.4).

45.6.1 Typical Means of Delivery MI is a flexible intervention in terms of time point for delivery within treatment and delivery mode format and this may be part of the explanation for its rapid international diffusion and dissemination. In terms of delivery time point, MI can be effectively used as a motivational prelude to other interventions, as a “stand-alone” intervention, or in combination with other interventions such as cognitive behavioral therapy (Lundahl et al., 2010; Miller & Rollnick, 2013). Some meta-analyses have suggested that MI may be well suited as a prelude to treatment (Burke et al., 2003; Hettema, Steele, & Miller, 2005; Lundahl et al., 2010), but a systematic review of reviews concluded that the evidence for significantly better adult engagement with mental health interventions, following MI, is still of low quality (Frost et al., 2018). Regarding delivery format, MI has most commonly been delivered as an individual face-to-face intervention (see examples of typical MI conversations in Appendix 45.3, supplemental materials). However, MI has also been provided via telephone (e.g., Gates et al., 2012), in text formats (e.g., Mason et al., 2015), as components in computer-

Table 45.2 Examples of motivational interviewing (MI) techniques matched with techniques from behavior change taxonomies with related techniques and descriptions and MI coding schemes

MI Technique

Description as Practiced in MI

Typical Quote

MITI Coding Scheme Michie et al. BCTTv1 (Michie et al., 2013); (Moyers et al., 2016) Matching as in Hardcastle et al. (2017)

Partial overlap with 15.1: Verbal persua“In your condition and with all Affirmation (added sion about capability this snow, it wasn’t easy for you value for Empathy) to get here. I really appreciate that you did. “ Is it OK if we talk about alco- Seeking Collaboration, Partial overlap with 5.1: Information about Elicit-ProvideP asks for permission to speak health consequences Questions, Giving hol?” (P) Elicit about a behavior, then asks C Information, what they know about it already. “ OK” (C) Seeking Then P delivers information that “ What do you know about alcoCollaboration hol effects on the body?” (P) is adapted to what C already (added value for knows. P then asks C what this “Not much” (C) P gives information about alcohol Partnership and might mean for them Cultivating Change and the body. Then: “How does this sound to you?” (P) Talk) Related to Valued self-identity 13.4: Advise Identify strengths P helps the client recognize their “ When I have a deadline, I really Affirmation (added strengths by reflecting strength- make sure to keep it, whatever the value for Cultivating the person to write or complete rating Change Talk) scales about a … personal strength as a based utterances circumstances” (C); means of affirming the person’s identity “You are good at carrying out as part of a behavior change strategy what you truly decide to do” (P) (includes self-affirmation) Develop a change P assists C in developing a specific “How would you like to start?” Question (added value Action Planning. 1.4: Prompt detailed for Cultivating planning of behavioral performance (C); “What are you ready to plan for change, one that C is plan (CATs) (must include at least one of context, do?” (A); “When and how will Change Talk) ready to accept from a motivaC= frequency, duration and intensity). you take this step?” (T) tional standpoint Commitment Context may be environmental (physiA= cal/social) or internal (physical, emoActivation tional or cognitive) T= Taking steps Affirmation

Practitioner (P) expresses appreciation for client (C) actions or personal attributes

Motivational Interviewing Interventions

based interventions (e.g., Christoff & BoerngenLacerda, 2015), and in group formats (e.g., D’Amico et al., 2015). However, inconsistent results have been reported in reviews on the mode of MI delivery (Frost et al., 2018). There is reasonable evidence that MI can be delivered effectively by telephone (Jiang, Wu, & Gao, 2017), but more research is needed regarding the effectiveness of both group-delivered MI (Lundahl et al., 2010) and different versions of technology-delivered MI (Shingleton & Palfai, 2016).

45.6.2 Target Audiences and Behaviors MI is appropriate for anyone who may need to change a specific behavior but has ambivalent or even reluctant feelings about doing so (see also Chapter 10, this volume). If a person is ready to change, in terms of formulating a change plan and taking concrete steps toward behavior change, the full range of MI components is not needed but the relational skills can still be used to support and reinforce the change process (Miller & Rollnick, 2013). MI should not be used in situations where it is not appropriate to influence and favor resolution of ambivalence in a specific direction – for example, when considering whether to continue or interrupt pregnancy (Miller & Rollnick, 2009). Although it is difficult to draw overall generalizable conclusions, MI can be delivered to clients of different ages, gender, ethnicity, and problem severity (Lundahl & Burke, 2009). MI may not, however, be as helpful for children and preteens, perhaps because a level of abstract reasoning that develops after puberty is needed for MI to be effective (Lundahl et al., 2010). Meta-analytic findings also suggest that MI may be particularly helpful with clients from minority ethnic groups, although it is not entirely clear why; for Native Americans, it might be because MI is more culturally congruent with this group’s communication style (Hettema et al., 2005). Another possible explanation is that minority groups have been exposed to

669

confrontational techniques in the health care system to a larger extent than majority groups and that they have therefore developed a more pronounced aversion to treatment providers. MI, with its clientcentered emphasis as conceptualized in MI spirit, has thus been experienced as a welcome contrast, yielding the effects identified in meta-analysis. The wide applicability of MI has also enabled it to spread from the health behavior field to other fields such as schools and organizations. The evidence for MI in these new settings, targeting new health and social problems, is still sparse but it is worthwhile reviewing its application in these novel contexts. MI has been adapted to facilitate return to work for people with disabilities (e.g., Page & Tchernitskaia, 2014), motivate children toward academic and behavioral improvements in schools (Terry et al., 2013), and enhance sustainable environmental behaviors (Forsberg, Wickström, & Källmén, 2014).

45.6.3 Enabling or Inhibiting Factors Research has shown that not all clients respond to MI and that some MI practitioners are more effective than others in delivering the same treatment, even when trained together and using the same manual (Miller & Rose, 2009). Links between MI delivery processes and client outcomes are not fully understood, although some enabling and inhibiting factors have been proposed in the literature. Research has shown that factors such as the absence of a practitioner’s manual, use of intervention fidelity measures, working with disadvantaged minority populations, and delivery of higher doses of MI treatment yield higher effect sizes of MI on behavior change (Miller & Rollnick, 2014). Although the overall evidence for using MI as a prelude or complement to further treatment has been found to be of low quality (Frost et al., 2018), individual studies have found associations with better outcomes in numerous behavioral contexts such as treatment of severe generalized anxiety disorder (Westra, Constantino, & Antony,

670

ANNE H. BERMAN, MARIA BECKMAN, AND HELENA LINDQVIST

2016), social anxiety (Buckner & Schmidt, 2009), and obsessive-compulsive disorder (Maltby & Tolin, 2005). MI seems to be equally effective for individuals with high and relatively low preexisting levels of distress (Lundahl et al., 2010).

45.6.4 Training and Skills Required MI training should lead to changed counseling practice and improved client outcomes (Miller & Rose, 2009), yet the exact form and amount of training required to achieve such changed counseling practices are unclear (Hall et al., 2015). Previous studies have shown that the common one-time workshop format, without subsequent supervision based on work samples, can be insufficient for producing sustained MI skills (e.g., Schwalbe, Oh, & Zweben, 2014), and that, even with similar formats, variability in practitioners’ MI skills is common after training (Miller & Moyers, 2017). However, neither professional background nor earlier work experience seems to predict practitioners’ MI skills after training, and some researchers have even found that greater work experience may obstruct MI learning (Miller & Moyers, 2017). A meta-analysis of MI training studies found that maintaining training effects requires about three to four subsequent supervision sessions over a period of six months (Schwalbe et al., 2014). Instead of relying on fixed training formats, other researchers have proposed ongoing training until proficiency is attained (Hall et al., 2015; Miller & Moyers, 2017; Miller & Rollnick, 2014). There are various instruments available for measuring practitioners’ skills in MI proficiency (see Section 45.6.6 on evaluation of fidelity and Appendix 45.2, supplemental materials, for a summary of coding instruments).

45.6.5 Intensiveness Numerous empirical studies indicate that small-tolarge effect sizes can be achieved with one to four sessions of MI. In general, a low dose of MI seems

to be effective in eliciting meaningful behavior change. With regard to session length, no exact time span is specified for MI and it can be delivered in single or multiple sessions. Adaptations in the form of brief MI interventions can range between one and sixty minutes spread across one to five sessions for problematic alcohol use (Kaner et al., 2018). In screening and brief intervention models for problematic alcohol use, interventions can be delivered in fewer than five to fifteen minutes by busy clinicians, and such interventions have been shown to be effective in primary health care settings (O’Donnell et al., 2014). An earlier systematic review showed that multiple brief contacts could more effective than single interventions (Jonas et al., 2012). Regarding length of session, a review revealed that sessions longer than a total of sixty minutes did not yield additional effects beyond briefer interventions (Kaner et al., 2018). Regarding multiple contacts with lengthier sessions, four MI-based sessions for problematic alcohol use were evaluated extensively in Project MATCH, where data on client change over twelve months indicated parity between four MI sessions of MI and twelve cognitive behavioral therapy or twelve-step sessions (Babor & Del Boca, 2003). In the area of problem gambling, a single ninety-minute “minimal intervention” consisting of adapted MI and including assessment feedback and practical advice yielded results equivalent to six sessions of either cognitive therapy, behavior therapy, or motivational therapy, in terms of reduced gambling frequency and expenditure (Toneatto, 2016). Overall, a meta-analysis of MI in medical settings showed that the total minutes of delivered MI might potentially moderate the effects of MI (Lundahl et al., 2013). To summarize, MI works as a brief intervention, but the question of dose effectiveness may vary across target behaviors, populations, and settings and merits continued research (Frost et al., 2018). More research investigating the specific components related to effectiveness should be conducted in the future (Kaner et al., 2018).

Motivational Interviewing Interventions

45.6.6 Evaluation of Fidelity Several instruments to assess MI fidelity have been developed. Each instrument differs in purpose as well as specific advantages and disadvantages (Dobber et al., 2015; Madson & Campbell, 2006). An instrument suited for research purposes has the advantage that it aims to assess all hypothesized active components of MI. However, it will likely be difficult to learn how to use and thus costly to implement. An instrument developed to assess MI skills in clinical supervision may assess fewer MI skills and be quite easy to learn how to use but may not give the whole picture of the practitioner’s MI proficiency. Treatment providers and researchers should thus exercise care in selecting the best instrument for specific settings or studies (Moyers et al., 2016). The MITI is the most commonly used instrument to assess practitioners’ MI skills (Moyers et al., 2016). The MITI has been used to assess MI skills in clinical trials and training studies across a range of settings and population groups (e.g., Beckman et al., 2017; Koken et al., 2012; McCambridge et al., 2011) and has also been used to assess MI fidelity in interventions where MI has been combined with

671

other treatments (e.g., Schmidt et al., 2019). Regarding psychometric properties, preliminary evidence has been found for predictive validity and satisfactory inter-rater reliability scores of the latest MITI version 4.2.1 (Moyers et al., 2016; Schmidt et al., 2019). Appendix 45.2 (supplemental materials) provides a summary of most of the current coding instruments.

45.6.7 Evaluation of Effectiveness The major in-session signals of practitioner effectiveness in MI are client change talk and sustain talk, as well as the proportion of change talk to sustain talk (Magill et al., 2018) (see Appendix 45.4, supplemental materials, and Table 45.3). When practitioners evaluate their own MI sessions, it is helpful if they decide beforehand what behaviors to focus on when listening to the recording, such as counting the number of reflections, or the number of change talk and sustain talk utterances, or to note how they responded to expressed change talk in the session (Miller & Rollnick, 2013). This is even more important when MI is implemented in settings where

Table 45.3 Categories of in-session client change talk Client language category

Example of change talk

Example of sustain talk

Desire

“I really want to quit smoking”

Ability Reason Need Commitment

“I would be able to cut down” “I would save a lot of money” “I have to quit now” “I will cut down without using NRT”

Activation

“I am ready to cut down the number of cigarettes I smoke” “I only smoked one cigarette yesterday”

“I would love to have a cigarette right now” “I can’t quit smoking” “Smoking helps me relax” “I need my cigarettes” “I will continue smoking while I enjoy my morning coffee” “I’m not willing to deal with the abstinence” “Last week I began to smoke even more than I used to”

Taking steps

Note. This table summarizes categories of client change talk derived from prior research (Amrhein et al., 2003; Miller & Rollnick, 2013).

672

ANNE H. BERMAN, MARIA BECKMAN, AND HELENA LINDQVIST

adaptations of the method are required (Miller & Rollnick, 2014). It is important to keep in mind, however, that practitioners’ self-assessments of MI skills have not been shown to be particularly reliable (e.g., Miller et al., 2004; Wain et al., 2015). Even MI supervisors seem to overestimate the MI performance of their supervisees (Martino et al., 2009). For more reliable and objective feedback on MI practice, MI sessions should be assessed by objective raters using instruments developed for that purpose (Miller & Rollnick, 2013). Ideally, this type of evaluation will take place within an organizational infrastructure that allows for audio-recording and objective assessment of MI practice using established instruments, as well as discussing individual sessions in designated supervision sessions.

45.6.8 Typical Materials Needed No typical materials are needed to conduct MI, only a skillful practitioner. Indeed, one earlier meta-analysis found that MI effect sizes were twice as large in studies without a specific manual to standardize the intervention, in comparison to studies using manuals (Hettema et al., 2005), a finding that could be explained by the fact that manuals are rarely adapted to the clinical flexibility required by the method. However, helpful worksheets are freely available (MINT, 2019). The personal values card sort, together with an instruction sheet, can help clients clarify their central values and then consider how these values are reflected in their daily lives. The card sorting can be useful in itself, but the value of the process is often in the subsequent dialogue. The change plan worksheet can help the client to sort out the different aspects of the change plan and can also help focus the client’s attention on various critical details. Additionally, it can be helpful to use rulers (see Table A45.1.1 in Appendix 45.1, supplemental materials) in conjunction with the change plan worksheet.

For practicing and learning MI, additional materials are available at the MINT (2019) website: The Ten Strategies for Evoking Change talk sheet provides practitioners with useful evocative questions and exercises that can be used in MI sessions. The MI Training New Trainers manual provides a variety of training ideas, specific exercises, metaphors and activities designed to help practitioners learn the method. A glossary of terms from the third and latest MI book by the method’s developers (Miller & Rollnick, 2013) can also be found at the website, and supplemental resources for that book are available separately, for example reflection questions on each chapter.

45.6.9 Typical Examples of Implementation MI texts have been published in at least twentyseven languages, and more than 3,000 practitioners speaking at least fifty languages have received preparation as trainers through the MINT organization (MINT, 2019; Miller & Moyers, 2017). There is no other evidence-based treatment that has achieved the same level of worldwide dissemination, including cognitive behavioral therapies (Carroll, 2016); see Appendix 45.5 for further details on basic training in MI via a massive online open course (MOOC) as well as typical training syllabi. One of the greatest challenges of MI implementation is training practitioners to implement the method with adequate levels of adherence and competence (Decker et al., 2013) and, as with all other evidence-based treatments, transferring the knowledge acquired in MI training to meaningful practitioner change in clinical practice has proved to involve a great amount of implementation efforts (Hall et al., 2015). An example of this is the extensive implementation of MI at the Swedish National Board of Institutional Care (SiS), a Swedish government agency for adults with substance use disorders. As part of the implementation of MI by the SiS, all employees who interact with clients are offered MI training, and extended training is

Motivational Interviewing Interventions

offered to practitioners who use MI in therapy treatment sessions within the agency. A project manager and four national trainers at SiS, all members of MINT, supervise implementation and MI quality assurance within the organization. Additionally, another team within the organization further supports the MI implementation, with the main purpose of detecting synergies and delimitations from other methods and government assignments and anchoring MI within all parts of the organization. However, in a study that assessed 134 SiS practitioners’ MI skills to evaluate the impact of different ways of providing MI supervision, not all participants met the benchmarks for MI proficiency (Beckman, 2018).

45.7 Conclusion This chapter has provided an evidence-based overview of MI as applied for behavior change, as well as offering a brief guide to the design, evaluation, and implementation of MI interventions. More than 100 reviews and meta-analyses of clinical trials applying MI to change behavior have been published, and the general trend suggests that MI is an effective means to change behavior in multiple behavioral contexts and populations (Frost et al., 2018). However, no MI-based intervention has been shown to be superior for all clients under all conditions (Prochaska & Norcross, 2007), and a comprehensive description of the effects of MI in all its applications and adaptations is beyond the scope of this chapter. As described in Sidebar 45.2 and Sections 45.4 and 45.6.2, reviews have been published on the use of MI in general, for specific behaviors, and for specific settings. Readers are directed to Frost et al.’s (2018) meta-analysis of reviews on the effectiveness of MI on adult behavior change in health and social care settings, but it is also recommended that readers attend to new reviews and research syntheses of MI-based behavior change interventions as they become available.

