The Neurotic Paradox, Vol 2: Progress in Understanding and Treating Anxiety and Related Disorders, Volume 2 (World Library of Mental Health) [Volume 2, 1 ed.] 1138659819, 9781138659810

This collection of David H. Barlow‘s key papers are a testimony to the collaborative research that he engendered and dir

303 82 11MB

English Pages 282 [301] Year 2016

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

The Neurotic Paradox, Vol 2: Progress in Understanding and Treating Anxiety and Related Disorders, Volume 2 (World Library of Mental Health) [Volume 2, 1 ed.]
 1138659819, 9781138659810

Citation preview

“What a wonderful overview of a program of research that carries with it much of the history of behavior therapy together with the development of an outstanding career. This volume makes for interesting reading for anyone at any level with an interest in clinical psychology and the development of behavior therapy. Moreover, there is a wide range of subject matter and of older papers that still merit reading.” —W. Stewart Agras, MD, Professor of Psychiatry Emeritus, Stanford University “Assembled in one place, here, is a collection of papers by Barlow and colleagues outlining the next paradigm in the psychological science of anxiety disorders. The Neurotic Paradox reflects one important way out of the proliferation of treatment manuals that plagues the dissemination of evidence based practice in clinical psychology. The impact of this work will be as profound as it will be broad based.” —Terry M. Keane, PhD, Professor of Psychiatry & Psychology, Boston University School of Medicine

The Neurotic Paradox

This collection of David H. Barlow’s key papers are a testimony to the collaborative research that he engendered and directed with associates who now stand with him at the forefront of experimental psychopathology research and in the treatment of anxiety and related disorders. His research on the nature of anxiety and mood disorders resulted in new conceptualizations of etiology and classification. This research led to new treatments for anxiety and related emotional disorders, most notably a new transdiagnostic psychological approach that has been positively evaluated and widely accepted. Clinical psychology will benefit from this collection of papers with its connecting commentary. David H. Barlow, PhD, is Professor and Founder of the Center for Anxiety and Related Disorders at Boston University. He has received numerous awards and has published over 600 articles and chapters and over 75 books; his research has been continuously funded by the National Institutes of Health for over 40 years.

Routledge World Library of Mental Health

The World Library of Mental Health celebrates the important contributions to mental health made by leading experts in their individual fields. Each author has compiled a career-long collection of what they consider to be their finest pieces: extracts from books, journals, articles, major theoretical and practical contributions, and salient research findings. For the first time ever the work of each contributor is presented in a single volume so readers can follow the themes and progress of their work and identify the contributions made to, and the development of, the fields themselves. Each book in the series features a specially written introduction by the contributor giving an overview of his career, contextualizing his selection within the development of the field, and showing how his own thinking developed over time.

Rationality and Pluralism—The selected works of Windy Dryden By Windy Dryden The Price of Love—The selected works of Colin Murray Parkes By Colin Murray Parkes Attachments: Psychiatry, Psychotherapy, Psychoanalysis—The selected works of Jeremy Holmes By Jeremy Holmes Passions, Persons, Psychotherapy, Politics—The selected works of Andrew Samuels By Andrew Samuels Towards a Radical Redefinition of Psychology—The selected works of Miller Mair Edited by David Winter and Nick Reed Living Archetypes—The selected works of Anthony Stevens By Anthony Stevens Soul: Treatment and Recovery—The selected works of Murray Stein By Murray Stein A Developmentalist’s Approach to Research, Theory, and Therapy—The Selected Works of Joseph Lichtenberg By Joseph D. Lichtenberg

The Neurotic Paradox

Progress in Understanding and Treating Anxiety and Related Disorders Volume 2

Edited by David H. Barlow

First published 2016 by Routledge 711 Third Avenue, New York, NY 10017 and by Routledge 2 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN Routledge is an imprint of the Taylor & Francis Group, an informa business © 2016 Taylor & Francis The right of the editor to be identified as the author of the editorial material, and of the authors for their individual chapters, has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. Library of Congress Cataloging-in-Publication Data A catalog record for this book has been requested. ISBN: 978-1-138-65981-0 (hbk) ISBN: 978-1-315-61999-6 (ebk) Typeset in Minion by Apex CoVantage, LLC

Epigram

In 1950, O. Hobart Mowrer described a mystery: [It is] the absolutely central problem in neurosis and therapy. Most simply formulated, it is a paradox—the paradox of behavior which is at one and the same time self-perpetuating and self-defeating! . . . Common sense holds that a normal, sensible man or even a beast to the limits of his intelligence, will weigh and balance the consequences of his acts: if the net effect is favorable, the action producing it will be perpetuated; and if the net effect is unfavorable, the action producing it will be inhibited, abandoned. In neurosis, however, one sees actions which have predominantly unfavorable consequences; yet they persist over a period of months, years, or a lifetime. (p. 486)

Contents

Permissions Acknowledgments Introduction Articles 1–12 (pp. 1–198) appear in Volume 1

xi xiii

Nature, Diagnosis, and Etiology of Anxiety and Related Disorders

199

13 Disorders of Emotion

201

DAVID H. BARLOW

14 The Development of Anxiety: The Role of Control in the Early Environment

227

BRUCE F. CHORPITA AND DAVID H. BARLOW

15 A Modern Learning Theory Perspective on the Etiology of Panic Disorder

265

MARK E. BOUTON, SUSAN MINEKA, AND DAVID H. BARLOW

16 A Proposal for a Dimensional Classification System Based on the Shared Features of the DSM-IV Anxiety and Mood Disorders: Implications for Assessment and Treatment

325

TIMOTHY A. BROWN AND DAVID H. BARLOW

17 The Nature, Diagnosis, and Treatment of Neuroticism: Back to the Future DAVID H. BARLOW, SHANNON SAUER-ZAVALA, JENNA R. CARL, JACQUELINE R. BULLIS, AND KRISTEN K. ELLARD

355

x Contents

The Ascendance of Evidence-Based Psychological Treatments 18 Health Care Policy, Psychotherapy Research, and the Future of Psychotherapy

391

393

DAVID H. BARLOW

19 The Dissemination and Implementation of Evidence-Based Psychological Treatments: A Review of Current Efforts

409

R. KATHRYN Mc HUGH AND DAVID H. BARLOW

20 Direct-to-Consumer Marketing of Evidence-Based Psychological Interventions: Introduction

433

LAUREN C. SANTUCCI, R. KATHRYN Mc HUGH, AND DAVID H. BARLOW

21 Evidence-Based Psychological Treatments: An Update and a Way Forward

441

DAVID H. BARLOW, JACQUELINE R. BULLIS, JONATHAN S. COMER, AND AMANTIA A. AMETAJ

Index

473

Permissions Acknowledgments

13 Barlow, D. H. (1991). Disorders of emotion. Psychological Inquiry, 2(1), Permissions acknowledgements need to be split for Vol 1 and Vol 2. 58–71. doi:10.1207/s15327965pli0201_15 14 Chorpita, B. F., & Barlow, D. H. (1998). The development of anxiety: The role of control in the early environment. Psychological Bulletin, 124(1), 3. doi:10.1037/0033-2909.124.1.3 15 Bouton, M. E., Mineka, S., & Barlow, D. H. (2001). A modern learning theory perspective on the etiology of panic disorder. Psychological Review, 108(1), 4–32. doi:10.1037/0033-295X.108.1.4 16 Brown, T. A., & Barlow, D. H. (2009). A proposal for a dimensional classification system based on the shared features of the DSM-IV anxiety and mood disorders: Implications for assessment and treatment. Psychological Assessment, 21(3), 256–271. doi:10.1037/a0016608 17 Barlow, D. H., Sauer-Zavala, S., Carl, J. R., Bullis, J. R., & Ellard, K. K. (2014). The nature, diagnosis, and treatment of neuroticism: Back to the future. Clinical Psychological Science, 2(3), 344–365. doi:10.1177/ 2167702613505532 18 Barlow, D. H. (1996). Health care policy, psychotherapy research, and the future of psychotherapy. American Psychologist, 51(10), 1050–1058. doi:10.1037/0003-066X.51.10.1050 19 McHugh, R. K., & Barlow, D. H. (2010). The dissemination and implementation of evidence-based psychological treatments: A review of current efforts. American Psychologist, 65(2), 73–84. doi:10.1037/a0018121 20 Santucci, L. C., McHugh, R. K., & Barlow, D. H. (2012). Direct-to-consumer marketing of evidence-based psychological interventions: Introduction. Behavior Therapy, 43(2), 231–235. doi:10.1016/j.beth.2011.07.003 21 Barlow, D. H., Bullis, J. R., Comer, J. S., & Ametaj, A. A. (2013). Evidence-based psychological treatments: An update and a way forward. Annual Review of Clinical Psychology, 9, 1–27. doi:10.1146/annurev-clinpsy-050212-185629

Introduction

The first volume of these collected papers begins with a chapter titled “Introduction: A Career in Psychology.” In this chapter is a description of my own odyssey—the intellectual and experiential influences that provide context to the substance of my contributions over the decades. Included is a recounting of the major settings in which I worked, how I came to be in those settings, my major collaborators, the inevitable politics that emerge in any active department, and my reflections on how it all came together to eventuate in a program of research. In subsequent sections of the chapter I describe the publications chosen from four major areas of research and writing over the past 50 years for inclusion in this book. The first area focuses on methodology and clinical research, particularly the use of single case experimental designs in evaluating the development of psychological treatments. The second area includes papers published over the past 50 years recounting the development of new treatments for anxiety and related disorders and the progress we have made over the decades. The third area incorporates research focusing on the development of knowledge regarding the nature of anxiety and temperament with implications for the study of the etiology, diagnosis, and classification of anxiety and related disorders. The fourth and final area includes papers describing developments in health care policy as well as the trials and tribulations of promoting and disseminating evidence-based psychological treatments. This volume picks up with the third of seven papers in the area focusing on the nature, etiology, diagnosis, and treatment of anxiety and related disorders. Specifically, and as described in the introductory chapter to the first volume, the first article appearing in volume 2 titled “Disorders of Emotion” (Barlow, 1991, Article 13) emerged from theoretical work first in a book published in 1988: Anxiety and Its Disorders: The Nature and Treatment of Anxiety and Panic. In this article relatively new conceptions of etiology of panic disorder were extended to other disorders of emotion such as depression, stress and anger, and mania (excitement). Research for the 1988 book Anxiety and Its Disorders also led to a focus on the construct of a sense of uncontrollability and unpredictability underpinning conceptions of anxiety, and later neuroticism itself. Expanding on ideas first presented in that book, the origins of