673

References Amrhein, P. C., Miller, W. R., Yahne, C. E., Palmer, M., & Fulcher, L. (2003). Client commitment language during motivational interviewing predicts drug use outcomes. Journal of Consulting and Clinical Psychology, 71, 862–878. https://doi .org/10.1037/0022-006X.71.5.862 Babor, T. F., & Del Boca, F. (2003). Treatment Matching in Alcoholism. Cambridge: Cambridge University Press. Bahner, J., & Stenqvist, K. (2019). Motivational Interviewing as evidence-based practice? An example from sexual risk reduction interventions targeting adolescents and young adults. Sexuality Research and Social Policy. https://doi.org/ 10.1007/s13178-019-00388-y Beckman, M. (2018). Dissemination of Motivational Interviewing: The role of workshop training and subsequent supervision in the development of competence in clinical practice. PhD thesis, Karolinska Institutet. Beckman, M., Forsberg, L., Lindqvist, H., Diez, M., Persson, J., & Ghaderi, A. (2017). The dissemination of motivational interviewing in Swedish county councils: Results of a randomized controlled trial. PLoS ONE, 12, 7. https://doi.org/ 10.1371/journal.pone.0181715 Bem, D. J. (1972). Self-perception theory. Advances in Experimental Social Psychology, 6, 1–62. https:// doi.org/10.1016/S0065-2601(08)60024–6 Berman, A. H., Biguet, G., Stathakarou, N. et al. (2017). Virtual patients in a behavioral medicine massive open online course (MOOC): A qualitative and quantitative analysis of participants’ perceptions. Academic Psychiatry, 41, 631–641. https://doi.org/ 10.1007/s40596-017-0706-4 Buckner, J., & Schmidt, N. (2009). A randomized pilot study of motivation enhancement therapy to increase utilization of cognitive-behavioral therapy for social anxiety. Behaviour Research and Therapy, 47, 710–715. https://doi.org/ 10.1016/j.brat.2009.04.009 Burke, B. L., Arkowitz, H., & Menchola, M. (2003). The efficacy of motivational interviewing: A meta-analysis of controlled clinical trials. Journal of Consulting and Clinical Psychology, 71, 843– 861. https://doi.org/10.1037/0022-006X.71.5.843

674

ANNE H. BERMAN, MARIA BECKMAN, AND HELENA LINDQVIST

Carroll, K. M. (2016). Training in vain? Addiction, 111, 1153–1154. https://doi.org/10.1111/add.13175 Christoff, A. D. O., & Boerngen-Lacerda, R. (2015). Reducing substance involvement in college students: A three-arm parallel-group randomized controlled trial of a computer-based intervention. Addictive Behaviors, 45, 164–171. https://doi .org/10.1016/j.addbeh.2015.01.019 Cushing, C. C., Jensen, C. D., Miller, M. B., & Leffingwell, T. R. (2014). Meta-analysis of motivational interviewing for adolescent health behavior: Efficacy beyond substance use. Journal of Consulting and Clinical Psychology., 82, 1212– 1218. https://doi.org/10.1037/a0036912 D’Amico, E. J., Houck, J. M., Hunter, S. B., Miles, J. N., Osilla, K. C., & Ewing, B. A. (2015). Group motivational interviewing for adolescents: Change talk and alcohol and marijuana outcomes. Journal of Consulting and Clinical Psycholology, 83, 68–80. https://doi.org/10.1037/a0038155 Decker, S. E., Carroll, K. M., Nich, C., Canning-Ball, M., & Martino, S. (2013). Correspondence of motivational interviewing adherence and competence ratings in real and role-played client sessions. Psychological Assessment, 25, 306–312. https://doi.org/10.1037/a0030815 Dobber, J., van Meijel, B., Barkhof, E. et al. (2015). Selecting an optimal instrument to identify active ingredients of the motivational interviewingprocess. Journal of Psychosomatic Research, 78, 268–276. https://doi.org/http://dx.doi.org/ 10.1016/j.jpsychores.2014.10.010 Festinger, L. (1962). A Theory of Cognitive Dissonance, Vol. 2. Paulo Alto, CA: Stanford University Press. Forsberg, L., Wickström, H., & Källmén, H. (2014). Motivational interviewing may facilitate professional interactions with inspectees during environmental inspections and enforcement conversations. PeerJ, 2, e508. https://doi.org/ 10.7717/peerj.508 Frost, H., Campbell, P., Maxwell, M. et al. (2018). Effectiveness of motivational interviewing on adult behaviour change in health and social care settings: A systematic review of reviews. PLoS ONE, 13, e020489. https://doi.org/10.1371/journal.pone .0204890

Gates, P. J., Norberg, M. M., Copeland, J., & Digiusto, E. (2012). Randomized controlled trial of a novel cannabis use intervention delivered by telephone. Addiction, 107, 2149–2158. https://doi.org/ 10.1111/j.1360-0443.2012.03953.x Gaume, J., McCambridge, J., Bertholet, N., & Daeppen, J.-B. (2014). Mechanisms of action of brief alcohol interventions remain largely unknown: A narrative review. Frontiers in Psychiatry, 5, 108. https://doi .org/10.3389/fpsyt .2014.00108 Hall, K., Staiger, P. K., Simpson, A., Best, D., & Lubman, D. I. (2015). After 30 years of dissemination, have we achieved sustained practice change in motivational interviewing? Addiction, 7, 1144– 1150. https://doi.org/10.1111/add.13014 Hardcastle, S. J., Fortier, M., Blake, N., & Hagger, M. S. (2017). Identifying content-based and relational techniques to change behaviour in motivational interviewing. Health Psychology Review, 11, 1–16. https://doi.org/10.1080/17437199.2016 .1190659 Heckman, C. J., Egleston, B. L., & Hofmann, M. T. (2010). Efficacy of motivational interviewing for smoking cessation: A systematic review and metaanalysis. Tobacco Control, 19, 410–416. https:// doi.org/10.1136/tc.2009.033175 Hettema, J., Steele, J., & Miller, W. R. (2005). Motivational interviewing. Annual Review of Clinical Psychology, 1, 91–111. https://doi.org/ 10.1146/annurev.clinpsy.1.102803.143833 Houck, J. M., Moyers, T. B., Miller, W. R., Glynn, L. H., & Hallgren, K. A. (2010). Motivational Interviewing Skill Code (MISC) Version 2.5. https://casaa.unm.edu/download/misc25.pdf Jiang, S., Wu, L., & Gao, X. (2017). Beyond face-toface individual counseling: A systematic review on alternative modes of motivational interviewing in substance abuse treatment and prevention. Addictive Behaviors, 73, 216–235. https://doi.org/ 10.1016/j.addbeh.2017.05.023 Jonas, D. E., Garbutt, J. C., Brown, J. M. et al. (2012). Screening, Behavioral Counseling, and Referral in Primary Care to Reduce Alcohol Misuse (Report No. 64). Rockville, MD: Agency for Healthcare Research and Quality. Kaner, E. F. S., Beyer, F. R., Muirhead, C. et al. (2018). Effectiveness of brief alcohol interventions in primary care populations. Cochrane Database of

Motivational Interviewing Interventions

Systematic Reviews, 2, CD004148. https://doi.org/ 10.1002/14651858.CD004148.pub4 Kohler, S., & Hofmann, A. (2015). Can motivational interviewing in emergency care reduce alcohol consumption in young people? A systematic review and meta-analysis. Alcohol and Alcoholism, 50, 107–117. https://doi.org/10.1093/alcalc/agu098 Koken, J., Naar-King, S., Umasa, S. et al. (2012). A crosscultural three-step process model for assessing motivational interviewing treatment fidelity in Thailand. Health Education and Behavior, 39, 574– 582. https://doi.org/10.1177/1090198111423679 Li, L., Zhu, S., Tse, N., Tse, S., & Wong, P. (2016). Effectiveness of motivational interviewing to reduce illicit drug use in adolescents: A systematic review and meta-analysis. Addiction, 111, 795–805. https:// doi.org/10.1111/add.13285 Lundahl, B., Moleni, T., Burke, B. L. et al. (2013). Motivational interviewing in medical care settings: A systematic review and meta-analysis of randomized controlled trials. Patient Education and Counseling, 93, 157–168. https://doi.org/ 10.1016/j.pec.2013.07.012 Lundahl, B. W., & Burke, B. (2009). The effectiveness and applicability of motivational interviewing: A practice-friendly review of four meta-analyses. Journal of Clinical Psychology, 65, 1–14. https:// doi.org/10.1002/jclp.20638 Lundahl, B. W., Kunz, C., Brownell, C., Tollefson, D., & Burke, B. L. (2010). A meta-analysis of motivational interviewing: Twenty-five years of empirical studies. Research on Social Work Practice, 20, 137–160. https://doi.org/10.1177/1049731509347850 Madson, M., & Campbell, T. (2006). Measures of fidelity in motivational enhancement: A systematic review. Journal of Substance Abuse Treatment, 31, 67–73. https://doi.org/10.1016/j.jsat.2006.03.010 Magill, M., Apodaca, T. R., Borsari, B. et al. (2018). A meta-analysis of motivational interviewing process: Technical, relational, and conditional process models of change. Journal of Consulting and Clinical Psychology, 86, 140–157. https://doi .org/10.1037/ccp0000250 Maltby, N., & Tolin, D. F. (2005). A brief motivational intervention for treatment-Refusing OCD patients. Cognitive and Behavioral Therapy, 34, 176–184. https://doi.org/10.1080/165060705100 43741

675

Martino, S., Ball, S., Nich, C., Frankforter, T., & Carroll, K. (2009). Correspondence of motivational enhancement treatment integrity ratings among therapists, supervisors, and observers. Psychotherapy Research, 19, 181–193. https://doi.org/10.1080/10503300802688460 Mason, M. J., Campbell, L., Way, T. et al. (2015). Development and outcomes of a text messaging tobacco cessation intervention with urban adolescents. Substance Abuse, 36, 500–506. https://doi.org/10.1080/08897077.2014.987946 McCambridge, J., Day, M., Thomas, B., & Strang, J. (2011). Fidelity to motivational interviewing and subsequent cannabis cessation among adolescents. Addictive Behaviors, 36, 749–754. https://doi.org/10.1016/j.addbeh.2011.03.002 Michie, S., Richardson, M., Johnston, M. et al. (2013). The behavior change technique taxonomy (v1) of 93 hierarchically clustered techniques: Building an international consensus for the reporting of behavior change interventions. Annals of Behavioral Medicine, 46, 81–95. Miller, W. R. (1983). Motivational interviewing with problem drinkers. Behavioural Psychotherapy, 11, 147–172. https://doi.org/10.1017/s014134730000 6583 Miller, W. R., & Moyers, T. (2017). Motivational interviewing and the clinical science of Carl Rogers. Journal of Consulting and Clinical Psychology, 85, 757–766. https://doi.org/10.1037/ccp0000179 Miller, W. R., & Rollnick, S. (2009). Ten things that motivational interviewing is not. Behavioural and Cognitive Psychotherapy, 37, 129–140. https:// doi.org/10.1017/S1352465809005128 Miller, W. R., & Rollnick, S. (2013). Motivational Interviewing: Helping People Change (3rd ed.). New York: Guilford Press. Miller, W. R., & Rollnick, S. (2014). The effectiveness and ineffectiveness of complex behavioral interventions: Impact of treatment fidelity. Contemporary Clinical Trials, 37, 234–241. https://doi.org/10.1016/j.cct.2014.01.005 Miller, W. R., & Rose, G. S. (2009). Toward a theory of motivational interviewing. American Psychologist, 64, 527–537. https://doi.org/10.1037/a0016830 Miller, W. R., Yahne, C. E., Moyers, T. B., Martinez, J., & Pirritano, M. (2004). A randomized trial of

676

ANNE H. BERMAN, MARIA BECKMAN, AND HELENA LINDQVIST

methods to help clinicians learn motivational interviewing. Journal of Consulting and Clinical Psychology, 72, 1050–1062. https://doi.org/ 10.1037/0022-006X.72.6.1050 MINT (Motivational Interviewing Network of Trainers). (2019). MINT. Website. https://motiva tionalinterviewing.org Moyers, T., Rowell, L., Manuel, J., Ernst, D., & Houck, J. (2016). The Motivational Interviewing Treatment Integrity Code (MITI 4): Rationale, preliminary reliability and validity. Journal of Substance Abuse Treatment, 65, 36–42. https:// doi.org/10.1016/j.jsat.2016.01.001 O’Donnell, A., Anderson, P., Newbury-Birch, D. et al. (2014). The impact of brief alcohol interventions in primary healthcare: a systematic review of reviews. Alcohol and Alcoholism, 49, 66–78. https://doi.org/10.1093/alcalc/agt170 Pace, B. T., Dembe, A., Soma, C. S., Baldwin, S. A., Atkins, D. C., & Imel, Z. E. (2017). A multivariate meta-analysis of motivational interviewing process and outcome. Psychology of Addictive Behaviors, 31, 524–533. https://doi.org/10.1037/ adb0000280 Page, K. M., & Tchernitskaia, I. (2014). Use of motivational interviewing to improve return-towork and work-related outcomes: A review. The Australian Journal of Rehabilitation Counselling, 20, 38–49. https://doi.org/10.1017/jrc.2014.5 Prochaska, J. O., & Norcross, J. C. (2007). Systems of Psychotherapy: A Transtheoretical Approach (7th ed.). New York: Brooks/Cole. Rogers, C. R. (1956). Becoming a person. Pastoral Psychology, 7, 9–13. https://doi.org/10.1007/ BF01560065 Schmidt, L., Andersen, K., Nielsen, A., & Moyers, T. (2019). Lessons learned from measuring fidelity with the Motivational Interviewing Treatment Integrity code (MITI 4). Journal of Substance Abuse Treatment, 97, 59–67. https://doi.org/ 10.1016/j.jsat.2018.11.004 Schwalbe, C. S., Oh, H. Y., & Zweben, A. (2014). Sustaining motivational interviewing: A metaanalysis of training studies. Addiction, 109, 1287– 1294. https://doi.org/10.1111/add.12558 Shingleton, R., & Palfai, T. P. (2016). Technologydelivered adaptations of motivational

interviewing for health-related behaviors: A systematic review of the current research. Patient Education and Counseling, 99, 17–35. https://doi .org/10.1016/j.pec.2015.08.005 Tanner-Smith, E. E., & Lipsey, M. W. (2015). Brief alcohol interventions for adolescents and young adults: A systematic review and meta-analysis. Journal of Substance Abuse Treatment, 51, 1–18. https://doi.org/10.1016/j.jsat.2014.09.001 Terry, J., Strait, G., Smith, B., & McQuillin, S. (2013). Replication of motivational interviewing to improve middle school students’ academic performance. Journal of Community Psychology, 41, 902–909. https://doi.org/doi.org/10.1002/jcop.21574. Toneatto, T. (2016). Single-session interventions for problem gambling may be as effective as longer treatments: Results of a randomized control trial. Addictive Behaviors, 52, 58–65. https://doi.org/ 10.1016/j.addbeh.2015.08.006 Vallabhan, M. K., Jimenez, E. Y., Nash, J. L. et al. (2018). Motivational interviewing to treat adolescents with obesity: A meta-analysis. Pediatrics, 142, e20180733. https://doi.org/ 10.1542/peds.2018-0733 Villarosa-Hurlocker, M. C., O’Sickey, A. J., Houck, J. M., & Moyers, T. B. (2019). Examining the influence of active ingredients of motivational interviewing on client change talk. Journal of Substance Abuse and Treatment, 96, 39–45. https://doi.org/10.1016/j.jsat.2018.10.001 Wain, R., Kutner, B., Smith, J. et al. (2015). Self-report after randomly assigned supervision does not predict ability to practice motivational interviewing. Journal of Substance Abuse Treatment, 57, 96–101. https://doi.org/10.1016/j.jsat.2015.04.006 Werner, H., Hakeberg, M., Dahlström, L. et al. (2016). Psychological interventions for poor oral health: A systematic review. Journal of Dental Research, 95, 506–514. https://doi.org/10.1177/00220345 16628506. Westra, H. A., Constantino, M. J., & Antony, M. M. (2016). Integrating motivational interviewing with cognitive-behavioral therapy for severe generalized anxiety disorder: An allegiancecontrolled randomized clinical trial. Journal of Consulting and Clinical Psychology, 84, 768–782.

46 The Science of Behavior Change: The Road Ahead Martin S. Hagger, Linda D. Cameron, Kyra Hamilton, Nelli Hankonen, and Taru Lintunen

46.1 Introduction The Handbook of Behavior Change adopts a theory- and evidence-based approach to scientific inquiry into, and the practice of, behavior change. Drawing from multiple disciplines and perspectives, the handbook provides comprehensive coverage of topics in three parts: Part I covers a selection of the most prominent theories that have been commonly applied to explain behavior change;1 Part II reviews models and processes that have been adopted to develop behavior change interventions and the methodological and pragmatic considerations that need to be accounted for when implementing and evaluating them; and Part III provides the evidence base, specific guidelines and considerations, and steps for developing and implementing behavior change interventions using particular approaches. Although the science of behavior change is a relatively new discipline, the handbook illustrates the intense interest, breadth of approaches, and complexity of issues and considerations that need to be accounted for when seeking to understand and change behavior. An “at a glance” summary of the key contributions to behavior change covered in the handbook is provided in Table 46.1 under five global themes: (1) individual approaches to behavior change; (2) social, ecological, and environmental perspectives on behavior change; (3) behavior change interventions: development, implementation, and evaluation; (4) considerations in developing behavior change interventions; and (5) innovative methods in behavior change. Each global theme is

organized into several subthemes that reflect major contributions in each area. Chapters relating to each subtheme along with further relevant articles and sources, key concepts and theories or models, and a summary of its significance for behavior change are also identified. Aside from summarizing the current state-ofthe-art in the science of behavior change, the handbook also aims to advance research on, and understanding of, behavior change. In keeping with this goal, this chapter identifies some emerging areas of behavior change that represent important topics going forward and outlines some priority questions and recommendations that will set the agenda for future research. The chapter is organized into three sections: trends, gaps, and issues in the development of behavior change theory; issues in intervention development; and suggestions for “best-practice” guidelines for behavior change.

46.2 Development of Theory on Behavior Change: Trends, Gaps, and Ongoing Issues 46.2.1 Moving on from Individual Approaches to Behavior Change Keen observers will note the striking preponderance of individual approaches to behavior among

1

For a comprehensive list and basic description of theories that have been applied to understand behavior change, the reader is directed to Michie, West, Campbell, Brown, and Gainforth’s (2014) book on the subject. https://doi.org/10.1017/9781108677318.046

Table 46.1 Themes emerging from the handbook with source chapters and further reading, key concepts, and a summary of the significance of each theme for behavior change Key Chapters and Citations

Global Theme

Theme

Individual approaches to behavior change

Social cogni- Chapters 2, 3, 4, 5, 10, tion theories 15, 31, 32, 34; Ajzen (1991); Bandura (1986); Leventhal et al. (1980); Rogers (1975); Rosenstock (1974) Chapters 6, 7, 15, 39; Dual-phase Heckhausen & approaches Gollwitzer (1987); and action Schwarzer (2008) planning

Implicit/ automatic processes

Key Concepts/Theories

Significance for Behavior Change

Identification of belief- and attitudeConcepts: information processing; based determinants of behavior and attitudes; beliefs; aelf-efficacy; behavior change techniques aimed anticipated affect; risk perceptions at changing beliefs (e.g., informaTheories/models: social cognitive tion provision and persuasion, theory; theories of reasoned action experiences of success, positive and planned behavior; health belief feedback, fear appeals) model; protection motivation theory; common sense model Concepts: motivational vs. volitional Resolving limitation of weak intention-behavior relations by specifyphases; implementation intentions; ing separate motivational and implemental mindsets volitional phases of action; planAction and coping planning ning constructs identified as key Theories/models: model of action means convert intentions into phases; health action process action; effectiveness of intervenapproach; integrated theories tions including planning techniques such as implementation intentions, and action and coping planning Chapters 12, 13, 14, Concepts: reflective-impulsive deter- Recognition of nonconscious, auto15, 34, 36, 41, 42; matic determinants of behavior; minants (system 1 vs. system 2); Utilization of strategies targeting habits; automaticity; implicit Gibbons et al. those who are highly vulnerable to processes (1998); Hagger & impulses temptations or strong Theories/models: prototype willingChatzisarantis maladaptive habits (e.g., environness model; integrated behavioral (2014); Montaño & mental restructuring, nudging), and model; integrated behavior change Kasprzyk (2015); promoting adaptive habits (e.g., model Strack & Deutsch repeated experience, cue (2004); Thaler & awareness) Sunstein, 2008; Triandis (1977) Continued

General theories of motivation that Chapters 8, 9, 11, 15, Concepts: incentives; intrinsic vs focus on quantity of motivation – extrinsic motivation; motivational 33, 35, 37, 38, 40, such as incentives and rewards – quality; self-control 45; Deci & Ryan can be powerful influences on (1985); Hofmann & Theories/models: self-determination behavior; goals also scaffold motitheory; integrated self-control theKotabe (2012); vation, particularly goals with perory; goal setting theory Locke & Latham sonal relevance and SMART (2019) features; motivational quality (e.g., intrinsic and self-determined motivation) can be important for persistence; individual differences in capacities to self-regulate (e.g., self-control) promote general motivation toward behaviors Behavior is determined by influences Concepts: multiple levels of influSocial, ecological, and environmental Ecological and Chapters 17, 18, 28; from multiple levels within society: ence; collective efficacy; interperBandura (1986); perspectives on behavior change community policy, community, organizational, sonal contact; community and Bronfenbrenner approaches social, and individual; accounting coalition building; social capital – (1977); McLeroy et to behavior for influences at each level is cenbonding, bridging, linking; social al. (1988) change tral to providing a comprehensive norms; community consciousness description of behavioral determiTheories/models: ecological systems nants. Changing behavior within theory; social cognitive theory; community contexts requires consocial ecological model; systems sideration of social norms, social theory; theories of power; empowsettings, and social policy erment theories; social network/ social support theories; theories of stigma and discrimination; diffusion of innovations theory; organizational development and change theories; social norms theories; theories of power and process; coalition frameworks; social capital and community capacity General theories of motivation and individual differences

Continued

Table 46.1 (Cont.)

Global Theme

Theme

Key Chapters and Citations

Key Concepts/Theories

Significance for Behavior Change

theories; conscientization; community organization; multiplex policy networks Concepts: incentives; rewards; choice Rewards and incentives are effective Environment Chapters 14, 41, 42; architecture; nudge; heuristics and environmental contingencies that Thaler & Sunstein and beha(2008); Kahneman biases motivate behavior change based on vioral eco(2011); Daw et al. Theories/models: economic theory; classic economic models. Nudge nomic approaches (2011) incentive theory; “model-based” and choice architecture intervenvs. “model-free” framework tions promote behavior change by restructuring the environment at the point of decision-making without limiting choice Social and Chapters 16, 43, 44; Concepts: group norms; social iden- Individuals’ behavior is a function of group-based Tajfel & Turner tity; group identification their identification with the groups to which they belong and the extent approaches (1986); Turner et al. Theories/models: social identity theory; self-categorization theory; to which they self-categorize (1987); Jetten et al. (2012) social identity model of behavior themselves as typical members of change; social “cure” the group; Individuals identifying with groups will behave consistently with norms, which can be manipulated and leveraged to change behavior Provides researchers and intervention Concepts: mechanisms of action; Behavior change interventions: Approaches to Chapters 19, 20, 21; designers with steps to intervention logic model; behavior change development, implementation, and behavior Abraham (2012), development beginning with protechniques evaluation change interBartholomew blem identification, identification Theories/models: intervention mapvention Eldredge et al. of techniques and mechanisms of ping; behavior change wheel; thedevelopment (2016); Sheeran et change, consideration of design oretical domains framework al. (2017); Presseau elements, and implementation et al. (2019); components Michie, van Continued

Straalen, & West (2011) Evaluation of Chapters 22, 46; behavior Glasgow et al. change (2019); Skivington interventions et al. (2019)

Considerations in developing behavior change interventions

Different approaches to behavioral intervention evaluation proposed: (1) efficacy; (2) “real-world” effectiveness; (3) process evaluation; or (4) effectiveness in the context of wider context and systems. High-quality evaluation includes efficacy and process evaluation, stakeholder involvement, consideration of wider context and systems, and economic evaluation Beyond efficacy and effectiveness, Evidence Chapters 23, 24; Concepts: implementation science; demonstrable evidence for the translation Damschroder et al. implementation theories; evidencetranslation and implementation of and imple(2009); Glasgow et based frameworks; translational behavior change interventions in mentation in al. (1999); Logan & research “real-world” contexts is needed. behavior Graham (2010) Theories/models: the Ottawa model Implementation science provides change of research use; consolidated framodels for translation to ensure mework for implementation optimal engagement and efficacy research; RE-AIM framework within the populations, contexts, and settings targeted by the intervention Efficacy and effectiveness of behavior Involving sta- Chapters 21, 22, 24, Concepts: tailoring; needs assesschange interventions can be optikeholders 25; Glasgow et al. ment; user attitudes and beliefs; mized by maximizing user engageand user (1999); May et al. person-based approach; intervenment and engaging relevant engagement (2007); Skivington tion co-design; patient and public stakeholders in the design process. et al. (2019) involvement Tailoring intervention content to Theories/models: normalization prothe target population based on data cess theory; RE-AIM framework; collected during development and UK Medical Research Council’s involving stakeholders such as Concepts: types of evaluation; program theory; implementation science; economic evaluation Theories/models: RE-AIM framework; UK Medical Research Council’s (MRC) guidance on complex interventions

Continued

Table 46.1 (Cont.)

Global Theme

Theme

Key Chapters and Citations

Key Concepts/Theories

Significance for Behavior Change

(MRC) guidance on complex interventions

Addressing disparities

Innovative methods

Qualitative and critical approaches

Use of technology to

patients and health professionals in the intervention development process will promote uptake among users and “real-world” effectiveness Chapter 27; Engel Concepts: inequity; social disparities; Observed disparities in behavior change necessitates a full under(1977); Kreuter et socioeconomic status; culturally standing of the mechanisms that al. (2003); McLeroy appropriate models underpin disparities and how they et al. (1988); Myers Theories/models: social ecological can be addressed. Research and (2009); Resnicow et model; biopsychosocial model; practice in behavior change needs a al. (1999) minority stress model; model of stronger focus on equity issues and cultural sensitivity; model of culthe development of behavior turally tailored health change interventions that are sencommunications sitive to culture and socioeconomic inequities Chapter 30; Concepts: qualitative methods; criti- Traditional theories and models Hargreaves (2011); cal perspectives; social context applied to behavior change tend to Lyons & Theories/models: social practice focus on unitary, individualized Chamberlain (2017) theory explanations of behavior change, and fail to adequately account for contextual and experiential influences. Qualitative approaches offer an alternative that takes into account that behaviors are contextualized, complexly situated, and socially and culturally enabled and patterned Chapters 29, 37, 43; de Interventionists can leverage digital Bruin et al. (2012); tools, mobile devices, and web- and Continued

change behavior

Heber et al. (2017); Concepts: digital tools and devices; mobile technology; eHealth; apps; Webb et al. (2010); web-based interventions Harkin et al. (2016) Theories/models: theories of selfregulation; integrated self-control theory; tailoring and digital coaching

online-based interventions to deliver tailored interventions, collect detailed real-time behavioral data, and allow for effective behavioral monitoring. Such interventions are cost-effective, have wide support, improve reach and precision of intervention delivery, and are increasingly supported by research. More systematic research is needed, and concerns surrounding data protection and selection of appropriate technology and intervention models need to be addressed

684

MARTIN S. HAGGER, LINDA D. CAMERON, KYRA HAMILTON, NELLI HANKONEN, AND TARU LINTUNEN

the theories reviewed in Part I. Many of the theories focus on the roles that socially defined intrapersonal beliefs, motives, and states play in determining behavior change, consistent with the social cognition approach (Conner and Norman, 2015). While these theories have made important contributions to understanding behavior change, the emphasis on individual theories highlights the relative dearth of broader perspectives that encompass group, social, ecological, and political determinants, and there have been calls for greater application of social theories (Moore, Cambon et al., 2019). Numerous alternatives to these predominantly individual approaches are also covered in the handbook, such as social identity theory (Tajfel & Turner, 1986; Chapter 16, this volume) and ecological and community models (Bronfenbrenner, 1977; Chapters 17 and 18, this volume). In addition, approaches emphasizing the importance of incorporating social demographic factors into explanations of behavior change, such as socioeconomic status and disparities, are also included (Chapter 27, this volume). However, interventions based on these broader approaches are relatively sparse and warrant greater attention (Chapters 43 and 44, this volume). Other perspectives that encompass these broader factors have been proposed (e.g., Borland, 2017; Johnson et al., 2010), and more research is required on how applications of such approaches can yield more comprehensive explanations of behavior change beyond theories that focus on individual determinants.

46.2.2 Clarity in Specifying and Operationalizing Theories An important issue arising from research on behavior change theories is the large number of theories available and the considerable variability in the quality of their descriptions of predictions (Davis et al., 2015; Michie, Carey et al., 2017; West et al., 2019). While many theories have good internal validity and clarity in their

specifications and predictions, others do not, making it difficult to establish the extent to which the theory is applicable and testable across behavior change contexts. A further issue is the vast number of constructs and mechanisms identified, which presents considerable challenges in synthesizing research on theories and identifying commonalities and redundancies across theories (Hagger, 2014; Michie et al., 2014). A related issue is the lack of clarity in describing the causal mechanisms that underpin relations among theory constructs (West et al., 2019). Further, few theory comparisons demonstrate the relative effectiveness of theories and predictions (e.g., Dzewaltowski, Noble, & Shaw, 1990; Weinstein, 1993), and few attempt to integrate and reconcile constructs and predictions across theories (Hagger, 2009; Rhodes, McEwan, & Rebar, 2019; Chapter 15, this volume). One solution to the issue of variability in theory specification is the development and application of reporting standards for describing theories. Such standards would entail the development of a common terminology or system to formally specify theories. For example, one research team is developing a set of formal terms and symbols based on systems theory to describe theories (West et al., 2019). Another approach is to develop formalized descriptions of theories using computational modeling (Fried et al., 2019), which can provide systematized descriptions of theory predictions that also encompass auxiliary assumptions and conditions on which the predictions depend (Trafimow, 2012). It is also important to develop standards to evaluate the quality of a theory in terms of its clarity and precision in description and potential to provide hypotheses that are not only empirically testable but testable using robust designs (Meehl, 1990; Trout, 2004). For example, Davis et al. (2015) have developed a theory quality checklist, which provides a preliminary means to evaluate theory specification and description. Furthermore, the field of behavior change should

The Science of Behavior Change: The Road Ahead

consider applying principles from philosophy of science to provide formal mechanistic descriptions of relations among theory constructs (Hedström & Ylikoski, 2010). Such an approach is highly relevant to providing theoretical explanations of how behavior change interventions work in changing behavior and guiding their process evaluation (Sheeran, Klein, & Rothman, 2017). Finally, means to deal with the vast number of constructs and mechanisms, many of which have similar content but different labels, have been developed (Michie et al., 2014). One approach has focused on developing classifications of links between theoretical components and behavior change techniques (Carey et al., 2019; Connell et al., 2019). Such an endeavor entails formal synthesis of constructs across theories alongside taxonomies of behavior change techniques. However, such research is in its relative infancy, and future research that applies such tools to behavior change theories is required to identify a core set of theoretical constructs and mechanisms capable of optimally explaining behavior change.

46.2.3 Beyond Silos: The Need for More Multidisciplinary Research This handbook illustrates the broad diversity in approaches to behavior change (see Table 46.1). The emerging science of behavior change has been informed by research and practice in traditional social science disciplines such as psychology, sociology, economics, and philosophy. However, comparatively new disciplines have also contributed to this understanding, including behavioral economics, behavioral medicine, translational medicine, and implementation science (e.g., Chapters 23 and 42, this volume). The diversity in approaches illustrates the intense interest in behavior change and a recognition that multiple disciplines can contribute to the development of behavioral solutions to many problems in society. Furthermore, an interdisciplinary

685

approach to behavior change could be considered a strength as leveraging methods and strategies across disciplines may afford novel solutions (see Spotswood, 2016). However, it is also clear that much of the research on behavior change tends to be conducted with relatively little interdisciplinary collaboration. Such “siloed” perspectives may impede progress in developing precise, comprehensive explanations of behavior and behavior change interventions. Nevertheless, there are good examples where an interdisciplinary approach has been effective in advancing knowledge of behavior change. For example, some perspectives on the development and implementation of behavior change interventions combine theory from psychology with design elements from translational medicine and implementation science (e.g., Chapters 21 and 23, this volume). However, such perspectives are relatively rare and there is a need to further facilitate initiatives in which teams from different disciplines collaborate on addressing priority issues on behavior change. For example, the Behavioral Medicine Research Council was founded by a multidisciplinary consortium of organizations including the American Psychological Association’s Society for Health Psychology (SfHP), the Academy of Behavioral Medicine Research (ABMR), the American Psychosomatic Society (APS), and the Society for Behavioral Medicine (SBM). The organization aims to identify research priorities and promote strategic goals for behavioral medicine research, of which behavior change is a key element (Freedland, 2019). Similarly, the Science of Behavior Change Research Network is a consortium of research organizations funded by the US National Institutes of Health that brings together basic and applied scientists from different disciplines to conduct research on behavior change. The consortium focuses on developing a better understanding of mechanisms and behavioral interventions in health contexts (National Institutes of Health, 2019) and has published a set of metareviews of current evidence on behavior change

686

MARTIN S. HAGGER, LINDA D. CAMERON, KYRA HAMILTON, NELLI HANKONEN, AND TARU LINTUNEN

interventions and their mechanisms of action (Hennessy et al., 2020; Suls et al., 2020; Wilson et al., 2020). These collaborative initiatives provide models for multidisciplinary research on behavior change that may facilitate novel solutions to behavior-related problems.