xiv Introduction

perceptions of uncontrollability were traced to early developmental experiences described in some detail in a Psychological Bulletin article authored with one of my students at the time, Bruce Chorpita (Chorpita & Barlow, 1998, Article 14). Almost 10 years later at the Center for Advanced Study in the Behavioral Sciences, my colleagues, Mark Bouton and Sue Mineka, and I wrote a paper updating in some detail the theory of the etiology of panic disorder integrating new findings from cognitive science and neuroscience (Bouton, Mineka, & Barlow, 2001, Article 15). In the next article in this volume Tim Brown and I began speculating on possible contributions of new findings from our long-running classification of anxiety disorders grant to DSM-5 and its successors. Taking a unified transdiagnostic perspective, we published an invited paper in 2009 proposing a new hybrid dimensional-categorical classification system based on the shared features of anxiety and mood disorders (Brown & Barlow, 2009, Article 16). This work continues to be the major focus of the classification grant under Tim’s direction. Finally, as also previously mentioned briefly, following the substantial honor of being awarded the James McKeen Cattell award from the Association for Psychological Science in 2012, our team fashioned a paper based on portions of my award address, entitled “The Nature, Diagnosis, and Treatment of Neuroticism: Back to the Future,” in which we proposed that earlier ideas from Barlow (1988) on the origins of anxiety required refocusing to higherorder dimensions of temperament, specifically neuroticism itself (Barlow, Sauer-Zavala, Carl, Bullis, & Ellard, 2014, Article 17). We note in that article that neuroticism may be more malleable than previously thought and would ideally be the target of direct therapeutic intervention. This constitutes the very heart of our research approach at the current time. The final section of this two-volume series contains selected articles on health care policy and dissemination under the heading “The ascendance of evidencebased psychological treatments.” As mentioned in the introductory chapter to the 1st volume “A Career in Psychology” describing my years at SUNY Albany, I served as President of the Society for Clinical Psychology (Division 12 of the American Psychological Association) in 1993 and created a task force on the “Promotion and Dissemination of Psychological Interventions.” This initiative grew out of efforts in the late 1980s, described in volume 1, that began when it had become very clear from a public policy point of view that the prevailing pharmacological approaches to mental disorders were being widely adopted and recommended in emerging health care policy statements. Most of these policies were striving for accountability under the relatively new concept of evidence-based practice, while at the same time addressing the spiraling costs of health care. These policies often took the form of clinical practice guidelines. As also noted in volume 1, medications were deemed the treatment of choice for panic disorder despite the fact that our data had already indicated that new psychological interventions were at least as effective as medications if not more so (e.g., Barlow & Cerny, 1988; Klosko, Barlow, Tassinari, & Cerny, 1990). Since these treatments were not readily available to clinicians or the public, and there were no efforts other than the occasional workshop to make them

Introduction xv

available, we formed a small publishing company “Graywind Publications” in order to disseminate new treatments. With unbridled optimism and a firmly established illusion of control over how easy this was going to be, we incorporated our company, invested money in printing several hundred copies of our treatment manual for panic disorder entitled Mastery of Your Anxiety and Panic, and set up shop in the basement of our house in 1988, with my wife Beverly running the show. After obtaining several mailing lists and testing the waters, the initial investment was quickly recouped, and the company became profitable. The business expanded rapidly, was moved into commercial space, and additional employees were hired. This experience convinced us that the demand existed for these programs and that the problem was little or no infrastructure for dissemination. After five years, the business required an infusion of substantial cash for expansion, a much advanced computer system for fulfillment, and a CEO to actually run the business—all steps that neither Beverly nor I were willing to take due to other commitments and a bit of fatigue from the very long hours and inevitable set of problems involved in any rapidly growing business. So we sold the business with it ultimately ending up in the hands of Oxford University Press where the series is known as “Treatments that Work.” But these early experiences made apparent the necessity of further promoting the existence of what we then called “empirically validated treatments” and of exploring methods of dissemination and implementation. In the first article in this section, published in 1996 (Barlow, 1996, Article 18), I identified emerging policies, the status of research, and the necessity of responding to developing clinical practice guidelines, such as they were at that time. By 2010, dissemination and implementation had become its own field of endeavor with appropriate funding mechanisms and emerging methodologies. Along with my student at the time, Kate McHugh, we detailed the status of those efforts offering judgments on where the field was lacking and what additional research was needed (McHugh & Barlow, 2010, Article 19). This topic was expanded into a book published in 2012 (McHugh & Barlow, 2012) and in 2013 our team once again updated the status of research on evidence-based psychological treatments with suggestions for more broad-based future efforts (Barlow, Bullis, Comer, & Ametaj, 2013, Article 21). To improve dissemination, one of those recommendations focused on taking advantage of the power of direct-to-consumer marketing of psychological interventions, a strategy that has proved enormously successful for the large pharmaceutical companies. Initial demonstrations of the potential of this strategy for psychological interventions were detailed in a special series (Santucci, McHugh, & Barlow, 2012, Article 20). Notably, another of my students, Katlin Gallo, conducted an important dissertation on this topic only recently published (Gallo, Comer, & Barlow, 2013; Gallo, Comer, Barlow, Clarke, & Antony, 2015). Thus, the articles in this last section, representing as they do substantial advances in policy, are perhaps one of the more remarkable developments over the course of my career inasmuch as we literally began with nothing. From

xvi Introduction

that humble beginning, we have now reached a point, detailed in some of the preceding publications, where governments and health care policy makers around the world, including the National Health Care System in the UK and the Veterans Health Administration in the United States, are spending billions of dollars to make evidence-based psychological treatments more readily available. Although we have a very long way to go, I believe we can all be gratified by this progress.

References Barlow, D.H. (1988). Anxiety and its disorders: The nature and treatment of anxiety and panic. New York: Guilford Press. Barlow, D. H. (1991). Disorders of emotion. Psychological Inquiry, 2(1), 58–71. doi:10.1207/ s15327965pli0201_15 Barlow, D. H. (1996). Health care policy, psychotherapy research, and the future of psychotherapy. American Psychologist, 51(10), 1050–1058. doi:10.1037/0003066X.51.10.1050 Barlow, D. H., Bullis, J. R., Comer, J. S., & Ametaj, A. A. (2013). Evidence-based psychological treatments: An update and a way forward. Annual Review of Clinical Psychology, 9, 1–27. doi:10.1146/annurev-clinpsy-050212-185629 Barlow, D.H., & Cerny, J.A. (1988). Psychological treatment of panic. New York: Guilford Press. Barlow, D.H., Sauer-Zavala, S., Carl, J.R., Bullis, J.R., & Ellard, K.K. (2014). The nature, diagnosis, and treatment of neuroticism back to the future. Clinical Psychological Science, 2(3), 344–365. doi:10.1177/2167702613505532 Bouton, M.E., Mineka, S., & Barlow, D.H. (2001). A modern learning theory perspective on the etiology of panic disorder. Psychological Review, 108(1), 4–32. doi: 10.1037/0033295X.108.1.4 Brown, T. A., & Barlow, D. H. (2009). A proposal for a dimensional classification system based on the shared features of the DSM-IV anxiety and mood disorders: Implications for assessment and treatment. Psychological Assessment, 21(3), 256–271. doi:10.1037/ a0016608 Chorpita, B.F., & Barlow, D.H. (1998). The development of anxiety: The role of control in the early environment. Psychological Bulletin, 124(1), 3. doi:10.1037/0033-2909.124.1.3 Gallo, K.P., Comer, J.S., & Barlow, D.H. (2013). Direct-to-consumer marketing of psychological treatments for anxiety disorders. Journal of Anxiety Disorders, 27(8), 793–801. doi:10.1016/j.janxdis.2013.03.005 Gallo, K. P., Comer, J. S., Barlow, D. H., Clarke, R. N., & Antony, M. M. (2015). Direct-toconsumer marketing of psychological treatments: A randomized controlled trial. Journal of Consulting and Clinical Psychology, 83(5), 994–998. doi: 10.1037/a0039470 Klosko, J.S., Barlow, D.H., Tassinari, R., & Cerny, J.A. (1990). A comparison of alprazolam and cognitive-behavior therapy in treatment of panic disorder. Journal of Consulting and Clinical Psychology, 58, 77–84. McHugh, R.K., & Barlow, D.H. (2010). The dissemination and implementation of evidence-based psychological treatments: A review of current efforts. American Psychologist, 65(2), 73–84. doi:10.1037/a0018121

Introduction xvii McHugh, R.K., & Barlow, D.H. (Eds.). (2012). Dissemination and implementation of evidence-based psychological interventions. New York: Oxford University Press. Santucci, L.C., McHugh, R.K., & Barlow, D.H. (2012). Direct-to-consumer marketing of evidence-based psychological interventions: Introduction. Behavior Therapy, 43(2), 231–235. doi:10.1016/j.beth.2011.07.003

Nature, Diagnosis, and Etiology of Anxiety and Related Disorders

Article 13

Disorders of Emotion David Harrison Barlow Center for Stress and Anxiety Disorders. State University of New York at Albany

A new preliminary model of emotional disorders, derived from basic tenets of emotion theory and new developments in cognitive science, is presented. It is suggested that tightly organized basic emotions stored in memory fire inappropriately on occasion. In individuals who are vulnerable both biologically and psychologically, these emotions may become the focus of anxiety or dysthymia in that the emotions themselves are experienced as uncontrollable and threatening with adequate coping being difficult or impossible. Early experiences with lack of control over one’s environment as well as biological vulnerabilities may well determine whether or not one becomes anxious/dysthymic over the experience of basic emotions in an inappropriate context. This model is illustrated in the context of panic disorder and then extended to depression (sadness/distress), stress (anger), and mania (excitement). Anxiety and mood disorders are fundamentally emotional disorders. Therefore, the study of anxiety, depression, and other affective states, both normal and pathological, falls within the more general purview of the study of emotion. For this reason, it is surprising that investigations of emotional disorders have only recently begun to consider some of the rich traditions in the study of emotion. The purpose of this article is to present, in summary form, a new model based in part on the accumulated wisdom of emotion theory including new and exciting developments in the area of cognitive science. I begin by considering the nature of panic and anxiety. I then review depression, taking into account evidence for the very close connection of anxiety and depression. I conclude with a look at other closely related emotions and emotional disorders that have received somewhat less attention. Panic Within the anxiety disorders, the most profound development in the past decade in terms of its impact on research and clinical practice has