46.3 Issues in Behavior Change Intervention Development, Implementation, and Evaluation 46.3.1 Intervention Fidelity Fidelity is a key determinant of intervention efficacy (Bellg et al., 2004; see Chapters 21 and 22, this volume). Intervention fidelity focuses on whether the intervention components (e.g., intervention content such as messages and behavior change techniques) are delivered to the target population in the intended manner or, if the intervention is self-administered, whether the recipient carries out the intervention according to protocol. Bellg et al. (2004) indicate that fidelity applies to multiple aspects of behavioral interventions: study design, provider training, treatment delivery, treatment receipt, and enactment of treatment skills. Fidelity has been identified as a key moderator of behavior change intervention efficacy in meta-analyses of randomized controlled trials (Durlak & DuPre, 2008; Hardeman et al., 2007). However, research suggests that intervention fidelity is a neglected aspect, with few behavioral intervention trials incorporating procedures to ensure adequate fidelity and measures to assess fidelity. Furthermore, many trials that have included fidelity checks have assessed only some aspects of fidelity (e.g., Rixon et al., 2016; Walton et al., 2017). These deficiencies occur despite the existence of frameworks to guide intervention fidelity procedures (Bellg et al., 2004) and the inclusion of intervention fidelity assessments as integral aspects of intervention development approaches (Abraham, 2012;

Presseau et al., 2019; see Chapter 21, this volume). As with initiatives to improve reporting of intervention design and content, advocacy to promote greater attention to fidelity issues is needed. In addition, researchers and intervention designers need prompts and guidance on evaluation methods to assess all salient aspects of behavior change intervention fidelity (Toomey et al., 2019).

46.3.2 New Approaches to the Translation, Feasibility, and Optimization of Behavior Change Interventions Findings from basic and applied research on behavior change indicate that interventions based on behavioral theory have considerable promise in addressing key behavior-related problems (Bartholomew & Mullen, 2011; Rhodes et al., 2019). However, there is, by comparison, relatively little research on the translation of these findings into workable interventions (often referred to as “Phase III” trials) that can be tested for effectiveness in ecologically valid, “realworld” settings. Increasing emphasis is being placed on processes and systems that outline how behavior change interventions with demonstrable efficacy in experiments and controlled trials can be translated into effective interventions that result in meaningful changes in target populations. Various models have been proposed that describe best-practice steps in translating the evidence base of behavior change interventions into workable behavior change solutions in real-world contexts, such as the Ottawa model of research use (Logan & Graham, 2010), the consolidated framework for implementation research (Damschroder et al., 2009), and the RE-AIM framework (Glasgow, Vogt, & Boles, 1999). Much of this work is informed by relatively new interdisciplinary fields such as translational medicine and implementation studies (see Chapter 23, this volume). Other frameworks have also been

The Science of Behavior Change: The Road Ahead

proposed, such as the ORBIT model aimed at developing behavioral interventions in chronic disease (Czajkowski et al., 2015). To date, however, translational activities have seldom been incorporated into behavior change intervention development protocols. In addition, utilization of innovative research designs focused on translation and real-world application in the early stages of intervention development has been advocated. For example, there have been calls for early-phase translational science practices as a routine part of intervention development. The ORBIT model, for example, offers step-by-step guidance for the development of translatable behavioral interventions, which includes key milestones for progression of intervention design and testing, with options to return to earlier stages for further refinement, feasibility testing, and optimization. The model proposes innovative study methods which, if used early in intervention development, may facilitate translation and optimization later down the track. These methods include human-centered design, behavioral event modeling, small-N studies, optimization methods (e.g., dose findings, optimizing treatment findings, developing adaptive treatments), and cluster randomized and pragmatic clinical trials (Naar, Czajkowski, & Spring, 2018). Utilization of such methods may yield more efficient interventions that are optimally effective for the desired context and target population. These procedures are relatively new, however, and few examples of behavior change interventions utilizing these designs exist. Research is needed to determine whether systematic adoption of these methods produces interventions that are optimally effective in evoking meaningful changes in behavioral outcomes in real-world settings.

46.3.3 Ethical Issues in Behavior Change The ethics underpinning behavior change campaigns and initiatives is an important but seldom

687

considered issue. For example, should governments and organizations implement means and strategies to change the behavior of a population? It is often assumed that the benevolent motives underpinning behavioral interventions, along with the substantive gains in terms of ameliorating problems faced by society, outweigh the moral and ethical concerns relating to freedom to choose and individual rights. However, such issues are rarely raised or debated. Interventions that change behavior through legislation and regulation (e.g., seatbelt use in motor vehicles, bans on tobacco smoking in public places, compulsory safety helmets for cyclists) are usually the consequence of overwhelming evidence supporting the benefits of the behavior, as well as years of lobbying work and political advocacy. Strong support for the legislation in public opinion polls is also important to allay politicians’ concerns over introducing unpopular measures. In the face of such universal public support, ethical concerns over personal freedoms become less imperative. Behavior change science could be used to help increase public acceptability of initially unpopular policies, such as various restrictions to tackle climate change (Marteau, 2017). Although legislation and regulation can be highly effective means to change behavior, they are often not possible, feasible, or sufficiently acceptable to be implemented and often do not have universal support. Other approaches may be necessary, including campaigns aimed at altering behavior through persuasion or other means. Nudge and choice architecture interventions are relatively recent approaches to behavior change (see Chapters 14 and 42, this volume). Such interventions are consistent with the philosophy of “liberal paternalism”: While they aim to change behavior by directing individuals toward a particular behavioral response, they do not negate individuals’ right to choose. Similarly, information campaigns that seek to persuade individuals to alter their behavior, or offer incentives to do so, do not undermine these rights. However, the

688

MARTIN S. HAGGER, LINDA D. CAMERON, KYRA HAMILTON, NELLI HANKONEN, AND TARU LINTUNEN

ethics of exposing individuals to particular choice scenarios or messages aimed at altering thought and behavior patterns still demands consideration. Furthermore, interventions that aim to change behavior at the population level can differentially affect certain segments of the population (see Chapter 27, this volume). For example, regulation strategies aimed at manipulating behavior through price increases, such as minimum prices for alcohol and taxes on sugar-sweetened beverages, can disproportionately affect individuals and families on lower incomes (Cawley et al., 2019; Ward, 2011). Given evidence that those on lower incomes are also most affected by the problems associated with the target behavior, this creates a moral dilemma for those tasked with developing and implementing such initiatives. Ethical considerations should be considered an important “meta-issue” pervading all aspects of behavior change. In the context of research, it is imperative that all trials of behavior change interventions are subjected to rigorous review by experts on ethics through institutional review boards, human research ethics committees, or similar organizational units. Such review primarily focuses on supporting participants’ ethical rights to choose – particularly in withdrawing from a trial or declining to engage in specific behaviors or assessment components without prejudice or cost – and on making decisions to approve trials by balancing the potential value and benefits of the research against the costs to participants. More broadly, development of behavior change interventions should involve user groups, that is, representative members of the target population, from the outset and include questions regarding the acceptability of the intervention from the standpoint of intrusion and personal choice (see Chapters 24 and 25, this volume). Similarly, interventionists should consider surveying the target population on the acceptability of introducing the intervention broadly in that community. Such work can assist in identifying potential ethical issues and

potential means to address such concerns in the population before the intervention has been developed and implemented. Ethical considerations should, therefore, form a routine part of the developmental procedures of behavior change interventions (see Chapter 21, this volume).

46.3.4 Evaluation of Mechanisms of Impact Behavior change interventions are predominantly evaluated through examination of effects on the primary outcome (e.g., changes in measures of behavior), while process evaluation is less frequently evaluated (see Chapter 22, this volume). Testing the mechanisms by which interventions lead to behavior change is an important component of process evaluation. Psychological constructs derived from behavioral theory are examples of process-related variables that have been proposed to explain or mediate effects of behavior change interventions on behavior. Changes in these constructs reflect the process or mechanism of action by which the behavior change technique(s) that comprise the intervention leads to change in the target behavior (see Chapters 19 and 20, this volume). Numerous authors make reference to a basic process model or theory of change (Sheeran et al., 2017 Hagger et al., 2020; Chapters 20 and 22, this volume; see Figure 46.1), which summarizes the relevant relations necessary for a process evaluation of interventions: (1) the effect of the behavior change technique on the theory-derived construct implicated in the mechanism (path a, Figure 46.1); and (2) the effect of the construct on behavior change (path b, Figure 46.1); and (3) the effect of the technique on behavior change, which represents the residual effect of the intervention independent of the indirect effect through the mediator (path c', Figure 46.1). The indirect effect of the intervention content on behavior change through the theoretical construct represents the mechanism of action of the intervention. A process evaluation of a behavior change intervention necessitates specification of a

The Science of Behavior Change: The Road Ahead

689

Mechanism of action

Effect of method or technique on measure of construct – should be experimentally verified

Change method or technique

Intervention content

a

A mediator

Relation between construct and behavior – should be experimentally verified

Modifiable factor (e.g., psychological construct)i

b

Behavior change

Behavioral outcome

c'

Residual effect of the intervention unaccounted for by the modifiable factor (mediator)

Figure 46.1 A basic model of a behavior change mechanism of action

process model, which will likely form part of the program theory or logic model of the intervention (Chapters 19 and 21, this volume). The mechanism is usually tested using mediation analyses, which test the extent to which the effect of change technique on behavior is “transmitted” through the theorybased construct (for more details see Hagger et al., 2020). In practice, the model is likely to be more elaborate because techniques change behavior through more than one construct, different individuals may change via different pathways, and interventions often comprise multiple techniques. However, the basic model provides a template for informing research that will contribute to an evidence base for behavior change techniques, the constructs they are purported to change, and change in behavior. Despite a growing literature on the importance of identifying mechanisms of action of behavior change interventions, evaluations of mechanisms of change are relatively rare. Many intervention reports do not specify a theoretical framework for the intervention (Michie, Carey et al., 2017;

Prestwich et al., 2014), and among those that do, few provide clear descriptions of the mechanisms of action. Among intervention trials that do measure theory-based constructs, relatively few conduct mediation analyses to test the process or conduct an a priori statistical power analysis ensuring that such process evaluation is feasible (Hennessy et al., 2020). In addition, mediation analyses that test intervention effects on behavior change through intermediate or interim measures of the theorybased mediator can be suboptimal to test the mechanism of change (Bullock, Green, & Ha, 2010; Fairchild & McDaniel, 2017). Instead, such analyses should estimate the indirect effect of the intervention on behavior change through change in the mediator itself (e.g., Renner et al., 2012). Research syntheses can contribute to knowledge of behavior change mechanisms of action. For example, Rhodes et al. (2020) conducted a metaanalysis of theory-based behavior change interventions in physical activity in which they tested the effects of intervention content on behavior change through the putative constructs implicated in the

690

MARTIN S. HAGGER, LINDA D. CAMERON, KYRA HAMILTON, NELLI HANKONEN, AND TARU LINTUNEN

theory-based mechanisms of action. This study provides a template for future syntheses of evidence on mechanisms of action and will contribute to future databases on how interventions work in changing behavior. A clear recommendation arising from this handbook is the imperative for researchers and interventionists to incorporate such process evaluations of behavioral interventions from the outset, and including analyses of the mechanisms of impact will advance the evidence base on the mechanisms involved in behavior change interventions. One barrier to process evaluation is the lack of formal terminology and descriptions linking theory constructs with the behavior change technique purported to change them as well as an appropriate means to describe them. To address this gap, researchers have proposed expertverified links between theory components and behavior change techniques based on published theories and the development of formal systems to describe those links (Carey et al., 2019; Connell et al., 2019). This work is part of broader projects (e.g., the Theories and Techniques of Behaviour Change Project and the Human Behaviour Change Project) aimed at developing ontological descriptions of behavior change interventions that comprise organized sets of relations between behavior change methods; theoretical techniques; intervention design components (e.g., means of delivery); features of the behavior, context, and population; and behavioral outcomes (Larsen et al., 2016; Michie, Aonghusa et al., 2019; Michie & Johnston, 2017; Michie, Rothman et al., 2019; Michie, Thomas et al., 2017). One of the aims of the projects is to develop a database of behavior change ontologies that is regularly modified and refined as new evidence emerges through machine learning. The database will enable researchers and interventionists to search for the specific sets of techniques, mechanisms of action, and intervention components that can inform the development and process evaluation of interventions.

46.3.5 Complex Systems and Behavior Change Interventions The complex systems approach is an emerging theme identified in many chapters of this handbook. Numerous authors have noted that behavior change interventions are not only complicated but also complex (Hawe, Shiell, & Riley, 2009; Moore, Evans et al., 2019). Complicated interventions may involve numerous interacting components but still can be divided into discrete sets of actions with predictable, stable, and linear consequences. However, many interventions are better defined as complex due to the emergent, unpredictable, and nonlinear associations between actions and outcomes. Humans are active agents, whose behavior continuously adapts in response to feedback from one another, and individuals’ behaviors are part of broader small group and community systems (Moore, Evans et al., 2019). Ideally, behavior change theory, interventions, and evaluations would take such aspects into account, including recursive causality (with reinforcing loops); disproportionate, nonlinear relationships (“tipping points”); and emergent outcomes (Rogers, 2008). This approach challenges the current mainstream view on behavior change interventions, where theories typically assume causal pathways with separate components, usually hypothesized to be linearly associated (see Figure 46.1 for an example). In the mainstream, psychological constructs are thought to be reducible to a set of independent components (component-dominant dynamics; Wallot & Kelty-Stephen, 2018). In the complex systems approach, the alternative view of causality assumes that consecutively measured values of a behavioral or physiological process are interdependent and irreducible to component parts (interaction-dominant dynamics). Thus far, available statistical approaches are limited in terms of their capacity to model complexity, so researchers have tended to study behavior with a toolbox of primarily linear methods (Wallot & Kelty-

The Science of Behavior Change: The Road Ahead

Stephen, 2018), but novel methods to evaluate mechanisms of behavior change appreciating its complex properties have emerged (Heino et al., 2019). In future, researchers are likely to further explore how complex systems theory can be utilized to better understand behavior change.

46.4 Considerations for “Best Practice” in Behavior Change 46.4.1 Other Intervention Approaches A major goal of this handbook is to provide up-todate, evidence-based, practical guidance on how to develop behavior change interventions. To this end, chapters in Part III provide broad coverage of prominent and emerging approaches to behavior change, with accompanying guidance on how to implement them. The approaches were selected on the basis of their prominence, frequency of use, and evidence base underpinning their use. However, it is important to note that some approaches have not been covered. Examples include mental contrasting (Oettingen, 2012) and cognitive behavioral therapy (CBT; Kendall & Hollon, 1979), both of which are briefly summarized here. Mental contrasting. Mental contrasting is a self-regulation technique in which individuals are prompted to visualize their desired future with respect to a given behavior or outcome and contrast it with their current state, identify obstacles responsible for the discrepancy, and put into place goals or behavioral strategies to overcome the obstacles to the desired outcome (Oettingen, 2012). A recent meta-analytic review of twelve studies applying mental contrasting interventions suggests that it can change health behaviors with small-to-medium effect sizes (Cross & Sheffield, 2019). As a relatively nascent strategy that extends techniques such as mental imagery and goal setting, it has not received full coverage in this handbook beyond a cursory mention (Chapter 33, this volume). CBT. CBT is a widely used strategy that aims to assist individuals in managing psychological

691

disorders and maladaptive behaviors by challenging and negating maladaptive beliefs and cognitions and developing problem-specific, goal-directed alternatives to the maladaptive behaviors (Kendall & Hollon, 1979). CBT comprises multiple techniques such as cognitive restructuring and goal setting. CBT has a long history and vast evidence base supporting its effectiveness (Butler et al., 2006; Tolin, 2010) but it is not covered in this handbook because of its predominant focus on the management of disorders in clinical populations rather than on behavior change more broadly.

46.4.2 Behavior Change Maintenance A key challenge facing interventionists is maintaining behavior change over time. Given that long-term maintenance is often requisite for adaptive outcomes to be realized (e.g., improvements in health, educational, environmental, and occupational outcomes), long-term evaluations of behavior maintenance and behavioral outcomes are paramount for interventions to be fit-for-purpose in offering solutions to problems. Many behavior change interventions have demonstrated efficacy and effectiveness in changing behavior in the short and medium term up to a few months post-intervention. However, relatively few intervention trials have demonstrated long-term maintenance of behavior change over many months or years. In many cases, issues around maintenance remain among the “unknowns” in the evidence base for behavior change interventions (Hagger et al., 2020), typically because time and budget constraints do not permit assessment of longterm (e.g., more than one year) maintenance. Researchers and stakeholders interested in behavior change maintenance should lobby research funders to provide sufficient resources for longitudinal intervention trials that can capture maintenance. In addition, interventionists must provide a clear rationale for evaluating behavior change maintenance when applying for research funds and a protocol on how they will do so in funding applications.

692

MARTIN S. HAGGER, LINDA D. CAMERON, KYRA HAMILTON, NELLI HANKONEN, AND TARU LINTUNEN

Resource availability for intervention components that promote behavior change maintenance is another key consideration. Interventions that aim for maximum efficacy in initiating behavior change are often complex and elaborate and therefore demand considerable resources, particularly human resource costs when the intervention is delivered in person (e.g., interactive client-practitioner sessions, group sessions). The allocation of intensive resources to behavior initiation is understandable, particularly in light of evidence that the length and dosage of behavior change interventions have been found to influence effectiveness (e.g., Burke, Arkowitz, & Menchola, 2003; Gillison et al., 2015). Consequently, however, interventionists may need to consider low-cost alternatives for assisting maintenance of behavior change, especially given that maintenance may not require the intensive methods used to initiate behavior change in the first place (e.g., Burke et al., 2003). Intervention designers can leverage alternatives such as digital and technology-based methods that deliver intervention content (e.g., “booster” messages, behavior monitoring) without the high costs associated with in-person delivery (Webb et al., 2010; see also Chapter 29, this volume). Such technologies may allow for extended delivery of intervention content to the target population and assist in maintaining intervention effects. Another alternative is to tap into existing alternative networks to deliver interventions involving professionals (e.g., health care professionals, community campaigners and leaders) who can be trained to deliver behavior change interventions. The development and evaluation of such alternatives for promoting maintenance, along with the need for long-term evaluations of behavior maintenance, should be priorities for future research.

46.4.3 Education and Training on Behavior Change As behavior change becomes a priority for organizations aiming to develop solutions to many societal

problems, training in the theories, principles, and practices of behavior change has become an integral part of the educational programs of many disciplines (e.g., psychology, sociology, economics) and professions (e.g., medicine, nursing, general practice, occupational management). Training in the science of behavior change is important to produce the next generation of researchers tasked with advancing knowledge of behavior change theory and practice. Behavior change should therefore become a key component of undergraduate and graduate degree programs for these disciplines. Further, training practitioners in multiple professions in the key principles of behavior change is essential to ensure their practice is evidence-based and consistent with the latest research and recommendations. Those tasked with providing in-service training of professionals whose jobs involve changing the behavior of clients need to incorporate training on behavior change practices. Such training should also be included in continuing professional development and top-up courses for qualified professionals whose purview includes changing the behavior of others (e.g., public health specialists and campaign managers, local government policy makers). The Handbook of Behavior Change can inform the content of behavior change training courses and serve as a reference for the latest evidence-based practices in behavior change. The three parts of the handbook offer a useful template for the development of academic training programs on the theory and evidence-based for behavior change as well as vocational and practice-based training on how to conduct and implement behavior change interventions. In addition, the multiple viewpoints presented in the chapters illustrate the diversity in the scientific disciplines that have been applied to understand behavior change and offer students a rounded, balanced perspective on the subject. Educators developing behavior change training programs and students of behavior change should also consult the many other resources on behavior change available that will augment and enrich learning. From a scientific perspective, cutting-edge

The Science of Behavior Change: The Road Ahead

research and evidence-based practice on behavior change are published in peer-reviewed publications in fields such as applied psychology, social science, translational medicine and implementation science, and behavioral economics (for examples, see Appendix 46.1). One of the optimal ways of identifying these publications is to use search engines (e.g., Google Scholar, Microsoft Academic) or databases (e.g., Web of Science, Scopus, PubMed) using relevant keywords. A further way of learning about recent developments in behavior change is to attend relevant scientific meetings, where scholars can be exposed to the latest research from scientists conducting research in behavior change. These scientific meetings also have special topics and interest groups relevant to behavior change, such as the Theories and Techniques of Behavior Change Interventions special interest group of the Society for Behavioral Medicine and the Intervention Science: Harnessing Psychology to Address Real-World Social Problems pre-conference of the Society of Personality and Social Psychology. In addition, scientists, students, and practitioners of behavior change may consider joining learned societies that represent key disciplines that conduct and promote work in behavior change such as applied psychology (e.g., International Association for Applied Psychology, Society for Personality and Social Psychology, Society for Health Psychology), behavioral medicine (e.g., International Society of Behavioral Medicine, Society of Behavioral Medicine), and motivation (e.g., Society of the Study of Motivation). Educators and scholars interested in the practice of behavior change within organizations should consider resources designed to train practitioners in behavior change such as the Improving Health: Changing Behavior – NHS Health Trainer Handbook (Michie et al., 2008). Further resources that could inform curricula on behavior change include major initiatives conducting large-scale research work and evidence syntheses on behavior change, including the Science of Behavior Change

693

Research Network (SOBC, 2019); the Human Behavior Change Project (Michie, Aonghusa et al., 2019); the Behavioral Medicine Research Council (Freedland, 2019); and the Behavioral Research Program of the National Cancer Institute’s Division of Cancer Control and Population Sciences (BRP, 2019). To date, however, there are no definitive materials or guidelines on the content of behavior change training courses and educational curricula in behavior change. The future of education on behavior change may lie in the development of common content that comprises expert-validated core and elective topics on behavior change. Such validated content will lead to more consistent, uniform training in behavior change and is the hallmark of a maturing discipline of study. The scientific community and learned bodies in behavior change have a key role to play in the development of such core curricula and this should be considered a future goal of this emerging science.

46.5 Conclusion Recognition of the behavioral origins of many problems in society today has led to a proliferation of interest and research in behavior change. Developing means to understand behavior change and design effective methods to change behavior is a priority agenda for governments and policy makers, research organizations and funders, and practitioners in multiple fields and disciplines. The increased emphasis placed on research inquiry and practice on behavior change is founded on the premise that changing behavior has been shown to offer effective solutions to many societal problems but has also been shown to be cost-effective. This chapter has identified key emerging issues and priority research directions arising from the handbook. From the perspective of theory development, there is a need for (1) a move away from individual theories and toward more integrative approaches that encompass social and ecological determinants of action; (2) clearer specification and operationalization of

694

MARTIN S. HAGGER, LINDA D. CAMERON, KYRA HAMILTON, NELLI HANKONEN, AND TARU LINTUNEN

behavior change theories; and (3) more interdisciplinary as opposed to siloed approaches to behavior change. In terms of behavior change intervention development, future research should consider (1) conducting more comprehensive and consistent evaluations of intervention fidelity; (2) utilizing innovative research methods, particularly in the design phase, for more effective translation, feasibility, and optimization of interventions; (3) ensuring ethical considerations are taken into account in the development and implementation of interventions; (4) conducting evaluations of the mechanisms of action of behavior change interventions; and (5) adopting a complex systems approach as an alternative paradigm in the development and evaluation of behavior change interventions and theories. In addition, ongoing development of behavior change intervention “best practice” should consider (1) broadening the scope to encompass approaches to behavior change from other disciplines; (2) evaluating the efficacy and effectiveness of interventions to produce long-term maintenance of behavior; and (3) developing core educational curricula to train the next generation of behavior change specialists. The growing interest in behavior change, and the research intensiveness in the field, suggests that the emerging science of behavior change is in good health and will continue to develop. The Handbook of Behavior Change represents a culmination of current work behavior change that can not only serve to provide a broad overview of theory and practice in this emerging science but also set the agenda for future research inquiry toward the development of optimal behavioral solutions to problems in society.

References Abraham, C. (2012). Mapping change mechanisms onto behaviour change techniques: A systematic approach to promoting behaviour change through text. In C. Abraham & M. Kools (Eds.), Writing Health Communication: An Evidence Based Guide (pp. 99–116). Thousand Oaks, CA: SAGE.

Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50, 179–211. https://doi.org/10.1016/ 0749-5978(91)90020-T Bandura, A. (1986). Social Foundations of Thought and Action: A Social-Cognitive Theory. Englewood Cliffs, NJ: Prentice Hall. Bartholomew, L. K., & Mullen, P. D. (2011). Five roles for using theory and evidence in the design and testing of behavior change interventions. Journal of Public Health Dentistry, 71, S20–S33. https:// doi.org/10.1111/j.1752-7325.2011.00223.x Bartholomew Eldredge, L. K., Markham, C. M., Ruiter, R. A. C., Fernández, M. E., Parcel, G. S., & Kok, G. (2016). Planning Health Promotion Programs: An Intervention Mapping Approach (4th ed.). San Francisco, CA: Jossey-Bass. Bellg, A. J., Borrelli, B., Resnick, B. et al. (2004). Special NIH committee report – Enhancing treatment fidelity in health behavior change studies: Best practices and recommendations from the NIH behavior change consortium. Health Psychology, 23, 443–451. Borland, R. (2017). CEOS theory: A comprehensive approach to understanding how to maintain behaviour change. Applied Psychology: Health and Well-Being, 9, 3–35. https://doi.org/10.1111/ aphw.12083 Bronfenbrenner, U. (1977). Toward an experimental ecology of human development. American Psychologist, 32, 513–531. https://doi.org/ 10.1037/0003-066X.32.7.513 BRP (Behavioral Research Program). (2019). Behavioral Research Program of the National Cancer Institute’s Division of Cancer Control and Population Sciences. Website. https://cancercon trol.cancer.gov/brp/about.html Bullock, J. G., Green, D. P., & Ha, S. E. (2010). Yes, but what’s the mechanism? (don’t expect an easy answer). Journal of Personality and Social Psychology, 98, 550–558. https://doi.org/10.1037/ a0018933 Burke, B. L., Arkowitz, H., & Menchola, M. (2003). The efficacy of motivational interviewing: A meta-analysis of controlled clinical trials. Journal of Consulting and Clinical Psychology, 71, 843–861. https://doi.org/10.1037/0022006x.71.5.843

The Science of Behavior Change: The Road Ahead

Butler, A. C., Chapman, J. E., Forman, E. M., & Beck, A. T. (2006). The empirical status of cognitivebehavioral therapy: A review of meta-analyses. Clinical Psychology Review, 26, 17–31. https:// doi.org/10.1016/j.cpr.2005.07.003 Carey, R. N., Connell, L. E., Johnston, M. et al. (2019). Behavior change techniques and their mechanisms of action: A synthesis of links described in published intervention literature. Annals of Behavioral Medicine, 53, 693–707. https://doi.org/10.1093/abm/kay078 Cawley, J., Thow, A. M., Wen, K., & Frisvold, D. (2019). The economics of taxes on sugar-sweetened beverages: A review of the effects on prices, sales, cross-border shopping, and consumption. Annual Review of Nutrition, 39, 317–338. https://doi.org/ 10.1146/annurev-nutr-082018-124603 Connell, L. E., Carey, R. N., de Bruin, M. et al. (2019). Links between behavior change techniques and mechanisms of action: An expert consensus study. Annals of Behavioral Medicine, 53, 708–720. https://doi.org/10.1093/abm/kay082 Conner, M. T., & Norman, P. (2015). Predicting and changing health behaviour: A social cognition approach. In M. T. Conner & P. Norman (Eds.), Predicting and Changing Health Behaviour: Research and Practice with Social Cognition Models (3rd ed., pp. 1–29). Maidenhead: Open University Press. >Cross, A., & Sheffield, D. (2019). Mental contrasting for health behaviour change: A systematic review and meta-analysis of effects and moderator variables. Health Psychology Review, 13, 209– 225. https://doi.org/10.1080/ 17437199.2019.1594332 Czajkowski, S. M., Powell, L. H., Adler, N. et al. (2015). From ideas to efficacy: The ORBIT model for developing behavioral treatments for chronic diseases. Health Psychology, 34, 971–982. https:// doi.org/10.1037/hea0000161 Damschroder, L. J., Aron, D. C., Keith, R. E., Kirsh, S. R., Alexander, J. A., & Lowery, J. C. (2009). Fostering implementation of health services research findings into practice: A consolidated framework for advancing implementation science. Implementation Science, 4, 50. https://doi.org/ 10.1186/1748-5908-4-50

695

Davis, R., Campbell, R., Hildon, Z., Hobbs, L., & Michie, S. (2015). Theories of behaviour and behaviour change across the social and behavioural sciences: A scoping review. Health Psychology Review, 9, 323–344. https://doi.org/ 10.1080/17437199.2014.941722 Daw, N. D., Gershman, S. J., Seymour, B., Dayan, P., & Dolan, R. J. (2011). Model-based influences on humans’ choices and striatal prediction errors. Neuron, 69, 1204–1215. https://doi.org/10.1016/j .neuron.2011.02.027 de Bruin, M., Sheeran, P., Kok, G. et al. (2012). Selfregulatory processes mediate the intentionbehavior relation for adherence and exercise behaviors. Health Psychology, 31, 695–703. https://doi.org/10.1037/a0027425 Deci, E. L., & Ryan, R. M. (1985). Intrinsic Motivation and Self-Determination in Human Behavior. New York: Plenum Press. Durlak, J. A., & DuPre, E. P. (2008). Implementation matters: A review of research on the influence of implementation on program outcomes and the factors affecting implementation. American Journal of Community Psychology, 41, 327. https://doi.org/10.1007/s10464-008-9165-0 Dzewaltowski, D. A., Noble, J. M., & Shaw, J. M. (1990). Physical activity participation: Social cognitive theory vs. the theories of reasoned action and planned behavior. Journal of Sport and Exercise Psychology, 12, 388–405. Engel, G. (1977). The need for a new medical model: A challenge for biomedicine. Science, 196, 129– 136. https://doi.org/10.1126/science.847460 Fairchild, A. J., & McDaniel, H. L. (2017). Best (but oftforgotten) practices: Mediation analysis. The American Journal of Clinical Nutrition, 105, 1259– 1271. https://doi.org/10.3945/ajcn.117.152546 Freedland, K. E. (2019). The Behavioral Medicine Research Council: Its origins, mission, and methods. Health Psychology, 38, 277–289. https:// doi.org/10.1037/hea0000731 Fried, E., Robinaugh, D., Quinn, K., & Tiokhin, L. (2019). Resource list: Formalizing theories in social sciences. https://osf.io/n9zta/ Gibbons, F. X., Gerrard, M., Blanton, H., & Russell, D. W. (1998). Reasoned action and social reaction: Willingness and intention as independent

696

MARTIN S. HAGGER, LINDA D. CAMERON, KYRA HAMILTON, NELLI HANKONEN, AND TARU LINTUNEN

predictors of health risk. Journal of Personality and Social Psychology, 74, 1164–1180. Gillison, F. B., Stathi, A., Reddy, P. et al. (2015). Processes of behavior change and weight loss in a theory-based weight loss intervention program: A test of the process model for lifestyle behavior change. International Journal of Behavioral Nutrition and Physical Activity, 12, 2. https://doi .org/10.1186/s12966-014-0160-6 Glasgow, R. E., Harden, S. M., Gaglio, B. et al. (2019). RE-AIM planning and evaluation framework: Adapting to new science and practice with a 20year review. Frontiers in Public Health, 7. https:// doi.org/10.3389/fpubh.2019.00064 Glasgow, R. E., Vogt, T. M., & Boles, S. M. (1999). Evaluating the public health impact of health promotion interventions: The RE-AIM framework. American Journal of Public Health, 89, 1322–1327. Hagger, M. S. (2009). Theoretical integration in health psychology: Unifying ideas and complimentary explanations. British Journal of Health Psychology, 14, 189–194. https://doi.org/10.1348/ 135910708X397034 Hagger, M. S. (2014). Avoiding the “déjà-variable” phenomenon: Social psychology needs more guides to constructs. Frontiers in Psychology, 5, 52. https://doi.org/10.3389/fpsyg.2014.00052 Hagger, M. S., & Chatzisarantis, N. L. D. (2014). An integrated behavior-change model for physical activity. Exercise and Sport Sciences Reviews, 42, 62–69. https://doi.org/10.1249/JES.000000000000 0008 Hagger, M. S., Moyers, S., McAnally, K., & McKinley, L. E. (2020). Known knowns and known unknowns on behavior change interventions and mechanisms of action. Health Psychology Review, 14, 199–212. https://doi.org/10.1080/ 17437199.2020.1719184 Hardeman, W., Michie, S., Fanshawe, T., Prevost, T., Mcloughlin, K., & Kinmonth, A. L. (2007). Fidelity of delivery of a physical activity intervention: Predictors and consequences. Psychology and Health, 23, 11–24. Hargreaves, T. (2011). Practice-ing behaviour change: Applying social practice theory to pro-environmental behaviour change. Journal of Consumer Culture, 11, 79–99. https://doi.org/10.1177/1469540510390500

Harkin, B., Webb, T. L., Chang, B. P. I. et al. (2016). Does monitoring goal progress promote goal attainment? A meta-analysis of the experimental evidence. Psychological Bulletin, 142, 198–229. https://doi.org/10.1037/bul0000025 Hawe, P., Shiell, A., & Riley, T. (2009). Theorising Interventions as events in systems. American Journal of Community Psychology, 43, 267–276. https://doi.org/10.1007/s10464-009-9229-9 Heber, E., Ebert, D. D., Lehr, D. et al. (2017). The benefit of web- and computer-based interventions for stress: A systematic review and meta-analysis. Journal of Medical Internet Research, 19, e32. https://doi.org/10.2196/jmir.5774 Heckhausen, H., & Gollwitzer, P. M. (1987). Thought contents and cognitive functioning in motivational and volitional states of mind. Motivation and Emotion, 11, 101–120. https://doi.org/10.1007/ BF00992338 Hedström, P., & Ylikoski, P. (2010). Causal mechanisms in the social sciences. Annual Review of Sociology, 36, 49–67. https://doi.org/10.1146/ annurev.soc.012809.102632 Heino, M. T. J., Noonan, C., Knittle, K., & Hankonen, N. (2019). Studying behaviour change mechanisms under complexity. Unpublished manuscript, University of Helsinki. Hennessy, E. A., Johnson, B. T., Acabchuk, R. L., McCloskey, K., & Stewart-James, J. (2020). Selfregulation mechanisms in health behaviour change: A systematic meta-review of metaanalyses, 2006–2017. Health Psychology Review, 14, 6–42. https://doi.org/10.1080/ 17437199.2019.1679654 Hofmann, W., & Kotabe, H. (2012). A general model of preventive and interventive self-control. Social and Personality Psychology Compass, 6, 707–722. https://doi.org/10.1111/j.1751-9004.2012.00461.x Jetten, J., Haslam, C., & Haslam, S. A. (Eds.). (2012). The Social Cure: Identity, Health and Well-Being. New York: Psychology Press. Johnson, B. T., Redding, C. A., DiClemente, R. J. et al. (2010). A network-individual-resource model for HIV prevention. AIDS and Behavior, 14, 204– 221. https://doi.org/10.1007/s10461-010-9803-z Kahneman, D. (2011). Thinking, Fast and Slow. New York: Farrar, Straus & Giroux.