202 David Harrison Barlow

been the emergence of the phenomenon of panic. Panic attacks are typically described as sudden bursts of emotion consisting of a large number of somatic symptoms and thoughts of dying and/or losing control. On the average, these symptoms are relatively consistent across people experiencing panic; however, individual panic attacks may present with a different “mix” or number of symptoms and may vary in intensity. A prominent feature of many of these attacks as they present in the clinic is the report by the client, at least initially, that no frightening situation or thought process (cue) was associated with the attack. Clients may also report the attack to be totally unexpected. Although the uncued, unexpected nature of panic attacks is clearly a construct of the client (i.e., cues are usually discovered after a systematic examination; Barlow, 1988), this phenomenon has resulted in the labeling of these attacks as spontaneous. Of course, panic attacks may also be expected and have cues, as with the specific phobic who is afraid to cross bridges (cue) and fully expects to panic if he or she must cross a bridge. These cued expected panic attacks are essentially similar in their presentation to “spontaneous” attacks (Barlow & Craske, 1990). Several years ago, my colleagues and I recorded detailed physiological changes preceding and accompanying “uncued, unexpected” panic attacks in two clients who happened to be undergoing physiological assessments at the time (Cohen, Barlow, & Blanchard, 1985). An example of physiological changes associated with one of these attacks is presented in Figure 13.1. A major goal of research over the past 3 to 6 years has been to discover the causes of panic. Given the preliminary nature of this undertaking, it is no accident that most of our ideas are simplistic. As is often the case in the early investigation of psychopathological phenomena, initial conceptualizations are one dimensional. These models may contain a grain of truth but generally espouse a linear model of causation with the emphasis on one discrete causal event. In the study of panic the most prominent one-dimensional conceptualizations specify either biological or cognitive causation. Early biological conceptualizations of panic assumed a discrete biological dysfunction as a causal mechanism. The search for this biological dysfunction has ranged far and wide during the last decade across both central and peripheral mechanisms. The most popular procedure in this quest has been the pharmacological provocation of panic attacks in the laboratory. Substances such as lactate, yohimbine, isoproteronal, carbon dioxide, and caffeine have all been used to provoke panic. During these procedures investigators have examined several neurobiological processes such as neurotransmitter action and, more recently, patterns of cerebral blood flow (e.g., Reiman et al., 1986; see Barlow, 1988, for a review). Nevertheless, despite a number of interesting findings emerging from these studies and the promise of more to come, no evidence has been forthcoming for a discrete biological marker of panic. Rather, any number of provocation procedures, many with fundamentally different biological actions, are capable of provoking panic in the laboratory. Furthermore, there is increasing evidence that psychological provocation procedures are also capable of provoking panic including the very

SUBJECT 1 BASELINE

RELAXATION

PANIC

104 100

Onset of Panic Attack

Heart Rate (BPM)

96 92 88 84 80 76 72 68 64

46 Frontalis EMG (mv)

42 38 34 30 26 22 18 14

Temperature

728 726 724 722 720 718 1

2

3

4

5

6

7

8

9

10 11 12 13 14 15

Time (min)

Figure 13.1 Physiological Changes from the Start of the Recording Session Through the Onset and Peak of the Panic Attack. (From “The Psychophysiology of Relaxation Associated Panic Attacks” by A.S. Cohen, D.H. Barlow, and E.B. Blanchard, 1985, Journal of Abnormal Psychology, 94, p. 98. Copyright 1985 by the American Psychological Association, Inc. Reprinted by permission.)

204 David Harrison Barlow

interesting and seemingly panic-inhibiting procedure of relaxation (Adler, Craske, & Barlow, 1987). In this regard, one will notice that the assessment procedure in effect in Figure 13.1 that was associated with a “spontaneous” panic was relaxation. Recent studies have also demonstrated very strong effects of psychological manipulations on biological provocation procedures such as carbon dioxide inhalations (Sanderson, Rapee, & Barlow, 1989). Another fruitful approach to the causation of panic is the search for a straightforward misattribution of otherwise normal somatic events. Currently, there seems little question that cognitive processes have an important role to play in panic (Beck, 1988; Beck & Emery, 1985; Clark, 1988; Sanderson et al., 1989). But more sophisticated cognitive theorists such as Beck and Clark would not now advocate a linear causal model. Rather they would recognize and incorporate existing strong evidence for biological contributions to panic. This evidence includes but is not limited to studies demonstrating a strong familial aggregation of panic attacks (Crowe, Noyes, Pauls, & Slymen, 1983), as well as preliminary twin studies suggesting that panic (or at least some vulnerability to panic) is inherited (e.g., Torgersen, 1983).1 It is also possible that greater specification of seemingly labile neurotransmitter systems in panickers (e.g., Charney & Heninger, 1986; Nutt, 1986) or cerebral blood flow patterns associated with panic will enlighten us further on biological bases of panic, even if specific markers are not found (e.g., Reiman et al., 1986). In any case, if the history of research in psychopathology is any guide, linear causal models will give way to multi-determined causal sequences. In the case of panic, the implication of such a systemic model is that both psychological and biological factors would contribute to causation interacting in the context of a feedback loop. A further implication of this model is that both psychological and pharmacological interventions could be effective for panic and that each intervention would affect both biological and psychological components of the panic cycle. A model in construction at our Center attempts to integrate some of the biological and psychological data emanating from the study of panic and anxiety in the context of emotion theory. This model has implications for the nature of anxiety, which I consider first. Subsequently, I return to considering the nature of panic. Anxiety Most emotion theorists who speak to the issue have concluded that anxiety is a construct that clearly differs from related emotions such as anger and fear. Contrary to conclusions in general psychology textbooks, anxiety is not just fear without a cue. Rather, theorists as diverse as Carol Izard (1977; Izard & Blumberg, 1985), Richard Hallam (1985), Peter Lang (1979, 1984, 1985), and neurobiological theorists such as Robert Cloninger (1986) and Jeffrey Gray (1982, 1985) have concluded that anxiety is a blend of different emotions and cognitions or perhaps a diffuse affective network stored in memory that is very difficult to define. Based on data developed in our Center and elsewhere, I think that the evidence

Disorders of Emotion 205

supports a conceptualization of anxiety as a loose cognitive-affective structure which is composed primarily of high negative affect, a sense of uncontrollability, and a shift in attention to a primarily self-focus or a state of self-preoccupation. The sense of uncontrollability is focused on future threat, danger, or other negative events. Thus, this negative affective state can be characterized roughly as a state of “helplessness” because of perceived inabilities to predict, control, or obtain desired results in certain upcoming situations or contexts. If one were to put anxiety into words, one might say, “That terrible event could happen again and I might not be able to deal with it, but I’ve got to be ready to try.” From this point of view, a better and more precise term for anxiety might be anxious apprehension. This conveys the notion that anxiety is a future-oriented mood state where one is ready or prepared to attempt to cope with upcoming negative events. This is best reflected in the state of chronic overarousal that seems to characterize anxiety and those who present with anxiety. This arousal may be the physiological substrate of “readiness” which may underlie an effort to counteract helplessness (e.g., Fridlund, Hatfield, Cottam, & Fowler, 1986). Vigilance (hypervigilance) is another characteristic of anxiety that suggest readiness and preparation to deal with negative events. The process of anxiety, as this model would have it, is presented in Figure 13.2. With more specific reference to the model in Figure 13.2, a variety of cues, or propositions in Langian (1985) terms, would be sufficient to evoke anxious apprehension without the necessity of a conscious rational appraisal. The cues may be broad based or very narrow, as in the case of test anxiety or sexual dysfunction. The state of negative affect with its associated arousal and negative valance is, in turn, associated with distortions in information processing. These cognitive distortions are characterized by an attentional shift to a self-evaluative focus (or a rapidly shifting focus of attention from external EVOCATION OF ANXIOUS PROPOSITIONS (situational contexts, unexplained arousal, or other cues)

(POSSIBLE) AVOIDANCE of situational context or other aspects of negative affect (e.g., arousal) if feasible.

DYSFUNCTIONAL PERFORMANCE

NEGATIVE AFFECT A. Unpleasant affective state associated with perceptions of future unpredictability and uncontrollability. B. Preparatory coping set accompanied by supportive physiology (arousal) or Gray’s behavioral inhibition system.

ATTENTIONAL SHIFT to self-evaluative focus (on physiological or other aspect of responding)

Intensification

and/or lack of concentration on task at hand

ADDITIONAL INCREASES IN AROUSAL Intensification

INTENSE WORRY (“hot” apprehensive cognition)

ATTENTION NARROWING on negative affective content, hypervigilance (enhanced recognition of threat related stimuli)

APPREHENSIVE HYPERVALENT COGNITIVE SCHEMA

Figure 13.2 The Process of Anxious Apprehension. From Anxiety and Its Disorders: The Nature and Treatment of Anxiety and Panic (p. 250) by D.H. Barlow, 1988, New York: Guilford. Copyright 1988 by Guilford Press. Reprinted by permission.

206 David Harrison Barlow

sources to internal self-evaluative content). Evidence suggests that this, in turn, further increases arousal forming its own small positive-feedback loop with negative affect. Continuing on in the larger feedback loop, attention narrows on to sources of threat or danger, setting the stage for additional distortions in the processing of information, and one becomes hypervigilant for cues or stimuli associated with sources of apprehension (e.g., MacLeod, Mathews, & Tata, 1986). This process results in arousal-driven worry that, at intense levels, is very difficult to control by clients (Borkovec, Shadick, & Hopkins, in press). At sufficient intensity, this process results in disruption of concentration and performance and, ultimately, avoidance of sources of apprehension if this method of coping is available. Arousal-driven anxious apprehension will, of course, only interfere with performance if some performance is required. In situations where performance may not be called for immediately, but where perceptions of loss of control or other negative affective content have become associated with a number of important life events (e.g., health, finances, and family concerns), the process of worry will emerge. The intensity of worry will increase or decrease depending on situational context, the amount of underlying autonomic arousal that is available at the time for transfer (Zillmann, 1983), and/or the presence of other “propositions” capable of calling forth this diffuse cognitive-affective structure. Evidence for each of these components of anxiety and their connection is presented in some detail elsewhere (Barlow, 1988). It is also important to note that this is not a description of the etiology of anxiety but rather an illustration of the process of anxious apprehension. The etiology of anxious apprehension and a description of the biological and psychological vulnerabilities that predispose this state follows later and is fully discussed elsewhere (Barlow, 1988). What is the purpose of anxiety? Why are we programmed to become anxious? This has prompted much speculation from philosophers and psychologists but also generated some data. We have known for over 80 years that organisms become more vigilant, learn more quickly, and perform better both motorically and intellectually if anxious (Yerkes & Dodson, 1908). We also know that both benzodiazepines and relaxation interfere with effective performance (Barlow, 1988). From this point of view, anxiety can be very adaptive, up to a point, and the adaptive purpose of anxiety would seem to be planning and preparation to meet a challenge or threat. As Liddell (1949) noted, The planning function of the nervous system, in the course of evolution, has culminated in the appearance of ideas, values, and pleasures—the unique manifestations of man’s social living. Man, alone, can plan for the distant future and can experience the retrospective pleasures of achievement. Man, alone, can be happy. But man, alone, can be worried and anxious. Sherrington once said that posture accompanies movement as a shadow. I have come to believe that anxiety accompanies intellectual activity as its shadow and that the more we know of the nature of anxiety, the more we will know of intellect. (p. 185) It is also well-known that anxiety distributes as a trait (Eysenck, 1967, 1981) and is expressed more or less by most individuals under certain situations.