The Science of Behavior Change: The Road Ahead

Kendall, P. C., & Hollon, S. D. (1979). Cognitivebehavioral interventions: Overview and current status. In P. C. Kendall & S. D. Hollon (Eds.), Cognitive-Behavioral Interventions: Theory, Research, and Procedures (pp. 1–9). New York: Academic Press. Kreuter, M. W., Lukwago, S. N., Bucholtz, D. C., Clark, E. M., & Sanders-Thompson, V. (2003). Achieving cultural appropriateness in health promotion programs: Targeted and tailored approaches. Health Education and Behavior, 30, 133–146. https://doi.org/10.1177/1090198102251021 Larsen, K. R., Michie, S., Hekler, E. B. et al. (2016). Behavior change interventions: The potential of ontologies for advancing science and practice. Journal of Behavioral Medicine, 40, 6–22. https:// doi.org/10.1007/s10865-016-9768-0 Leventhal, H., Meyer, D., & Nerenz, D. (1980). The common sense model of illness danger. In S. Rachman (Ed.), Medical Psychology, Vol. 2 (pp. 7–30). New York: Pergamon Press. Locke, E. A., & Latham, G. P. (2019). The development of goal setting theory: A half century retrospective. Motivation Science, 5, 93–105. https://doi.org/10.1037/mot0000127 Logan, J., & Graham, I. D. (2010). The Ottawa model of research use. In J. R. M. Bucknall (Ed.), Models and Frameworks for Implementing EvidenceBased Practice: Evidence to Action (pp. 83–108). Oxford: Wiley Blackwell. Lyons, A. C., & Chamberlain, K. (2017). Critical health psychology. In B. Gough (Ed.), The Palgrave Handbook of Critical Social Psychology (pp. 533– 555). London: Palgrave Macmillan. Marteau, T. M. (2017). Towards environmentally sustainable human behaviour: Targeting nonconscious and conscious processes for effective and acceptable policies. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 375, 20160371. https://doi.org/10.1098/ rsta.2016.0371 May, C., Finch, T., Mair, F. et al. (2007). Understanding the implementation of complex interventions in health care: The normalization process model. BMC Health Services Research, 7, 148. https:// doi.org/10.1186/1472-6963-7-148

697

McLeroy, K. R., Bibeau, D., Steckler, A., & Glanz, K. (1988). An ecological perspective on health promotion programs. Health Education Quarterly, 15, 351–377. https://doi.org/10.1177/10901981880 1500401 Meehl, P. E. (1990). Why summaries of research on psychological theories are often uninterpretable. Psychological Reports, 66, 195–244. https://doi .org/10.2466/pr0.1990.66.1.195 Michie, S., Aonghusa, P. M., Johnston, M. et al. (2019). The Human Behaviour Change Project. Website. www.humanbehaviourchange.org Michie, S., Carey, R. N., Johnston, M. et al. (2017). From theory-inspired to theory-based interventions: A protocol for developing and testing a methodology for linking behaviour change techniques to theoretical mechanisms of action. Annals of Behavioral Medicine, 52, 501– 512. https://doi.org/10.1007/s12160-016-9816-6 Michie, S., & Johnston, M. (2017). Optimising the value of the evidence generated in implementation science: The use of ontologies to address the challenges. Implementation Science, 12, 131. https://doi.org/10.1186/s13012-017-0660-2 Michie, S., Rothman, A. J., Kelly, M., & de Bruin, M. (2019). Theories and techniques of behaviour change project. https://www.ucl.ac.uk/pals/research/ clinical-educational-and-health-psychology/ research-groups/behaviour-change-techniques-and Michie, S., Rumsey, N., Fussell, A. et al. (2008). Improving Health: Changing behaviour – NHS Health Trainer Handbook. London: Department of Health. Michie, S., Thomas, J., Johnston, M. et al. (2017). The human behaviour-change project: Harnessing the power of artificial intelligence and machine learning for evidence synthesis and interpretation. Implementation Science, 12, 121. https://doi.org/ 10.1186/s13012-017-0641-5 Michie, S., van Straalen, M., & West, R. (2011). The behaviour change wheel: A new method for characterising and designing behaviour change interventions. Implementation Science, 6, 42. https://doi.org/10.1186/1748-5908-6-42 Michie, S., West, R., Campbell, R., Brown, J., & Gainforth, H. (2014). ABC of Behaviour Change Theories. London: Silverback Publishing.

698

MARTIN S. HAGGER, LINDA D. CAMERON, KYRA HAMILTON, NELLI HANKONEN, AND TARU LINTUNEN

Montaño, D. E., & Kasprzyk, D. (2015). Theory of reasoned action, theory of planned behavior, and the integrated behavioral model. In K. Glanz, B. K. Rimer & K. Viswanath (Eds.), Health Behavior and Health Education: Theory, Research, and Practice (5th ed., pp. 95–124). San Francisco, CA: Jossey-Bass. Moore, G. F., Cambon, L., Michie, S. et al. (2019). Population health intervention research: The place of theories. Trials, 20, 285. https://doi.org/ 10.1186/s13063-019-3383-7 Moore, G. F., Evans, R. E., Hawkins, J. et al. (2019). From complex social interventions to interventions in complex social systems: Future directions and unresolved questions for intervention development and evaluation. Evaluation, 25, 23–45. https://doi .org/10.1177/1356389018803219 Myers, H. F. (2009). Ethnicity- and socio-economic status-related stresses in context: An integrative review and conceptual model. Journal of Behavioral Medicine, 32, 9–19. https://doi.org/ 10.1007/s10865-008-9181-4 Naar, S., Czajkowski, S. M., & Spring, B. (2018). Innovative study designs and methods for optimizing and implementing behavioral interventions to improve health. Health Psychology, 37, 1081–1091. https://doi.org/ 10.1037/hea0000657 National Institutes of Health. (2019). Science of behavior change. Website. https://scienceofbeha viorchange.org/ Oettingen, G. (2012). Future thought and behaviour change. European Review of Social Psychology, 23, 1–63. https://doi.org/10.1080/10463283.2011 .643698 Presseau, J., Byrne-Davis, L. M. T., Hotham, S. et al. (2019). Following through on health behaviour change breakthroughs: The synergy between implementation science and health psychology. Unpublished manuscript, University of Ottawa, Canada. Prestwich, A., Sniehotta, F. F., Whittington, C., Dombrowski, S. U., Rogers, L., & Michie, S. (2014). Does theory influence the effectiveness of health behavior interventions? Meta-analysis. Health Psychology, 33, 465–474. https://doi.org/ 10.1037/a0032853

Renner, B., Hankonen, N., Ghisletta, P., & Absetz, P. (2012). Dynamic psychological and behavioral changes in the adoption and maintenance of exercise. Health Psychology, 31, 306–315. https:// doi.org/10.1037/a0025302 Resnicow, K., Baranowski, T., Ahluwalia, J. S., & Braithwaite, R. L. (1999). Cultural sensitivity in public health: Defined and demystified. Ethnicity and Disease, 9. Rhodes, R. E., Boudreau, F., Weman Josefsson, K., & Ivarsson, A. (2020). Mediators of physical activity behavior change interventions among adults: A systematic review and meta-analysis. Health Psychology Review. https://doi.org/10.1080/ 17437199.2019.1706614 Rhodes, R. E., McEwan, D., & Rebar, A. L. (2019). Theories of physical activity behaviour change: A history and synthesis of approaches. Psychology of Sport and Exercise, 42, 100–109. https://doi .org/10.1016/j.psychsport.2018.11.010 Rixon, L., Baron, J., McGale, N., Lorencatto, F., Francis, J., & Davies, A. (2016). Methods used to address fidelity of receipt in health intervention research: A citation analysis and systematic review. BMC Health Services Research, 16, 663. https://doi.org/10.1186/s12913-016-1904-6 Rogers, P. J. (2008). Using programme theory to evaluate complicated and complex aspects of interventions. Evaluation, 14, 29–48. https://doi .org/10.1177/1356389007084674 Rogers, R. W. (1975). A protection motivation theory of fear appeals and attitude change. Journal of Psychology, 91, 93–114. https://doi.org/10.1080/ 00223980.1975.9915803 Rosenstock, I. M. (1974). Historical origins of the health belief model. Health Education Monographs, 2, 328–335. Schwarzer, R. (2008). Modeling health behaviour change: How to predict and modify the adoption and maintenance of health behaviors. Applied Psychology: An International Review, 57, 1–29. https://doi.org/10.1111/j.1464-0597.2007.00325.x Sheeran, P., Klein, W. M. P., & Rothman, A. J. (2017). Health behavior change: Moving from observation to intervention. Annual Review of Psychology, 68, 573–600 https://doi.org/10.1146/ annurev-psych-010416-044007

The Science of Behavior Change: The Road Ahead

Skivington, K., Matthews, L., Craig, P. et al. (2018). MRC guidance: Developing and evaluating complex interventions. Updating Medical Research Council guidance to take account of new methodological and theoretical approaches. The Lancet, 392, S2. Meeting abstract: Public Health Science 2018, Belfast, Northern Ireland, November 23, 2018. http://dx.doi .org/10.1016/s0140-6736(18)32865-4 SOBC (Science of Behavior Change Research Network). (2019). Science of Behavior Change Research Network. Spotswood, F. (Ed.). (2016). Beyond Behaviour Change. Bristol: Bristol University Press. Strack, F., & Deutsch, R. (2004). Reflective and impulsive determinants of social behavior. Personality and Social Psychology Review, 8, 220– 247. https://doi.org/10.1207/s15327957pspr0803_1 Suls, J., Mogavero, J. N., Falzon, L., Pescatello, L. S., Hennessy, E. A., & Davidson, K. W. (2020). Health behaviour change in cardiovascular disease prevention and management: Meta-review of behavior change techniques to affect selfregulation. Health Psychology Review, 14, 43–65. https://doi.org/10.1080/17437199.2019.1691622 Tajfel, H., & Turner, J. C. (1986). The social identity theory of intergroup behaviour. In S. G. Worchel & W. Austin (Eds.), Psychology of Intergroup Relations (2nd ed., pp. 7–24). Chicago: Nelson-Hall. Thaler, R., & Sunstein, C. R. (2008). Nudge: Improving Decisions about Health, Wealth, and Happiness. New Haven, CT: Yale University Press. Tolin, D. F. (2010). Is cognitive–behavioral therapy more effective than other therapies?: A metaanalytic review. Clinical Psychology Review, 30, 710–720. https://doi.org/https://doi.org/10.1016/j .cpr.2010.05.003 Toomey, E., Hardeman, W., Hankonen, N. et al. (2019). Focusing on fidelity: Recommendations for improving intervention fidelity within trials of health behavioral interventions. https://doi.org/ 10.13140/RG.2.2.22329.11367 Trafimow, D. (2012). The role of auxiliary assumptions for the validity of manipulations and measures. Theory and Psychology, 22, 486–498. https://doi .org/10.1177/0959354311429996 Triandis, H. C. (1977). Interpersonal Behavior. Monterey, CA: Brookes/Cole.

699

Trout, J. D. (2004). The philosophical legacy of Meehl (1978): Confirmation theory, theory quality, and scientific epistemology. Applied and Preventive Psychology, 11, 73–76. https://doi.org/10.1016/j .appsy.2004.02.013 Turner, J. C., Hogg, M. A., Oakes, P. J., Reicher, S. D., & Wetherell, M. S. (1987). Rediscovering the Social Group: A Self-Categorization Theory. Cambridge, MA: Basil Blackwell. Wallot, S., & Kelty-Stephen, D. G. (2018). Interactiondominant causation in mind and brain, and its implication for questions of generalization and replication. Minds and Machines, 28, 353–374. https://doi.org/10.1007/s11023-017-9455-0 Walton, H., Spector, A., Tombor, I., & Michie, S. (2017). Measures of fidelity of delivery of, and engagement with, complex, face-to-face health behaviour change interventions: A systematic review of measure quality. British Journal of Health Psychology, 22, 872–903. https://doi.org/10.1111/bjhp.12260 Ward, P. (2011). A minimum price for alcohol? House of Commons Standard Note SN/HA/5021, London. Webb, T. L., Joseph, J., Yardley, L., & Michie, S. (2010). Using the internet to promote health behavior change: A systematic review and metaanalysis of the impact of theoretical basis, use of behavior change techniques, and mode of delivery on efficacy. Journal of Medical Internet Research, 12, e4. https://doi.org/10.2196/ jmir.1376 Weinstein, N. D. (1993). Testing four competing theories of health-protective behavior. Health Psychology, 12, 324–333. https://doi.org/10.1037/ 0278-6133.12.4.324 West, R., Godinho, C. A., Bohlen, L. C. et al. (2019). Development of a formal system for representing behaviour-change theories. Nature Human Behaviour, 3, 526–536. https://doi.org/10.1038/ s41562-019-0561-2 Wilson, T. E., Hennessy, E. A., Falzon, L., Boyd, R., Kronish, I. M., & Birk, J. L. (2020). Effectiveness of interventions targeting selfregulation to improve adherence to chronic disease medications: A meta-review of metaanalyses. Health Psychology Review, 14, 66– 85. https://doi.org/10.1080/ 17437199.2019.1706615

Index

Aarts, Henk, 6, 182, 602 Abraham, Charles, 4, 7, 91, 210, 269, 293, 345, 433, 656, 686 acceptability. See intervention evaluation action. See transtheoretical model action control. See health action process approach (HAPA) action planning. See health action process approach (HAPA) action self-efficacy. See health action process approach (HAPA) Adriaanse, Marieke A., 83, 85, 154, 171, 181, 186–187, 578–580, 588 affect-based interventions, 495–504 affective attitudes, 496, 497–501, 504 anticipated affect, 496, 497, 501–504 behavior change techniques, 497 overview and definitions, 495–497 affective attitudes. See attitudes and affect-based interventions Ajzen, Icek, 17, 18–20, 22, 24, 26, 35, 91, 151, 193, 210, 352, 451, 452, 453, 462, 580 Albarracín, Dolores,151, 448, 456, 468, 496 amotivation, 107, 512, 515 Araújo-Soares, Vera, 301, 311 Armitage, Christopher J., 208, 484, 562–564, 566–569, 573, 579 associative learning, 194, 450, 499, 600 attitudes, 210, 213, 243, 445, 625 affective attitudes, 216, 496–501 attitude change, 144, 169, 226, 229, 276, 445, 482 cognitive dissonance, 445, 447, 449, 454–455 definition and measurement, 446–447 determinant of intentions, 20, 91, 220, 240 ecological theory, 244 imagery interventions, 484, 490 implicit attitudes, 182, 215, 456 mere exposure effect, 455 persuasion,451–454, 497, 502 social judgment theory, 448 theories of attitude change, 447–449 theory of planned behavior, 20–22, 232, 274 Yale model of attitude change, 448

automatic processes. See nonconscious processes automaticity. See nonconscious processes, habit autonomous motivation, 107–110, 113, 115 as a mediator, 113–114 autonomy support, 108–112 autonomy support, 116, 218, 511–513, 514, 519 autonomy-support training program, 111, 113, 114, 513–518 behavior change techniques, 112–113, 514–515 dual-process model, 514 mechanism of action, 513–514 Bagozzi, Richard P., 131, 212, 220–221, 576 Bandura, Albert, 2, 5, 18, 32–35, 38, 49, 91–92, 141, 210, 238–239, 247, 275, 461–467, 469, 482, 580, 634–638, 653 Bargh, John A., 6, 168, 194 Barker, Mary, 272, 434, 436, 437, 438 Bartholomew Eldredge, L. Kay, 2, 4, 9, 247, 253–257, 270, 272, 301, 310, 313, 335 basic psychological needs, 104, 105, 110, 111, 113, 116, 510, 512 basic needs theory, 108 Baumeister, Roy F., 151, 154, 158, 539, 587–589 behavior change method. See behavior change techniques behavior change techniques, 3, 4, 272–273, 293–294, 308, 311–312, 361, 363, 392, 393, 394, 685, 688–690 affective interventions, 497, 501 associative leading and evaluative conditioning, 499 attitudes and persuasion, 449–450 autonomy support, 514 common-sense model, 61 dyadic interventions, 633–637, 639 experimental medicine approach, 287–289 goal setting, 557–560, 561 habit formation, 600–605 health action process approach, 94, 96 imagery, 482, 485–486 integrated theories, 219 integrative self-control theory, 154–156 monitoring interventions, 540, 547 motivational interviewing, 666–668