Disorders of Emotion 207

Therefore, it is only very intense anxiety or perhaps anxiety with an inappropriate focus that comes to the attention of clinicians. In addition, as noted earlier, strong evidence supports the existence of both biological and psychological vulnerabilities to developing anxiety. I return to these vulnerabilities later. Fear If anxiety is, in fact, a diffuse cognitive-affective structure serving the adaptive purpose of preparation, what is fear? Emotion theorists generally agree that fear is a distinct, primitive, basic emotion or perhaps a tightly organized, cohesive affective structure stored in memory, that is fundamentally a behavioral act. Fear is associated with intense neurobiological and cognitive features. Fear occurs when we are directly and imminently threatened whether by wild animals, which was so often the case when our distant ancestors lived in caves, or by more modern-day dangers such as a vehicle careening out of control. The action tendency that is at the heart of this emotion is the well-known “fight-or-flight” response. It is essential that this response be instantaneous because the survival of the organism may depend on it. This is Cannon’s (1927) “emergency response” or “alarm reaction.” Most theorists would agree that this response is evolutionarily favored, ancient, and found far down the phylogenetic scale. Subjectively, the response is characterized by an overwhelming urge to escape, most often expressed as “I’ve got to get out of here.” This seems to reflect the basic action tendency of escape. What is the clinical manifestation of fear? Based on accumulating developmental and phenomenological evidence, it appears that panic attacks represent the clinical manifestation of the basic emotion of fear. The hair-trigger, instantaneous quality of fear is very well illustrated in Figure 13.1. Here one will notice that this unexpected surge of autonomic arousal with accompanying subjective manifestations of fear peaked and then diminished substantially in the space of 3 min. A Model of Panic Disorder One remarkable fact that has come to light in recent years, and has been confirmed independently around the world, is that the experience of panic is relatively common. The first study to suggest that panic attacks occur frequently in the general population was reported by Norton, Harrison, Hauch, and Rhodes (1985), who administered questionnaires to 186 normal young adults. Fully 34.4% of these subjects reported having one or more panic attacks in the past year. Norton, Doward, and Cox (1986) cross-validated this study, coming up with very similar data. Both sets of results are presented in Table 13.1. Additional data from our clinic and around the world (Brown & Cash, 1990; Rapee, Ancis, & Barlow, 1987; Salge, Beck, & Logan, 1988; Wittchen, 1986) suggest that approximately 9% to 15% of the population has experienced uncued, unexpected “spontaneous” panic attacks. Therefore, a high percentage of the population experiences panic although the majority report panic attacks in response to some identifiable trigger or cue such as taking an exam or speaking in public as in the Norton et al. (1985, 1986) data. But a substantial minority experienced the unexpected, uncued panic attacks that are the hallmark of panic disorder.

208 David Harrison Barlow Table 13.1 Panic in the normal population

One or More Panics in Last Year

Norton, Harrison, Hauch, and Rhodes (1985)a

Norton, Dorward, and Cox (1986)b

34.4%

35.9%

Panic Frequency Last Year 1 or 2 3 or 4 5 or more

17.2% 11.3% 6.0%

Panic Frequency Last 3 Weeks 1 2 3 or more

17.2% 4.8% 2.1%

12.5% 7.0% 3.1%

Note: From Anxiety and Its Disorders: The Nature and Treatment of Anxiety and Panic (p. 103) by D.H. Barlow, 1988, New York: Guilford. Copyright 1988 by Guilford Press. Adapted by permission. a N = 186. b N = 256.

Another factor that builds confidence in the validity of these findings from the general population is the strong aggregation of panic and other psychopathology in the families of nonclinical infrequent panickers. For example, Norton et al. (1986) found that a significantly greater proportion of panickers than nonpanickers reported fathers, mothers, brothers, and sisters who had panic attacks. These findings are particularly strong because the results were statistically significant for each class of relatives rather than just in the aggregate. These data resemble results demonstrating a high familial aggregation of panic in the families of patients with panic disorder (Crowe et al., 1983). The major difference in the presentation of these nonclinical spontaneous panickers is quite striking. These individuals show little or no concern over the attacks or the possibility of experiencing another one. Rather, they dismiss the attacks as associated with some passing event or episode such as something they ate or a difficult day at work. They seem to quickly put the experience behind them. Even if they occur at relatively high frequencies, these panic attacks do not seem to elicit the type of concern that is readily apparent in those presenting with a clinical disorder. This reflects a fundamental difference between clinical and nonclinical panickers that is receiving confirmation from a variety of sources. The distinguishing characteristic of individuals with panic disorder is the development of anxious apprehension over the next unexpected panic attack. That is, individuals with panic disorder apprehensively anticipate the next attack, perceive the attacks as uncontrollable, and are extremely vigilant to somatic symptoms that might signal the beginning of the next attack. For this reason at least a 1-month period of anxiety over having another panic attack is a defining characteristic of panic disorder in the Diagnostic and Statistical Manual of Mental Disorders (3rd ed., rev.; DSM–III–R; American Psychiatric Association, 1987).

Disorders of Emotion 209

This leads one, inevitably, to the conclusion that the fundamental difficulty in panic disorder is not necessarily the panic attacks themselves, but rather the development of anxiety in association with the attacks. Panic Versus Anxiety In addition to evidence reviewed so far, there are several other reasons for considering that panic, as a clinical manifestation of fear, is best conceptualized as distinct from anxiety. First, both physiological and phenomenological evidence suggests that panic attacks are descriptively and functionally unique events when compared to anxiety. Panic attacks present differently from anxious apprehension and are experienced differently by clients (Barlow et al., 1985; Cohen et al., 1985; Rapee, 1985; Taylor et al., 1986). Second, research demonstrating a strong functional relationship between panic attacks and subsequent anticipatory anxiety is now appearing (e.g., Rachman & Levitt, 1985). That is, whether a panic attack occurs or not in a specific circumstance and whether it is expected or not will influence subsequent anxiety (and avoidance) associated with the phobic situation. In this regard, it is conceptually difficult to consider the alternative of developing “anxiety” focused on the future experience of “anxiety.” Third, a strong (although not universal) consensus from the basic study of emotion, as noted earlier, suggests a fundamental difference between anxiety and fear. Fear is thought to be a basic action tendency of fight or flight that is a tightly organized, affective structure in memory. This contrasts with anxiety which is seen as a blend of basic emotions or, more likely, as a diffuse cognitive-affective structure associated with preparation for future threat or challenge. Unlike biological models of panic, this hypothetical separation of anxiety and panic does not imply that panic itself involves a biological dysregulation. Rather, panic is conceptualized as a normal fear response firing inappropriately. Further circumstantial evidence for this conceptual separation is present in genetic studies and studies of the aggregation of emotion-related action tendencies. Although family studies of panic in clinical context have been confounded somewhat, due to definitional problems in DSM–III and the lack of consideration of severity of the anxiety disorder, evidence from animal laboratories (as well as other clinical disorders reviewed later) is more striking (Barlow, 1988). For example, ancient and seemingly innate defensive reactions such as freezing when under attack by a predator and fainting at the sight of blood or injections are highly familial and almost certainly heritable. For example, fully 67% of blood phobics report biological relatives with the same reaction (Ost, Lindahl, Sterner, & Jerremalm, 1984). Remember also that panic attacks in nonclinical panickers are strongly familial (Norton et al., 1986) despite the fact that little or no anxiety over these attacks is present. Thus, the action tendency of panic may have a distinct genetic component that differs from heritable qualities associated with the trait of anxiety (Barlow, 1988; Eysenck, 1970). Nevertheless, other viable, heuristic models of panic disorder exist, some of them already mentioned, that do not hold that panic and anxiety should be conceptualized as distinct (e.g., Clark, 1988; Ehlers & Margraf, in press). But let us assume, for the sake of argument, that panic is best considered distinct from anxiety and represents the clinical manifestation of fear. With these

210 David Harrison Barlow

models of anxiety, panic, and panic disorder in mind, let us now turn our attention to depression. Depression What is depression and how does it differ from anxiety? Many different points of view on this distinction have appeared (e.g., Kendall & Watson, 1989). Several theorists have concluded that anxiety and depression are variable expressions of the same pathology. Yet another group of theorists suppose that anxiety and depression are fundamentally different and distinct. Other theorists have expressed views falling somewhere in between, suggesting, for example, a common diathesis with subsequent divergence occurring for any one of a number of reasons (Weissman, 1985). The evidence for a unitary view is very strong. For example, on a neurobiological level the process seems to be very similar or perhaps identical. Gray (1985), reviewing evidence in the animal laboratories, suggests that learned helplessness construed as an animal model of depression may be identical in its neurobiological underpinnings to models of anxiety. Specifically, enhanced hippocampal function as a result of increased noradrenergic input to this and other regions of the forebrain seem to underlie both anxiety and depression. Interestingly, Breier, Charney, and Heninger (1985) found identical, greatly enhanced, underlying noradrenergic activity in clinical populations of either panic disorder on the one hand or major depressive disorder on the other. Several recent studies have also shown that patients with relatively pure cases of generalized anxiety disorder or panic disorder show similar rates of nonsuppression on the dexamethasone suppression test when compared to cases of major depressive disorder (e.g., Avery et al., 1985; Coryell, Noyes, Clancy, Crowe, & Chaudhry, 1985; Schweizer, Swenson, Winokur, Rickels, & Maislin, 1986). In addition to this evidence, data from family studies strongly suggest that anxiety, depression, and panic are closely related. Generally, the more signs and symptoms of anxiety and depression, the greater the rate of anxiety or depression or both in first-degree relatives and children (Leckman, Weissman, Merikangas, Pauls, & Prusoff, 1983; Puig-Antich & Rabinovich, 1986; Weissman, 1985). Finally, the emerging evidence from pharmacological treatments for both anxiety and depressive disorders is that tricyclic antidepressants seem to be the treatment of choice. Although these drugs are known as antidepressants, it now seems clear that they are effective with panic disorder and not just for panic disorder with accompanying depression (Mavissakalian, 1987; Mavissakalian & Michelson, 1986). Furthermore, evidence now exists that tricyclics such as imipramine are effective with anxiety disorders where panic is not a prominent feature, such as generalized anxiety disorder (Kahn et al., 1986; Klein, Rabkin, & Gorman, 1985). In addition, a close analysis of the literature suggests that the effectiveness of tricyclics in cases of panic disorder may not necessarily be due to direct blockade of panic, as commonly assumed, but rather to therapeutic effects on anxious apprehension associated with panic (Barlow, 1988). There is also evidence that new high-potency benzodiazepines

Disorders of Emotion 211

such as alprazolam are effective not only for panic disorder (Ballenger et al., 1988) but also for some types of depressive disorders (e.g., Rickels, Feighner, & Smith, 1985). Thus, evidence from both basic neurobiological studies and drug treatment studies support the unitary view of anxiety and depression. Do questionnaires and inventories provide any help in discriminating anxiety and depression? In fact, an examination of popular rating scales for either anxiety or depression demonstrates that whether one is measuring anxiety or depression, these scales correlate very highly. For example, Dobson (1985) examined scales such as the State-Trait Anxiety Inventory (Spielberger, Gorsuch, & Lushene, 1970) and the Zung Self-Rating Depression Scale (Zung, 1965), among others. The correlation among anxiety scales was .66; among depression scales, it was .69; and among anxiety and depression scales, it was .61. The amount of shared variance for each correlation was very high, ranging from .37 to .47. This degree of shared variance suggests that these questionnaires are not useful in measuring the intensity of two different affective states, anxiety and depression; rather, they are measuring the same or very similar affective states. These correlations are displayed in Table 13.2. This reflects a finding also reported by Gotlib (1984), who found that self-report measures of depression and anxiety correlated highly in college students (see Gotlib & Cane, 1989, for a thorough discussion of this issue). Do clinical rating scales discriminate anxiety and depression? The most popular and widely used rating scales for anxiety and depression are the Hamilton scales (Hamilton, 1959, 1960). But on these scales the overlap on specific questions exceeds 70% (Barlow, 1985). Therefore it is not surprising that these scales are highly correlated, share considerable variance, and demonstrate poor discriminant validity. For example, groups with depressive diagnoses score as high or higher on the Hamilton Anxiety scale than any anxiety-disorder group (Di Nardo & Barlow, 1990). Similarly, many anxiety disorders do not differ significantly from depressive disorders on the Hamilton Depression scale. Of course, discriminant validity or diagnosis is not the purpose of these scales. Nevertheless, in our clinic since the early 1980s we have identified several items from the Hamilton scales that seem to discriminate consistently anxious and depressed clients (Barlow, 1983). Primary among these items, in addition to suicidal thoughts and depressed mood, are feelings of hopelessness and motor retardation. These findings are presented in Table 13.3. This finding is not unique to our clinic but has been reported since the 1950s by Sir Martin Roth and colleagues (e.g., Roth, Gurney, Garside, & Kerr, 1972). Table 13.2 Average correlation

Anxiety-Anxiety Depression-Depression Anxiety-Depression Shared Variance in Each Correlation

.66 .69 .61 37% to 48%

Note: From “The Relationship Between Anxiety and Depression” by K.S. Dobson, 1985, Clinical Psychology Review, 5, (p. 314). Copyright 1985 by Pergamon Press plc. Adapted by permission.

212 David Harrison Barlow Table 13.3 Hamilton depression items for the diagnostic groups

Agoraphobia Social Simple Panic Generalized Obsessive- Major Phobia Phobia Anxiety Compulsive Affective Retardation 1.09a Suicide 1.55b Helplessness 1.79a

1.13a 1.50ab 1.57a

1.00a 1.00a 1.00a 1.36ab 1.18a 1.13a 1.57a 1.58a 1.45a

1.25a 1.33ab 2.00a

1.58b 2.33c 2.67b

Note: Means having the same subscript are not significantly different at p < .05.

More recently, Riskind, Beck, Brown, and Steer (1987) revised the two Hamilton scales to discriminate more clearly between anxiety and depressive disorders in clinical populations. This revision is one of the first attempts to construct scales based on what is essential and important about anxiety and depression when compared to each other, as opposed to scale construction based simply on descriptors of unpleasant mood or dysphoria that are common to both anxiety and depression. Items were reassigned after factor analyses and point-biserial correlational analyses based on scores from relatively pure groups of anxious and depressed patients without comorbid depressive or anxiety diagnoses. As a result of this analysis, all items associated with arousal or activation were reassigned to the Anxiety scale (Riskind et al., 1987). From this point of view, fundamental differences between clearly defined anxiety and depression may be found in action tendencies and underlying associated physiology. Anxiety suggests engagement and activation; depression suggests disengagement and inactivity. Anxiety implies an effort to cope with difficult situations and the physiology is there to support active attempts at coping. Depression is characterized by behavioral retardation and an associated lack of arousal. With this clarification, we can also bid farewell to the conceptually muddled concept of “agitated depression.” Agitation clearly groups with activation and chronic overarousal which are at the core of anxiety. This contrasts with motor retardation and loss of pleasurable engagement which are characteristics unique to depression. Nevertheless, even with these revisions, revised Hamilton scores reflect the fact that patients earning a diagnosis of depression are anxious. We now have two different samples of anxiety-disorder patients from our clinic to whom we have administered these revised Hamilton scales (Di Nardo & Barlow, 1990; McCauley, Di Nardo, & Barlow, 1988). Notice in Table 13.4 that the depressive disorders are no longer the most anxious on the Anxiety scale, but they are still more “anxious” than simple or social phobia. However, on the Hamilton Depression scale, the depressive disorders are now clearly significantly different from all anxiety disorders. These developments also underscore the value of studying relatively pure groups of patients in an effort to uncover the nature of anxiety and depression. We have now begun this investigation along several lines. In one study, we looked at attributions of dysthymic patients compared to normal controls. To these groups we compared the attributions of anxiety patients who were arrayed on the Beck Depression Inventory (BDI, which contains ratings of

Disorders of Emotion 213 Table 13.4 Comparison of anxiety disorder and depressed groups on revised hamilton anxiety and depression scales Diagnosis Simple Social Major Dysthymia Generalized Obsessive- Panic Mixed Agoraphobia Phobia Phobia Depression (n = 15) Anxiety Compulsive (n = 19) (n = 19) With (n = 20) (n = 19) (n = 15) (n = 20) (n = 17) Panic (n = 18) Anxiety 17.3a Depression 16.3a

19.3ab 19.9bc

20.3abc 29.9f

21.0bcd 28.9f

21.2bcd 20.3bcd

23.4cd 24.4de

23.7de 20.2bc

25.3ef 23.9de

27.lf 22.6de

Note: Means having the same subscript are not significantly different at p < .05.

the intensity of core depressive symptoms) as either nondepressed or moderately depressed. Nevertheless, no anxiety-disorder patients, even those scoring in the moderately depressed range on the BDI, received an additional depressive-disorder diagnosis. Attributions were assessed by the well-known Attributional Style Questionnaire (Peterson et al., 1982). The results indicated clear differences between anxious and depressed patients once the presence and severity of depressive symptoms were controlled for by covariation procedures. That is, depressed patients displayed internal, global, and stable attributions for negative outcomes. Some anxiety patients also demonstrated this attributional style but only if they were also depressed. Anxiety patients, despite the severity of their anxiety disorder, had scores almost identical to those of the normal group if they were not depressed (Heimberg, Vermilyea, Dodge, Becker, & Barlow, 1987). Thus, scores on rating scales, as well as our study of attributional style, tend to converge on one conclusion. Depressive signs and symptoms but not anxiety signs and symptoms discriminate these groups. To put it another way, almost all depressed patients are anxious, but not all anxious patients are depressed. Evidence from a variety of other sources and methods in studies with both adults and children also reflects this general conclusion. For example, a close examination of patterns of comorbidity in our clinic suggests that individuals presenting with anxiety disorders do not necessarily present with comorbid depressive diagnoses, but clients with diagnosable depressive disorders very often present with comorbid anxiety diagnoses (Barlow, 1988; Barlow, Di Nardo, Vermilyea, Vermilyea, & Blanchard, 1986; Di Nardo & Barlow, 1990; Sanderson, Di Nardo, Rapee, & Barlow, 1990). This general conclusion is also supported when one examines the broad mood factors of negative and positive affect as isolated by Auke Tellegen (Tellegen, 1985). (See Figure 13.3.) For example, Watson, Clark, and Carey (1988), working within this framework, found that negative affect correlated broadly with symptoms and diagnoses of both anxiety and depression. But positive affect was related (negatively) only to symptoms and diagnoses of depression, indicating that the loss of pleasurable engagement or anhedonia, which is the hallmark of low positive affect, is a distributive feature of depression. This information, along with the oft-observed finding that anxiety tends to precede

214 David Harrison Barlow

Figure 13.3 The Two-Factor Structure of Self-Rated Mood. From “Structures of Mood and Personality and Their Relevance to Assessing Anxiety with an Emphasis on Self-Report” by A. Tellegen, 1985, in A.H.Turna and J.D. Maser (Eds.), Anxiety and the Anxiety Disorders (p. 691), Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Copyright 1985 by Lawrence Erlbaum Associates, Inc. Reprinted by permission.

the occurrence of depressive episodes, suggests one possible general conclusion. At least certain types of depression grow out of anxiety, or are a complication of anxiety occurring in some people under some conditions. The Causes of Anxiety and Depression A detailed discussion of etiology is not possible in the context of this article, but I have suggested elsewhere (Barlow, 1988) that anxiety and depression do share common biological vulnerabilities best described as an overactive neurobiological response to stressful life events. This would be conceptualized as a general proneness to develop anxiety or depression rather than a specific vulnerability for anxiety and depression itself. In addition to this biological vulnerability, I have also suggested that these individuals possess a psychological vulnerability based on early experiences with controllability. Here the pioneering work of Susan Mineka in reinterpreting the importance of the variable of uncontrollability in some of the early experimental neurosis work is extremely relevant (Mineka, 1985a, 1985b; Mineka & Kihlstrom, 1978). More recent work in elucidating the effects in primates of early experiences with uncontrollability on later manifestation of fear and anxiety is also crucial (Mineka, Gunnar, & Champoux, 1986). In fact, the concept of control or lack of it has

Disorders of Emotion 215 Table 13.5 Symptom predictions of the hopelessness theory

Causal Cognition

Symptoms

Uncertain Helplessness

Certain Helplessness

Hopelessness

Pure “Aroused” Anxiety Syndrome

Mixed “Retarded” Anxiety/ Depression Syndrome

Pure Depression Syndrome

Note: From “Comorbidity of Anxiety and Depressive Disorders: A Helplessness-Hopelessness Perspective” by L.B. Alloy, K.A. Kelly, S. Mineka, and C.M. Clements, 1990, in J.D. Maser and C.R. Cloninger (Eds.), Comorbidity of Mood and Anxiety Disorders (p. 526), Washington D.C.: American Psychiatric Press. Copyright 1990 by the American Psychological Association, Inc. Adapted by permission.