Index

protection motivation theory, 51 self-determination theory, 112–113 social cognitive theory and self-efficacy, 36–38, 463–467, 468, 472 social identity interventions, 655–656 behavior change wheel, 4, 22, 151, 157, 301 behavioral choice theory, 242, 245 behavioral contract, 558, 568, 636 behavioral determinant, 1, 2, 5, 63, 90, 92, 220, 238, 257, 294, 385, 386, 391, 393, 684 affective interventions, 496, 497 community theories, 258 culture, 394 ecological models, 239, 244, 684 goals, 35 habit, 189 imagery interventions, 483 integrated theories, 212 intention, 20, 21, 48, 215 modifiable determinant, 302, 309, 311–312 motivation, 104, 215 risk perception, 211 self-control, 153 self-efficacy interventions, 473 behavioral economics, 2, 5, 8, 272, 617–618, 622, 626 choice architecture. See nudge nudge. See nudge utility-maximizing behavior, 619, 621 behavioral epidemiology framework behavioral insights, 622, 627, 628 behavioral intention. See intention behavioral measure. See outcome measurement Berman, Anne H., 661 Biddle, Stuart J. H., 18, 107, 468 Borland, Ron, 55, 212, 276, 684 Branscombe, Nyla R., 229 Brewer, Noel T., 55, 292 Bronfenbrenner, Uri, 238, 402, 637, 684 Brown, Paul M., 626 Cameron, Linda D., 62, 64, 66, 67, 68, 70, 479, 480, 488 Carver, Charles S., 120, 275, 538, 613 Chamberlain, Kerry, 431, 432 change technique. See behavior change techniques Chatzisarantis, Nikos L. D., 18, 52, 107, 113, 114, 171, 208, 212, 215, 431, 451, 575 Cheon, Sung Hyeon, 114–115, 513, 514–515, 516–517, 519 choice architecture, 155, 196, 687, See also nudge overview and definition, 194–196 theory and mechanisms, 197–199

701

cognitive behavioral therapy (CBT), 38, 67, 353, 356, 394, 422, 667, 670, 672, 691 cognitive dissonance theory, 445, 447–448, 449 use in behavior change, 454–455 COM-B. See commitment-opportunity-motivation behavior model commitment-opportunity-motivation behavior model, 151, 157, 311, 433 common-sense model of self-regulation, 60, 352 action planning, 60, 62, 65, 68, 69 coherence, 62–63, 64, 65, 67 emotion regulation, 63, 64, 65, 66–67 illness representations, 62, 64–66, 69, 70 overview and description, 61–64 use in behavior change, 64–69 community context, 401 community social norms, 401–403 cultural norms, 238, 403 ecological perspectives, 401–403 organizational change, 404, 407, 408, 410 social policy and community change, 408–409 complex interventions, 7, 319, 334, 350, 470, 627, 654 MRC framework for complex interventions, 7, 301, 306, 319, 324, 349–350, 352, 353, 356 complex systems, 247, 315, 691 overview and definition, 306 conditioning, 9, 201, 600 Conner, Mark, 217–218 Conner, Mark T., 84, 208, 212, 220–221, 496, 497–500, 501, 504, 573, 684 Conroy, David E., 212, 213 Conroy, Dominic, 68, 479–480, 484–486 consolidated framework for implementation research, 336, 338–339, 686 Consolidated Health Economic Evaluation Reporting Standards (CHEERS) checklist, 373, 379–380 contemplation. See transtheoretical model control theory, 60, 120, 275, 538–540, 547, 560, 562 awareness, 127, 129–130, 131 behavioral illusion, 129, 194 conflict, 125, 126, 127, 128, 129, 131 controlled variable, 121, 125, 129 feedback function, 121, 125, 130 input function, 128 learned and unlearned mechanisms, 125–126 negative feedback loop, 121–122, 125 output function, 126 overview and description, 121–127 perceptual function, 125, 126 perceptual input, 122 perceptual signal, 122, 128, 130 reference value, 121–123, 125, 128, 537–539, 547

702

Index

control theory (cont.) reorganization, 125, 126–127, 128, 130 standards, 125–126 use in behavior change, 127–131 controlled motivation, 107, 116 coping planning. See health action process approach (HAPA) coping self-efficacy. See health action process approach (HAPA) counterconditioning, 139, 143 critical perspectives on behavior change, 430–434, 436–439 overview and description, 431–432, 436–438 social practice theory, 431, 435–436 Crutzen, Rik, 256, 275, 345 de Bruijn, Gert-Jan, 6, 215, 572–573, 575–576, 577 de Bruin, Marijn, 538, 547, 548 de Ridder, Denise T. D., 171, 587–588, 589–590, 593, 594, 625 Deci, Edward L., 104–110, 111, 511, 512–513, 608 Deutsch, Roland, 6, 164, 168, 169, 171, 276 DiClemente, Carlo C., 6, 17, 136–139, 142, 144–145, 462, 580 diffusion of innovations theory, 253, 254, 405 digital interventions, 353, 366, 368, 369, 417 acceptability and engagement, 419–421 efficacy, 421–422 eHealth, 99, 354, 416 gamification, 420, 502 implications for research and practice, 424–425 participatory codesign, 422–424 quality assessment, 424–425 self-guided interventions, 416, 419, 420, 422, 424 Dombrowski, Stephan U., 311, 547, 633, 656 dual-phase models, 77–78 motivational phase, 78, 215 volitional phase, 35, 78, 215, 220 dual-process models, 173, 213, 276, 449, 452, 512 critique, 168, 627 heuristic processing, 213, 448, 621 overview and definition, 6–7, 276 system 1 and 2 processes, 621 dual-process theories. See dual-process models dual-system models. See dual-process models dyadic interventions, 632, 638 collaborative planning, 644–645 couple-based dyadic interventions, 638 dyadic action control, 643–644 dyadic behavior change techniques, 632, 633–637, 639 dyadic constellations, 632, 638, 640

dyadic peer-based interventions, 638, 639 dyadic planning, 639, 640, 642 mechanisms of change, 638 parent-child dyads, 638, 640, 642 romantic couples, 632, 643 social support, 634–638, 641–642 ecological approach. See ecological models ecological models, 237, 387, 389, 403, 408 behavior change techniques, 239–241, 242 coalition frameworks, 253, 255 community level, 253, 255, 386, 404, 406 empowerment theories, 242, 254, 407 environmental level, 252–253, 257 interpersonal level, 244, 254 organizational level, 253, 254, 257, 260–261, 409 overview and definition, 237–239, 252–253 societal level, 7, 253, 256, 259, 273, 389 stakeholder frameworks, 253, 255 theories of behavior change, 241–242 ecological systems theory. See systems theory ecological theory. See ecological models economic evaluation, 7, 279, 280, 319, 320, 321, 326, 327, 658 cost-effectiveness, 419, 625 cost-effectiveness analysis, 373–379 cost-effectiveness evaluation, 7, 270, 312, 328, 350, 368, 373 economic evaluation framework, 373–376 elaboration likelihood model, 213, 448–449, 451 environmental influence, 37, 261, 621 environment-level theories. See ecological models Epton, Tracy, 289, 562–564, 566–569 equity in diversity and disparities, 387, 394 in intervention delivery, 311 evaluation design, 319, 320, 321–326, 328 evaluative conditioning, 155, 170, 456, 496, 497 mechanisms, 499 overview and definition, 165 expected utility theory, 617, 621, 623, 627 affect heuristic, 624 anchoring, 624 availability heuristic, 624 commitment consistency principle, 624 decision fatigue, 624 loss and regret aversion, 624 optimism and overconfidence, 624 overview and description, 618–619 present bias, 624 social norms, 624

Index

status quo bias, 624 experiential attitudes. See attitudes experimental design. See intervention design experimental medicine approach, 286–289, 294–296 use in behavior change, 289–291 extrinsic motivation, 107, 608 fear appeals, 9, 60, 62, 64, 66, 67, 69, 91, 275, 496, 497, 502 mechanisms, 54–55 fear arousal. See fear appeals feasibility study, 300, 301, 312, 314, 320, 353–356 Festinger, Leon, 2, 447, 454, 663 financial incentives. See incentives, monetary incentives Fishbach, Ayelet, 151, 155–156, 539 Fishbein, Martin, 17, 18–20, 24, 26, 35, 91, 193, 210, 274, 391, 462 Fisher, Jeffrey D., 19, 575 Fisher, William A., 19, 575 Fletcher, Paul C., 193, 627 French, David P., 466–468, 473–474 Friese, Malte, 587, 589, 592, 594 functional imagery training. See imagery interventions Gardner, Benjamin, 187, 291, 599–607, 611–613 Gergen, Kenneth J., 431 Gergen, Kenneth, J., 438 Gibbons, Frederick X., 212, 292 Gillebaart, Marleen, 587–588, 589–590, 594 Glasgow, Russell E., 278, 341–342, 408–409, 686 Gneezy, Uri, 523 goal intention. See intention goal pursuit, 35, 79, 90–91, 94, 182, 519, 538–540, 562, 594, See also model of action phases goal setting, 35, 691 goal setting interventions, 566–569 goal setting theory, 90, 538, 555 overview and definitions, 556–557 SMART goals, 555, 568–569 techniques and mechanisms, 557–562 goal systems theory, 154, 560, 562 Godin, Gaston, 83, 151, 501 Gollwitzer, Peter M., 6, 47, 77–78, 79, 80, 82, 83, 186, 277, 547, 575, 579, 602 graphic threat messages. See fear appeals group identification. See social identity approach group processes. See social identity approach guided imagery. See imagery interventions habit, 437 automaticity, 182–183, 611, 613 cue-response association, 179, 181, 182, 184, 187, 604

703

definition, 599–600 discontinuity, 180, 184–186, 188, 600–602, 604, 608, 609, 611 formation, 183, 184, 188, 525, 527, 533, 576, 578, 590, 600–605, 609, 610, 613 inhibition, 602, 608, 609 measurement, 182, 183, 187–188, 189, 611, 613 past behavior, 183 substitution, 600, 602, 605, 609, 612 habit theory, 179, 187, 599 mechanisms, 600–601 overview and definition, 599–600 overview and description, 179–183 use in behavior change, 183–187 Hagger, Martin S., 18, 52, 68, 93–94, 96, 107, 113–114, 171, 208–218, 431, 451, 484–490, 573–578, 580, 667, 684 Hamilton, Kyra, 92, 93, 94, 96, 212, 213, 215, 451, 452–453 Hankonen, Nelli, 113, 301, 306, 310, 391, 689, 691 Hardeman, Wendy, 389 Hargreaves, Tom, 436, 543, 549 Haslam, Catherine, 649–651, 655, 656 Haslam, S. Alexander, 227, 231, 649–651, 652, 653, 655, 656 health action process approach (HAPA), 89, 215, 216, 462, 575, 580 action control, 94, 97–98 action planning, 92–94, 96, 489, 575 action self-efficacy, 92, 94 coping planning, 92–94, 96, 575, 580 coping self-efficacy, 92 motivational phase, 90, 96 outcome expectancies, 91, 94, 96, 98 overview and description, 90–94 phase-specific self-efficacy, 91–92, 96 recovery self-efficacy, 92, 94–95 risk perceptions, 91, 94, 96, 98 use in behavior change, 96–97 volitional phase, 90, 91, 97, 98 health belief model, 47, 209, 462, 580 cues to action, 47 overview and description, 47–48 perceived barriers, 47, 51, 53, 55 perceived benefits, 47, 49, 51, 53, 55 use in behavior change, 53–56 Heckhausen, Heinz, 6, 77–78 Hofmann, Wilhelm, 151, 153, 154, 158, 170, 499, 587, 588 Hollands, Gareth J., 193, 195, 499, 627 Houben, Katrijn, 155–156, 170, 588

704

Index

if-then plans. See implementation intentions imagery interventions, 8, 38, 64, 68, 127, 463, 464, 480–482, 691 behavior change techniques, 486 functional imagery training, 418, 480–482 guided imagery, 480–481, 487, 488, 489–491 mechanisms, 482–484 mental simulation, 481, 482, 484, 487, 489, 490 overview and description, 480–482 implementation determinants, 336, 338–339, 340, 341, 344 implementation intentions, 39, 78, 132, 168, 215, 277, 489, 496, 575, 602 affective interventions, 505 collaborative implementation intentions, 642 habit formation, 186–187 overview and definition, 80, 577–578 use in behavior change, 82–84, 577–579 implementation science, 2, 5, 157, 334–335, 350–352, 685, 693 implicit attitudes. See attitudes implicit processes. See nonconscious processes incentives, 9, 157, 340, 523, 622, 687 breaking habits, 528–529 mechanisms, 524–533 micro-incentives, 625 monetary incentives, 181, 287, 525, 526, 527, 530, 609 overview and definition, 524, 619–620 social incentives, 524 switching costs, 531–533 upfront incentives, 529–531 information campaign, 619, 687 ingroup norms. See social identity approach, group norms integrated behavior change model, 215, 216, 217, 575 integrated behavioral model, 213, 216 integrated theories, 208, 576 overview and definition, 209 use in behavior change, 215–218 integrative self-control theory, 6, 151, 153, 154, 155, 156, 157 desire, 152–153, 154, 155, 157 overview and description, 152–154 self-control capacity, 152, 153, 154, 155, 156, 157 self-control conflict, 152, 153, 154, 155, 156, 158 self-control goal, 152, 153, 154, 156, 158 self-control motivation, 152, 153, 154, 155, 157 use in behavior change, 154–157 intention, 20, 35, 48, 82, 90, 105, 137, 152, 167, 181, 213, 232, 240, 286, 369, 420, 437, 487, 555, 573, 588, 600, 625, 643, 663 change, 452, 484, 500, 502, 504 definition, 19–20, 90–91, 575 determinants, 20–21, 48, 455, 462

dual-phase models, 215, 220, 575 dual-process models, 213, 220 formation, 18, 36, 38, 92, 94, 96, 215, 575 habit theory, 181 integrated theories, 210 integrative self-control theory, 152 intention change, 18, 26, 292, 451 intention-behavior gap, 26, 92, 94, 215, 216, 538, 572–573, 575, 663 intention-behavior relationship, 6, 18, 90, 91, 93, 95 major theorists’ model, 213 measurement, 23–24 reflective-impulsive model, 167, 168 self-determination theory, 114 social cognitive theory, 35 social identity processes, 232 stability, 340 theory of planned behavior, 19–20, 193 transtheoretical model, 137, 140 internal cost-benefit calculations, 621, 627 intervention design experimental designs, 24, 115, 321–324, 395, 410, 580 nonexperimental designs, 115, 321, 324–325 quasi-experimental designs, 24, 26, 321, 323–324, 410 randomized controlled trial, 290, 301, 321–324, 394, 396, 422, 469 intervention development frameworks, 306, 314 intervention effectiveness. See intervention evaluation intervention efficacy. See intervention evaluation intervention evaluation acceptability, 279, 300, 308, 310, 311, 312, 314, 320, 324, 334, 336, 342, 344, 394, 613 effectiveness, 2, 4, 8, 10, 25, 39, 46, 70, 83, 110, 120, 170, 172, 173, 189, 195, 238, 269, 270, 271, 277, 278, 279, 290, 311, 318, 319–320, 328, 345, 350, 356, 358, 361, 392, 394, 622, 691, 694 efficacy, 2, 4, 10, 70, 318, 319–320, 321, 323, 333, 335, 342, 343, 344, 345, 356, 361, 394, 396, 417, 421, 424, 454, 491, 510, 530, 533, 563, 622, 686, 688, 691, 694 realist, 319–320 systems, 319–320 intervention fidelity, 4, 8, 270, 278, 279, 308, 312, 314, 320, 321, 325, 334, 343, 420, 423, 473, 480, 486, 577 affective interventions, 500, 502 autonomy-support interventions, 517 dyadic interventions, 642 goals-setting interventions, 568 habit interventions, 611, 612 imagery interventions, 490, 491 monitoring interventions, 550 motivational interviewing, 666, 671 self-control interventions, 593–594