an important history in psychology, as is evident in the work of Julian Rotter (1954) and George Mandler (1966), as well as in Martin Seligman’s work on learned helplessness (see Seligman, 1975). But somehow the relationship of controllability to anxiety seems to have been overlooked until its recent revival. In any case, I suggest that early experiences with lack of control provide a psychological vulnerability for anxiety. This psychological vulnerability, when combined with a biological vulnerability and triggered by the stress of negative life-events, leads to clinical anxiety and, sometime later, to possible depression. Depression then may simply reflect an extreme psychological vulnerability to experiences of unpredictability and uncontrollability based on early experiences with controllability and coping. In other words, whether one becomes anxious and stays that way, or also becomes depressed, depends on the extent of one’s psychological vulnerability, the severity of the current stressor, and coping mechanisms at one’s disposal. I noted earlier that one could put anxiety into words by saying, “That terrible event is not my fault but it may happen again and I may not be able to cope with it, but I’ve got to be ready to try.” The depressed response would be, “That terrible event might happen again and I won’t be able to cope with it, and it’s probably my fault anyway, so there’s really nothing I can do.” In other words, the anxious individual continues to fight the good fight, but the depressed individual has given up. What is particularly interesting about this conceptualization is that Lauren Alloy, Susan Mineka, and their students coming at it from the viewpoint of hopelessness depression have arrived at essentially the same conclusion as depicted in Table 13.5. Disorders of Emotion Double Depression and Double Anxiety

But another important factor seems to be complicating the presentation of depression. This factor brings us beyond the notion of depression (and anxiety) as the diathesis of a common set of biological and psychological vulnerabilities. To understand this factor it now becomes important to refer back to

216 David Harrison Barlow

the model of panic disorder outlined earlier. Specifically, panic attacks seem to occur with some frequency in the normal population, but panic disorder develops when anxiety is focused on the possibility of another “spontaneous” uncontrollable attack occurring. Based on our analyses of anxiety, the vulnerabilities to become anxious and, most likely, the actual presence of generalized anxiety itself, with its characteristic early onset and chronicity (Barlow et al., 1986), would predate the first panic. But once the panic occurs, it becomes the focus of anxious apprehension. What is fascinating is that a similar phenomenon seems to occur in depressive disorders. It would be symmetrical for the model if some individuals experienced a fundamental basic emotion of sadness/distress in the same way they seem to experience the basic emotion of fear or panic. That is, much as one seems to experience a false alarm, one could feel sadness/distress for no discernible reason (unexpected and uncued) which would then be experienced as unpredictable and/or uncontrollable. Based on models presented earlier, the experience of this emotion would be sufficient to trigger anxiety and/or depression in those vulnerable to this affective state. Others without those vulnerabilities would not experience anxiety or depression. Clinically this seems to happen. Keller and Shapiro (1982) and others (e.g., Miller, Norman, Dow, 1986) have described the phenomenon of double depression where bouts of acute and deep depression are superimposed on a preexisting chronic dysthymic state. This is widely observed in clinical practice. But John Teasdale (1985) seemed to capture the essence of this phenomenon most completely when he wrote: It is not uncommon for depressed patients to misinterpret symptoms of depression as signs of irremediable personal inadequacy: for example, the lack of energy, irritability or loss of interest and affection that characterizes depression are seen as signs of selfishness, weakness, or as evidence that a person is a poor wife or mother. Such interpretations, as well as making the symptoms more aversive, imply that they are going to be very difficult to control. (p. 160) In other words, the experience of the basic emotion of sadness or depression itself is perceived as uncontrollable in these patients. Terminology now becomes problematic. It is easy to say that one becomes apprehensively anxious over panic attacks, but it is not helpful to say that one becomes depressed over bouts of depression. For this reason, I have suggested that the already well-utilized term dysthymia be employed to refer to that stage of anxious apprehension where coping mechanisms break down and hopelessness takes over. This term would encompass what Abramson, Metalsky, and Alloy (1989) as well as Alloy, Kelly, Mineka, and Clements (1990) have referred to as hopelessness depression. Using this terminology, dysthymic patients may become increasingly hopeless, but very few of them go into a profound psychomotor retardation with accompanying neurobiological and psychophysiological shutdown and anhedonia for more than a brief period of time. It is also not unusual for these periods of extreme sadness to appear “out of the blue” or for no apparent

Disorders of Emotion 217

reason. Finally, these bouts of sadness or major depression can occur outside of the context of a clinical disorder, as when a loved one dies. The profound sadness with all the accompanying symptoms are there but, as the grief process proceeds, it passes. In addition, Keller and Lavori (1984) observed that 97% of patients they had studied with double depression had recovered from their major depressive episode at a follow-up period. But very few had recovered from their dysthymia (Keller, Lavori, Endicott, Coryell, & Klerman, 1983). These observations reflect the fact that both panic attacks as well as periods of sadness or major depression are fundamentally self-limiting. In addition, 50% of people who are dysthymic or hopeless experience panic attacks based on recent evidence from our clinic (Benshoof, 1987). Bouts of major depression may occur across the helplessness-hopelessness spectrum. This suggests a relative independence of these basic emotions of panic (fear) and major depression (sadness/distress) from the more chronic states of anxiety and dysthymia or dysphoria. The origins of the first panic attack or major depressive episode are discussed elsewhere (Barlow, 1988), but suffice it to say, there are marked similarities in their origins. In any case, what becomes the central focus of perceptions of uncontrollability, in my view, are the unpredictable, unexpected, and uncued negative emotions themselves. I suggest that this is the core of emotional disorders as reflected in Figure 13.4. Of course, one does not need the very visible and central feature of a panic attack or major depressive episode to present with a severe emotional disorder. Earlier we have described how chronic and severe anxious apprehension in and of itself can result in substantial disruptions in performance and interference with functioning even if the focus is rather diffuse. As I have speculated elsewhere (Barlow, 1988), generalized anxiety disorder may be the more

Figure 13.4 The Relationship of Fear (Alarm), Depression (Sadness), Anxiety, and Dysthymia. From Anxiety and Its Disorders: The Nature and Treatment of Anxiety and Panic (p. 282) by D.H. Barlow, 1988, New York: Guilford. Copyright 1988 by Guilford Press. Reprinted by permission.

218 David Harrison Barlow

chronic and severe disorder in the long run in that it has an earlier onset, runs a longer course, and seems more difficult to treat when compared to other anxiety disorders such as panic disorder (which is certainly more problematic for a brief period of time). Akiskal (1983) suggested that dysthymic disorders are also chronic, run a longer course, and can be, more difficult to treat when compared to major depressive episodes. Now Daniel Klein and his colleagues (Klein, Taylor, Dickstein, & Harding, 1988) have demonstrated that clients presenting with primary early-onset dysthymia exhibit greater global impairment, an earlier age of onset of major depression, and a greater likelihood of having recurrent major depressive episodes when compared to a group of clients without dysthymia who are undergoing a major depressive episode. Interestingly, the dysthymic patients also present with more substantial family histories of affective and antisocial personality disorders and report higher levels of negative stressful life-events. This led Klein et al. to conclude that in the long run, dysthymia is the more severe form of affective disorder. There is little question that dysthymia subsumes a range of presentations from subthreshold mood disorders to deeply embedded traits of character (Akiskal, 1983). Similarly, anxiety can range from rather mild anxious symptomatology to a more stable trait (Eysenck, 1970). But what is important is that transitory emotional states of panic or depression can develop an intricate functional relationship with these more chronic affective states. Stress and Anger Disorder

It is also possible to extend this model to stress disorders. At our Center for Stress and Anxiety Disorders we see any number of individuals who come in with rather serious stress-related psychophysiological disorders such as hypertension. Nevertheless, many of these individuals display little or no anxiety or dysthymia whatsoever over the stress in their lives. In fact, their method of coping with stress, including any psychophysiological disorders they might have, is characterized by hard work, continual attention to achievement, and remarkable confidence in their ability to deal with problems. Many of these individuals would fit the loosely defined “Type A” behavior pattern. Often these clients seem to have an exaggerated sense of mastery and control and exercise their control or coping style at the expense of their physical well-being. It is well-known that these individuals are at risk for cardiovascular disease. But what investigators such as Chesney (1985, 1986) have demonstrated recently is that it is not hard work or the sense of being “driven” that puts these people at risk, but rather their aggressive, angry outbursts. Investigators working in this area have found that it is important to distinguish periods of anger (state anger) from more enduring trait characteristics (Spielberger et al., 1985; Spielberger, Krasner, & Solomon, 1988). On any circumplex of emotion, anger and fear are very closely related. Both are characterized by high negative affect and an unpleasant valence. This is also seen in the emergency reaction where flight is paired with fight. The one differentiating characteristic in factor-analytic studies of anger and fear is the sense of control or lack of it. Fear is characterized by a lack of control, anger by a sense of control and mastery. Thus, these individuals might share the same

Disorders of Emotion 219

biological vulnerability to be overresponsive to stress or challenge. Similarly, they might also experience “false alarms” in response to negative life-events much as do clients with panic disorder. But with their sense of mastery and control, these alarms would be experienced as anger (fight) rather than fear (flight). In this conceptualization, stress patients would never lose their illusion of control, even over their emotions (Alloy, Abramson, & Viscusi, 1981). Therefore, a more accurate (and symmetrical) name for these periodic angry outbursts might be anger disorder. In this light, anger disorder might be analogous to anxiety and mood disorders as yet another emotional disorder. Hypothetical similarities and differences among anxiety, mood, and stress disorders are represented in Table 13.6. As with anxiety and depressive disorders, the close relationship between anxiety and stress disorders ensures that individuals may present with considerable overlap between the two. For example, Lee and Cameron (1986) reported that almost 75% of a group of anxiety-disorder patients had Type A behavior (based on a questionnaire). Type A behavior was significantly more prevalent in male anxiety patients (92%) than in females (52%). Clinically, excessive anger is treated in a very similar fashion to anxiety, both psychologically (e.g., Biaggio, 1987; Novaco, 1975) and pharmacologically (Mattes, 1986), although the use of tricyclics for anger attacks has not yet been attempted. This overlap is best reflected in the shared biological vulnerability of an overresponsive, or perhaps labile, neurobiological response to stress, as well as in the experience of false alarms (the false firing of a fightor-flight response is widespread both in clinical and nonclinical populations; Barlow, 1988). Only in psychological vulnerabilities resulting in a sense of unpredictability or uncontrollability, as well as in coping action tendencies and associated arousal supporting these action tendencies, would differences be found. Unlike anxiety and depressive disorders, individuals with stress disorders would present with a sense of control and mastery. Unlike stress and anxiety disorders, where attempts at coping and associated arousal are always present, depressive disorders would present with disengagement and a lack of coping. It is also fascinating in our society that anxiety and depression are largely female disorders. The vast majority of individuals with Type A syndrome and other anger-linked stress disorders, on the other hand, are male. The crucial etiological factor in this sex difference may be reflected in Table 13.6. Rather than fundamental biological differences, psychological vulnerabilities associated with an acquired sense of uncontrollability and ineffective coping may be differentially distributed among men and women. That is, women with the same biological vulnerabilities as certain men, may be overrepresented among those with anxiety and depressive disorders because they do not experience the same level of mastery and control over their environment as men do. Excitement Disorder

Finally, it now seems possible to extend this model in a rather unexpected direction to address mania. For example, Post et al. (1989) reported that a substantial number of patients studied at the peak severity of a manic episode

220 David Harrison Barlow Table 13.6 Hypothetical similarities and differences among anxiety, depressive, and stress disorders

Biological Vulnerability Stressful Events Stress-Related Neurobiological Reactions False Alarms Psychological Vulnerability Resulting in a Sense of UnpredictabilityUncontrollability “Coping” Action Tendencies and Associated Arousal

Stress

Anxiety

Depressive

Disorders

Disorders

Disorders

Yes

Yes

Yes

Yes Yes

Yes Yes

Yes Yes

Yes No

Yes Yes

Yes Yes

Yes

Yes

No

Note: From Anxiety and Its Disorders:The Nature and Treatment of Anxiety and Panic (p. 284) by D.H. Barlow, 1988, New York: Guilford. Copyright 1988 by Guilford Press. Adapted by permission.

present with marked degrees of dysphoria (dysthymia and anxiety). Much as with Klein et al.’s (1988) dysthymics, these dysphoric manics were generally more severe than clients with manic episodes without dysphoria in that they demonstrated a significantly greater number of previous hospitalizations and displayed less rapid cycling. They also responded less well to treatment. Furthermore, much as with basic emotions of fear and sadness/distress, recent data indicate that manic reactions can be conditioned to a variety of internal and external events or stimuli (Post, Rubinow, & Ballenger, 1986). This would suggest that much as with fear, sadness, and anger, a time-limited basic emotion of excitement can emerge frequently and inappropriately, and that it can be perceived as unpredictable and out of one’s control, resulting in the emergence of negative affect (anxiety, dysthymia) focused on manic episodes. The phenomenon of dysphoric mania completes the better part of a circumplex of emotions associated with emotional disorders. What is common to each of these disorders is a discrete emotion occurring unexpectedly or inappropriately in a manner that is experienced as at least somewhat out of control. This model of emotional disorders is depicted in Figure 13.5. Clearly, this is a preliminary statement of a model of disorders of emotion. At many points the model is summarized, although fuller elaboration can be found elsewhere (Barlow, 1988). As a summary, it is characterized by oversimplification. If this model is to advance, the complexities must be accounted for

Disorders of Emotion 221

Figure 13.5 Model of Emotional Disorders.

and data must be produced to address the many connections that at present are not supported by data. Nevertheless, a comprehensive theory of disorders of emotion emanating from emotion theory has obvious appeal and should do much to deepen our understanding of emotional disorders. I hope this model is a step in that direction. Note 1

An earlier rendition of this article was presented as an invited address to the Society for Research in Psycho-pathology, November 1988, Cambridge, Massachusetts, and to the Association for Advancement of Behavior Therapy, November 1989, Washington D.C. My appreciation to colleagues at these meetings for feedback and to my students and colleagues in Albany for their input. Evidence on the heritability of anxiety and panic is strong in some instances but subject to various confounds and complexities. See Barlow (1988) for a review.

References Abramson, L. Y., Metalsky, G. I., & Alloy, L. B. (1989). Hopelessness and depression: A theory based subtype of depression. Psychological Review, 96, 358–392. Adler, C. M., Craske, M. G., & Barlow, D. H. (1987, November). The use of modified relaxation in the experimental induction of anxiety and panic. Paper presented at the meeting of the Association for the Advancement of Behavior Therapy, Boston. Akiskal, H. S. (1983). Dysthymic disorder: Psychopathology of proposed chronic depressive subtypes. American Journal of Psychiatry, 140, 11–20. Alloy, L. B., Abramson, L. Y., & Viscusi, D. (1981). Induced mood and the illusion of control. Journal of Personality and Social Psychology, 41, 1129–1140.

222 David Harrison Barlow Alloy, L. B., Kelly, K. A., Mineka, S., & Clements, C. M. (1990). Comorbidity of anxiety and depressive disorders: A helplessness-hopelessness perspective. In J. D. Maser & C. R. Cloninger (Eds.), Comorbidity of mood and anxiety disorders (pp. 499–543). Washington D.C.: American Psychiatric Press. American Psychiatric Association. (1987). Diagnostic and statistical manual of mental disorders (3rd ed., rev.). Washington D.C.: Author. Avery, D. H., Osgood, T. B., Ishiki, D. M., Wilson, L. G., Kenny, M., & Dunnar, D. L. (1985). The DST in psychiatric outpatients with generalized anxiety disorder, panic disorder, or primary affective disorder. American Journal of Psychiatry, 142, 844–848. Ballenger, J. C, Burrows, G. D., DuPont, R. L., Lesser, I. M., Noyes, R., Pecknold, J. C., Rifkin, A., & Swinson, R. P. (1988). Alprazolam in panic disorder and agoraphobia: Results from a multicenter trial: I. Efficacy in short term treatment. Archives of General Psychiatry, 45, 423–428. Barlow, D. H. (1983, October). The classification of anxiety disorders. Paper presented at the conference DSM-HI: An Interim Appraisal, sponsored by the American Psychiatric Association, Washington D.C. Barlow, D. H. (1985). The dimensions of anxiety disorders. In A. H. Tuma & J. D. Maser (Eds.), Anxiety and the anxiety disorders (pp. 479–500). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Barlow, D. H. (1988). Anxiety and its disorders: The nature and treatment of anxiety and panic. New York: Guilford. Barlow, D. H., & Craske, M. G. (1990). “Unexpected” panic and DSM–IV. Unpublished DSM–IV position paper, American Psychiatric Association, Washington D.C. Barlow, D. H., Di Nardo, P. A., Vermilyea, B. B., Vermilyea, J. A., & Blanchard, E. B. (1986). Co-morbidity and depression among the anxiety disorders: Issues in diagnosis and classification. Journal of Nervous and Mental Disease, 174, 63–72. Barlow, D. H., Vermilyea, J. A., Blanchard, E. B., Vermilyea, B. B., Di Nardo, P. A., & Cerny, J. A. (1985). The phenomenon of panic. Journal of Abnormal Psychology, 94, 320–328. Beck, A. T. (1988). Cognitive approaches to panic disorder: Theory and therapy. In S. Rachman & J. D. Maser (Eds.), Panic: Psychological perspectives (pp. 91–109). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Beck, A. T., & Emery, G. (1985). Anxiety disorders and phobias: A cognitive perspective. New York: Basic. Benshoof, B. G. (1987). A comparison of anxiety and depressive symptomatology in the anxiety and affective disorders. Unpublished doctoral dissertation, State University of New York at Albany. Biaggio, M. K. (1987). Therapeutic management of anger. Clinical Psychology Review, 7, 663–675. Borkovec, T. D., Shadick, R., & Hopkins, M. (in press). The nature of normal and pathological worry. In R. Rapee & D. H. Barlow (Eds.), Chronic anxiety and generalized anxiety disorder. New York: Guilford. Breier, A., Charney, D. S., & Heninger, G. R. (1985). The diagnostic validity of anxiety disorders and their relationship to depressive illness. American Journal of Psychiatry, 142, 787–797. Brown, T. A., & Cash, T. F. (1990). The phenomenon of nonclinical panic: Parameters of panic, fear and avoidance. Journal of Anxiety Disorders, 4, 15–29. Cannon, W. B. (1927). Bodily changes in pain, hunger, fear and rage (2nd ed.). New York: Appleton-Century-Crofts. Charney, D. S., & Heninger, G. R. (1986). Abnormal regulation of noradrenergic function in panic disorders. Archives of General Psychiatry, 43, 1042–1054. Chesney, M. A. (1985). Anger and hostility: Future implications for behavioral medicine. In M. A. Chesney & R. H. Rosenman (Eds.), Anger and hostility in cardiovascular and behavioral disorders (pp. 277–290). New York: Hemisphere.