Index

intervention implementation, 300, 341–345, 346, 350 diffusion, 334, 667 dissemination, 287, 335, 336, 338, 339, 341, 344, 366, 393, 667 hybrid trial, 335 implementation trial, 335 intervention mapping, 4, 9, 270–271, 272, 307, 309, 313, 473, 541, 557, 633 intervention development, 301 overview and definition, 253–256 intervention prototypes, 270, 278 integrated theories approaches to integration, 209–211 intrinsic motivation, 66, 105–107, 527–528, 532, 608 cognitive evaluation theory, 105–107 organismic integration theory, 107 Johnson, Blair T., 151, 448, 449, 451–452, 454, 684, 689 Johnston, Marie, 48, 293, 433, 688–690 Kahneman, Daniel, 6, 621 Kanfer, Frederick H., 538 Kavanagh, David J., 152, 154, 417–420, 483 Kelly, Michael P., 272, 433, 434, 436, 437, 438, 627 Klein, William M. P., 2, 286, 685 Knittle, Keegan, 116, 294, 306, 470, 566 Knoll, Nina, 93, 94, 96, 472, 634–637, 640–641 Kok, Gerjo, 3, 91, 252, 253–257, 432 Kruglanski, Arie W., 154, 168–169, 448, 560 Kwasnicka, Dominika, 93, 291, 314, 323, 580–581 Lally, Phillippa, 187, 434, 599–602 Latham, Gary P., 463, 538, 555–557, 560–566, 573, 613 Leventhal, Howard, 6, 60–62, 66, 68, 352, 393, 575 Lewin, Kurt, 77, 78, 153, 185, 238, 401–402, 538, 575 libertarian paternalism, 618, 620, 627 Locke, Edwin A., 469, 538, 555–557, 560–566, 573, 613 logic model, 4, 216, 218, 257, 277, 291, 301, 326, 490, 689 dark logic model, 308, 319 intervention development, 306, 311, 313 overview and definition, 272–275, 309–310 Luszczynska, Aleksandra, 39, 40, 41, 54, 93, 95, 99, 343, 345, 575, 577, 578, 580 Lyons, Antonia C., 431, 432 maintenance. See transtheoretical model major theorists’ model, 210, 212 Mansell, Warren, 123, 130–131 Marteau, Theresa M., 6, 202, 627 Matthews, Linsay, 304, 325

705

mechanism of action, 290, 291–292, 449, 468, 500, 501 overview and definition, 287–289, 688–690 Meier, Stephan, 527–528 mental contrasting, 85, 482, 691 mental imagery. See imagery interventions mental simulations. See imagery interventions mere exposure effect, 200 message framing, 624 mHealth. See digital interventions Michie, Susan, 2–4, 22, 151, 157, 209, 210, 289, 292–294, 301, 311, 340, 432, 433, 684–685, 688–690, 693 mixed methods, 325, 410 mode of delivery, 280, 302, 305, 312, 313, 314, 320, 549, 566, 577, 633, 656, 667, 669 model of action phases, 77, 215, 575, 578 deliberative and implemental mindsets, 79–82 goal striving and goal pursuit, 77, 79, 80, 82 implementation intentions. See implementation intentions overview and description, 79–80 use in behavior change, 80–83 model-based/model free framework, 194, 197, 202 applications, 199–201 overview and definition, 197–199 monitoring interventions behavior change mechanisms, 537–538 behavior change techniques, 547 definitions, 537–538 Moore, Graham F., 279, 325, 349, 684, 690 Moss-Morris, Rona, 352–353, 356 motivating style. See autonomy support motivational interviewing, 9, 64, 66, 70, 142, 156, 418, 423, 547, 641, 661–673 behavior change techniques, 666–668 mechanisms of change, 663–665 overview and definitions, 662 multiphase optimization strategy (MOST), 289, 295, 314, 368 need frustration, 512, 513–514 need support, 105, 110–112, 114, 116 need thwarting. See need frustration needs assessment. See user engagement neoclassical economic approaches, 375–379, 380–381, 617–618, 626–628 Nilsen, Per, 260, 334–336, 342, 345 Noar, Seth M., 6, 66, 209, 210, 216, 467 nonconscious processes, 6, 9, 41, 193, 562, 578, 600, 604 dual-process models, 194–195 integrated models, 213–215 nonexperimental designs. See intervention design Ntoumanis, Nikos, 513–516

706

Index

nudge behavioral choice, 157, 621 choice architecture, 153, 154, 156, 187, 193, 201, 625 cognitive bias, 623, 627 commitment techniques, 625 default options, 625 framing, 624 heuristics, 623, 624, 627 information salience, 624 micro-incentives, 625 overview and definition, 620 social incentives, 625 theory and mechanisms, 621–622 Ogden, Jane, 432–433, 434, 438 open science, 201, 642 optimization of interventions, 289, 295, 301, 303, 314, 365–368, 687 Orbell, Sheina, 6, 48, 52, 56, 179–182, 183, 208 Ottawa model for research use, 336–338, 341, 686 outcome evaluation, 279, 319, 423, 451 outcome expectancies. See self-efficacy outcome measurement, 2, 25, 329, 344, 473, 549 behavior, 327–328 objective behavior measure, 54, 84, 327, 328, 490, 500, 540, 611, 642 validity, 24, 51, 54, 292, 293, 672 Owen, Neville, 7, 237 past behavior, 141, 166, 183, 213, 463 perceived behavioral control determinant of intentions, 20 perceptual control theory. See control theory person-based approach. See user engagement persuasive communication, 22, 27, 66, 232, 243, 273, 363, 445, 448, 451–454, 497, 501, 502, 634 Perugini, Marco, 56, 212, 220–221 phase-specific self-efficacy. See health action process approach (HAPA) pilot study, 301, 320, 345 population, intervention, comparator, and outcome (PICO) framework, 373–375 Powers, William T., 120–121, 123, 127, 129, 538 precontemplation. See transtheoretical model preparation. See transtheoretical model Presseau, Justin, 92–93, 611, 686 Prestwich, Andrew, 2–3, 40, 217–218, 644–645 process evaluation, 279, 304, 319, 321, 325, 327, 653, 658 implementation, 279, 341–342 logic model, 273 mechanism of change, 113, 114, 116, 685, 688–690

processes of change. See transtheoretical model Prochaska, James O., 136–139, 140, 145, 462, 580 Proctor, Enola K., 333, 336, 342–343, 345 program theory, 301, 308, 319, 320, 321, 689 intervention development, 313 intervention evaluation, 326–329 overview and definition, 309–310 protection motivation theory, 47, 209, 210, 213, 218, 462 coping appraisal, 48, 50, 51, 52 overview and description, 48–49 perceived severity, 48, 50, 91 perceived susceptibility, 48, 50, 91 response costs, 49 response efficacy, 49 threat appraisal, 48, 50 use in behavior change, 53–56 prototype willingness model, 213–215, 216 prototype favorability, 215, 216 prototype similarity, 215, 216 qualitative methods, 311, 320, 324–325, 352, 362, 365–368, 395, 410, 434–439 quantitative methods, 8, 311, 320, 325, 352, 362, 368 quasi-experimental design. See intervention design randomized controlled trial. See intervention design RE-AIM framework, 278, 341–342, 686 reasoned action approach. See theory of planned behavior Rebar, Amanda L., 96, 187, 213, 601, 602–604, 684 recovery self-efficacy. See health action process approach (HAPA) Reeve, Johnmarshall, 111–113, 114–115, 511–512, 513, 514–515, 516–517, 519 reflective-impulsive model, 164, 165 critique, 169 dual-process models, 168, 173 impulsive system, 165–167, 168, 170, 171 overview and description, 165–168 reflective system, 167–168, 170 use in behavior change, 169–173 regulation ban, 1, 182 restriction, 22, 127, 620, 687 Reynolds, Katherine J., 229–230, 232 Rhodes, Ryan E., 6, 41, 215, 495, 496–497, 500, 503, 504, 567, 572–573, 575–576, 580, 684 risk perception, 9, 40, 90, 98, 211, 388 Rogers, Carl R., Rogers, R. W., 6, 47, 462 Rosenstock, Irwin M., 6, 47, 462, 580 Rothman, Alexander J., 2, 286, 291, 454, 685

Index

Ruiter, Robert A. C., 54, 91, 252, 256, 541 Rutter, Harry, 247, 335 Ryan, Richard M., 6, 104–110, 111, 511, 512–513, 608 Sallis, James F., 7, 237, 240, 408 Salmon, Jo, 245–246 Scheier, Michael, 120, 275, 538, 613 Schmidt, Peter, 18, 21, 451, 671 Scholz, Urte, 580, 634–637, 643–645 Schüz, Benjamin, 8, 61, 391–392, 580 Schwarzer, Ralf, 6, 40, 41, 92, 93, 94, 95, 99, 461, 472, 573, 575, 580 self-control, 150–151, 153–158, 165, 169, 420, 433, 538, 539, 586–587, 600 behavior change techniques, 591–595 definition, 587 mechanism of change, 589 self-control and inhibitory control training, 587, 589–595 self-determination theory, 6, 132, 217, 242, 510 interventions, 113–116 motivation and behavior change techniques, 112–113 overview and description, 105–108, 512–513 use in behavior change, 110–113 self-efficacy, 20, 33, 49, 90, 274, 275 behavior change techniques, 36–38, 275, 462–466 control theory, 123 definition, 33–34 definitions, 461–462 dyadic interventions, 634–637 ecological models, 243, 245 health action process approach (HAPA), 91–92, 94–96, 97–98 health belief model, 51–52 imagery interventions, 482 integrated theories, 210, 212, 215 mastery experiences, 463–464 measurement, 41, 473–474 outcome expectancies, 34–35, 90 protection motivation theory, 48–49, 54–55 reflective-impulsive model, 168 social cognitive theory, 33–34, 36 social identity interventions, 653 somatic and affective states, 463–467 sources of self-efficacy, 34, 462–467, 468–470 transtheoretical model, 140–141, 144–145 verbal persuasion, 463–466 vicarious experiences, 463–465 self-monitoring, 340, 417, 548 behavior change techniques, 540 common-sense model, 68–69 ecological models, 243, 246

707

health action process approach (HAPA), 94, 96, 97 integrative self-control theory, 156 mastery experiences, 464, 468 overview and description, 538 sequential multiple assignment randomized trial (SMART), 368 Sheeran, Paschal, 2, 4, 7, 18, 26, 47, 78, 80, 83, 91, 94, 210, 277, 286, 289, 291, 292, 496, 500, 502, 504, 505, 547, 573, 575, 578, 685 Simpson, Sharon A., 304, 325 situationism, 194 Sniehotta, Falko F., 93, 301, 575, 578, 580, 611 social agents, 107, 110–111, 113, 114–116, 252, 310, 511 social cognition models, 5–7, 105, 193, 208, 210, 211–215, 219, 438, 575, 684 social cognitive theory, 32, 209, 210, 212, 239, 243, 245, 246, 254, 461, 469, 474, 482, 484, 580 distal goals, 35 impediments (barriers), 33, 35, 36 opportunities (facilitators), 33, 35, 36 outcome expectancies, 34–35 overview and description, 33–36 proximal goals, 35, 38, 97, 481 reciprocal determinism, 33, 238, 637 self-efficacy, 33–34 use in behavior change, 36–38, 462–467 social context, 2, 32, 213, 404, 408, 432 social disparities, 385, 392 barriers to intervention effectiveness, 392–393 behavioral inequities, 389–391 culturally-specific behavior change, 393–396 definition and description, 386–387 frameworks of health inequities, 387–389 minority stress model, 389 social identification, 200, 225, 226, 229, 230, 650, 653 definition, 229 mechanisms, 232, 656 social identity approach, 226, 228, 649, 684 group identification, 227, 232 group norms, 225–226, 231, 233, 405, 650, 656 overview and description, 226–228, 650–652 use in behavior change, 229–232, 652–655 social identity model of behavior change, 649, 653, 656 behavior change techniques, 656 overview and description, 650–653 social support, 650, 653 social identity theory. See social identity approach social learning theory. See social cognitive theory social network, 233, 245, 389, 406, 435, 632, 633 social network analysis, 227, 232 social network theories, 253, 254

708

Index

social norms, 21, 213, 227, 232, 253, 255, 257, 286, 401, 403–405, 510, 620, 623, 624 social policy and community change, 403 social support, 36, 37, 95, 107, 138, 245, 276, 368, 420, 424, 472, 514, 633, 640, 643 theories of social support, 253, 254 socioecological approach. See ecological models stages of change. See transtheoretical model stakeholder engagement, 350, 352, 353, 356, 369 normalization process theory, 350–352, 353–356, 358 overview and description, 350–352 stakeholder involvement, 7, 8, 261, 301, 305–306, 307, 309, 310, 312, 313, 320, 325, 326, 327, 328, 340, 344, 345, 350, 351, 370, 392 intervention development stage, 353–356 intervention evaluation and implementation, 358 methods, 352, 353 overview and definition, 351 stigma, 146, 229, 363, 417 theories of stigma and discrimination, 253, 257 Stokols, Daniel, 239, 402 Strack, Fritz, 6, 164, 168, 169, 276 subjective norm determinant of intentions, 20 Sunstein, Cass R., 6, 125, 151, 153, 154, 156, 193, 194, 270, 604, 618, 620, 621, 628 systems theory, 219, 239, 253, 254, 684 tailoring interventions, 144, 145, 310, 344, 362, 364–365, 424 Tajfel, Henri, 7, 226, 227, 649, 684 Tarrant, Mark, 272, 653, 655, 656, 658 tax subsidy, 619, 620, 625 taxonomy behavior change technique, 3, 157, 219, 294, 296, 454, 463–467, 480, 485–486, 497, 499–501, 540, 557–560, 573–575, 600–605, 622–623, 633–637, 655–656 implementation, 335–336, 339, 342–343 Taylor, Shelley E., 68, 79, 81, 479–481, 484, 487, 489–490, 563 Thaler, Richard H., 6, 125, 151, 153, 154, 156, 193, 194, 270, 604, 618, 620, 621, 628 theoretical domains framework, 4, 311, 340–341 theories of power, 254, 258 theory of planned behavior, 36, 210, 212, 215, 217, 232, 352, 391, 451, 462 behavioral beliefs, 20, 23–25, 451–454 control beliefs, 20, 23–25 intention, 19–20 normative beliefs, 20, 23–25, 275 overview and description, 18–21 perceived behavioral control, 20–22

salient beliefs and belief elicitation, 23–25 subjective norm, 20–22, 24–25 use in behavior change, 21–25 theory of reasoned action. See theory of planned behavior think-aloud studies, 366 TIPPME, 195–197 availability, 195, 196 functionality, 195 information, 195, 196–197 presentation, 195 size, 195, 196 Trafimow, David, 6, 22, 209, 226, 684 transtheoretical model, 17–18, 40, 136, 462, 580 action, 17, 137, 139, 141, 145 contemplation, 17, 137 context of change, 140, 345 maintenance, 17, 137, 139 markers of change, 140–141 overview and description, 137–141 precontemplation, 17, 137, 140 preparation, 17, 137, 139, 140 processes of change, 136, 140 self-change, 140, 141, 146 stages of change, 17, 136, 140, 141–145 use in behavior change, 137 Triandis, Harry C., 19, 179, 213 Trickett, Edison J., 410 Turner, John C., 649, 684 Tversky, Amos, 618, 621 user engagement, 361, 364, 392 needs assessment, 362–363, 364, 395, 405 nonjudgmental language, 354, 365 person-based approach, 352, 353, 362–363, 366, 369 tailoring, 364–365 Verplanken, Bas, 6, 179–182, 183, 185–186, 602 visualization. See imagery interventions Vohs, Kathleen D., 154, 158, 539, 588, 589 Warner, Lisa M., 34, 461, 474 Webb Hooper, Monica, 385, 387, 392, 393–394 Webb, Thomas L., 26, 83, 91, 94, 292, 505, 537, 547–549, 573, 692 Weinstein, Neil D., 51, 273, 292, 684 West, Robert, 2, 22, 151, 157, 301, 311, 433, 684–685 Williams, David M., 41, 495, 496 Williams, Geoffrey C., 113, 517 Wood, Wendy, 6, 178–181, 194, 578 Yardley, Lucy, 354, 361–363, 365, 368