Disorders of Emotion 223 Chesney, M. A. (1986, November). Type A behavior. Keynote address presented at the meeting of the Association for the Advancement of Behavior Therapy, Chicago. Clark, D. M. (1988). A cognitive model of panic attacks. In S. Rachman & J. D. Maser (Eds.), Panic: Psychological perspectives (pp. 71–89). Hillsdale: NJ: Lawrence Erlbaum Associates, Inc. Cloninger, C. R. (1986). A unified biosocial theory of personality and its role in the development of anxiety states. Psychiatric Development, 3, 167–226. Cohen, A. S., Barlow, D. H., & Blanchard, E. B. (1985). The psycho-physiology of relaxation associated panic attacks. Journal of Abnormal Psychology, 94, 96–101. Coryell, W., Noyes, R., Clancy, J., Crowe, R., & Chaudhry, D. (1985). Abnormal escape from dexamethasone suppression in agoraphobia with panic attacks. Psychiatry Research, 15, 301–311. Crowe, R. R., Noyes, R., Pauls, D. L., & Slymen, D. J. (1983). A family study of panic disorder. Archives of General Psychiatry, 40, 1065–1069. Di Nardo, P. A., & Barlow, D. H. (1990). Syndrome and symptom comorbidity in the anxiety disorders. In J. D. Maser & C. R. Cloninger (Eds.), Comorbidity of mood and anxiety disorders (pp. 205–230). Washington D.C.: American Psychiatric Press. Dobson, K. S. (1985). The relationship between anxiety and depression. Clinical Psychology Review, 5, 307–324. Ehlers, A., & Margraf, J. (in press). The psychophysiological model of panic attacks. In P. M. G. Emmelkamp (Ed.), Anxiety disorders, Annual series of European research in behavior therapy. Vol. 4. Amsterdam: Swets. Eysenck, H. J. (1967). (Ed.). The biological basis of personality. Springfield, IL: Thomas. Eysenck, H. J. (1970). The structure of human personality. London: Methuen. Eysenck, H. J. (1981). (Ed.). A model for personality. New York: Springer-Verlag. Fridlund, A. J., Hatfield, M. E., Cottam, G. L., & Fowler, J. C. (1986). Anxiety and striate-muscle activation: Evidence from electromyographic pattern analysis. Journal of Abnormal Psychology, 95, 228–236. Gotlib, L. H. (1984). Depression and general psychopathology in university students. Journal of Abnormal Psychology, 93, 19–30. Gotlib, I. H., & Cane, D. B. (1989). Self-report assessment of depression and anxiety. In P. C Kendall & D. Watson (Eds.), Anxiety and depression: Distinctive and overlapping features (pp. 131–169). New York: Academic. Gray, J. A. (1982). The neuropsychology of anxiety. New York: Oxford University Press. Gray, J. A. (1985). Issues in the neuropsychology of anxiety. In A. H. Tuma & J. D. Maser (Eds.), Anxiety and the anxiety disorders (pp. 5–25). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Hallam, R. S. (1985). Anxiety: Psychological perspectives on panic and agoraphobia. New York: Academic. Hamilton, M. (1959). The assessment of anxiety states by rating. British Journal of Medical Psychology, 32, 50–55. Hamilton, M. (1960). A rating scale for depression. Journal of Neurology, Neurosurgery and Psychiatry, 23, 56–62. Heimberg, R. G., Vermilyea, J. A., Dodge, C. S., Becker, R. E., & Barlow, D. H. (1987). Attributional style, depression, and anxiety: An evaluation of the specificity of depressive attributions. Cognitive Therapy and Research, 11, 537–550. Izard, C. E. (1977). (Ed.). Human emotions. New York: Plenum. Izard, C. E., & Blumberg, M. A. (1985). Emotion theory and the role of emotions in anxiety in children and adults. In A. H. Tuma & J. D. Maser (Eds.), Anxiety and the anxiety disorders (pp. 109–129). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Kahn, R. J., McNair, D. M., Lipman, R. S., Covi, L., Rickeis, K., Downing, R., Fisher, S., & Frankenthaler, L. M. (1986). Imipramine and clordiazepoxide in depressive and anxiety disorders. Archives of General Psychiatry, 143, 1172–1173.

224 David Harrison Barlow Keller, M. B. & Lavori, P. W. (1984). Double depression, major depression and dysthymia: Distinct entities or different phases of a single disorder? Psychopharmacology Bulletin, 20, 399–402. Keller, M. B., Lavori, P. W., Endicott, J., Coryell, W., & Klerman, G. L. (1983). “Double depression”: Two-year follow-up. American Journal of Psychiatry, 140, 689–694. Keller, M. B., & Shapiro, R. W. (1982). “Double depression”: Superimposition of acute depressive episodes on chronic depressive disorders. American Journal of Psychiatry, 139, 438–442. Kendall, P. C., & Watson, D. (Eds.). (1989). Anxiety and depression: Distinctive and overlapping features. San Diego: Academic. Klein, D. F., Rabkin, J. G., & Gorman, J. M. (1985). Etiological and pathophysiological inferences from the pharmacological treatment of anxiety. In A. H. Turna & J. D. Maser (Eds.), Anxiety and the anxiety disorders (pp. 501–532). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Klein, D. N., Taylor, E. B., Dickstein, S., & Harding, K. (1988). Primary early-onset dysthymia: Comparison with primary non-bipolar non-chronic major depression on demographic, clinical, familial, personality, and socioenvironmental characteristics and short-term outcome. Journal of Abnormal Psychology, 97, 387–398. Lang, P. J. (1979). A bio-information theory of emotional imagery. Psychophysiology, 16, 495–512. Lang, P. J. (1984). Cognition in emotion: Concept and action. In C. Izard, J. Kagan, & R. B. Zajonc (Eds.), Emotion, cognition and behavior (pp. 192–225). New York: Cambridge University Press. Lang, P. J. (1985). The cognitive psychophysiology of emotion: Fear and anxiety. In A. H. Turna & J. D. Maser (Eds.), Anxiety and the anxiety disorders (pp. 131–170). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Leckman, J. F., Weissman, M. M., Merikangas, K. R., Pauls, D. L., & Prusoff, B. A. (1983). Panic disorder and major depression. Archives of General Psychiatry, 40, 1055–1060. Lee, M. A., & Cameron, O. G. (1986). Anxiety disorders, Type A behavior, and cardio-vascular disease. International Journal of Psychiatry in Medicine, 16, 123–129. Liddell, H. S. (1949). The role of vigilance in the development of animal neurosis. In P. Hoch & J. Zubin (Eds.), Anxiety (pp. 183–197). New York: Grune & Stratton. MacLeod, C., Mathews, A., & Tata, P. (1986). Attentional bias in emotional disorders. Journal of Abnormal Psychology, 95, 15–20. Mandler, G. (1966). Anxiety. In D. L. Sills (Ed.), International encyclopedia of the social sciences. New York: Macmillan. Mattes, J. A. (1986). Psychopharmacology of temper outbursts: A review. Journal of Nervous and Mental Disease, 174, 464–470. Mavissakalian, M. (1987). Initial depression and response to imipramine in agoraphobia. Journal of Nervous and Mental Disease, 175, 358–361. Mavissakalian, M., & Michelson, L. (1986). Agoraphobia: Relative and combined effectiveness of therapist-assisted in vivo exposure and imipramine. Journal of Clinical Psychiatry, 47, 117–122. McCauley, P. A., Di Nardo, P. A., & Barlow, D. H. (1988, November). Cluster analysis application in anxiety disorders. Poster session conducted at the meeting of the Association for the Advancement of Behavior Therapy, Washington D.C. Miller, I. W., Norman, W. H., & Dow, M. G. (1986). Psychosocial characteristics of “double depression.” American Journal of Psychiatry, 143, 1042–1044. Mineka, S. (1985a). Animal models of anxiety-based disorders: Their usefulness and limitations. In A. H. Turna & J. D. Maser (Eds.), Anxiety and the anxiety disorders (pp. 199–244). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Mineka, S. (1985b). The frightful complexity of the origins of fears. In Bruch & J. B. Overmier (Eds.), Affect, conditioning, and cognition: Essays on the determinants of behavior (pp. 55–73). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.

Disorders of Emotion 225 Mineka, S., Gunnar, M., & Champoux, M. (1986). Control and early socioemotional development: Infant rhesus monkeys reared in controllable versus uncontrollable environments. Child Development, 57, 1241–1256. Mineka, S. & Kihlstrom, J. (1978). Unpredictable and uncontrollable aversive events. Journal of Abnormal Psychology, 87, 256–271. Norton, G. R., Dorward, J., & Cox, B. J. (1986). Factors associated with panic attacks in nonclinical subjects. Behavior Therapy, 17, 239–252. Norton, G. R., Harrison, B., Hauch, J., & Rhodes, L. (1985). Characteristics of people with infrequent panic attacks. Journal of Abnormal Psychology, 94, 216–221. Novaco, R. W. (1975). Anger control: The development and evaluation of an experimental treatment. Lexington, MA: Heath. Nutt, D. J. (1986). Increased central alpha adrenoceptor sensitivity in panic disorder. Psychopharmacology, 90, 268–269. Öst, L. G., Lindahl, I. L., Sterner, U., & Jerremalm, A. (1984). Exposure in vivo vs. applied relaxation in the treatment of blood phobia. Behavior Research and Therapy, 22, 205–216. Peterson, C., Semmel, A., von Baeyer, C., Abramson, L. Y., Metalsky, G. I., & Seligman, M. E. P. (1982). The Attributional Style Questionnaire. Cognitive Therapy and Research, 6, 287–299. Post, R. M., Rubinow, D. R., & Ballenger, J. C. (1986). Conditioning and sensitization in the longitudinal course of affective illness. British Journal of Psychiatry, 149, 191–201. Post, R. M., Rubinow, D. R., Uhde, T. W., Roy-Byrne, P., Linnoila, M., Rosoff, A., & Cowdry, R. (1989). Dysphoric mania: Clinical and biological correlates. Archives of General Psychiatry, 46, 353–358. Puig-Antich, J., & Rabinovich, H. (1986). Relationship between affective and anxiety disorders in childhood. In R. G. Helman (Ed.), Anxiety disorders of childhood (pp. 136–156). New York: Wiley. Rachman, S., & Levitt, K. (1985). Panics and their consequences. Behaviour Research and Therapy, 23, 585–600. Rapee, R. (1985). A case of panic disorder treated with breathing retraining. Journal of Behavioral and Experimental Psychiatry, 16, 63–65. Rapee, R., Ancis, J., & Barlow, D. H. (1987). Emotional reactions to physiological sensations: Comparison of panic disorder and nonclinical subjects. Behaviour Research and Therapy, 26, 265–269. Reiman, E. M., Raichle, M. E., Robins, E., Butler, F. K., Herscovitch, P., Fox, D., & Perlmutter, J. (1986). The application of position emission tomography to the study of panic disorder. American Journal of Psychiatry, 143, 469–476. Rickels, K., Feighner, J. P., & Smith, W. T. (1985). Alprazolam and amitriptyline, doxepin and placebo in the treatment of depression. Archives of General Psychiatry, 42, 134–141. Riskind, J. H., Beck, A. T., Brown, G., & Steer, R. A. (1987). Taking the measure of anxiety and depression: Validity of the reconstructed Hamilton scale. Journal of Nervous and Mental Disease, 175, 474–479. Roth, M., Gurney, C., Garside, R. F., & Kerr, T. A. (1972). Studies in the classification of affective disorders: Relationship between anxiety states and depressive illness—I. British Journal of Psychiatry, 121, 147–161. Rotter, J. B. (1954). Social learning and clinical psychology. Englewood Cliffs, NJ: Prentice-Hall. Salge, R. A., Beck, J. G., & Logan, A. C. (1988). A community survey of panic. Journal of Anxiety Disorders, 2, 157–167. Sanderson, W. C, Di Nardo, P. A., Rapee, R. M., & Barlow, D. H. (1990). Syndrome comorbidity in patients diagnosed with a DSM-III-Revised anxiety disorder. Journal of Abnormal Psychology, 99, 308–312.

226 David Harrison Barlow Sanderson, W. C., Rapee, R. M., & Barlow, D. H. (1989). The influence of an illusion of control on panic attacks induced via inhalation of 5.5% carbon dioxide-enriched air. Archives of General Psychiatry, 46, 157–164. Schweizer, E. E., Swenson, C. M., Winokur, A., Rickels, K., & Maislin, G. (1986). The dexamethasone suppression test in generalized anxiety disorder. British Journal of Psychiatry, 149, 320–322. Seligman, M. E. P. (1975). Helplessness. San Francisco: Freeman. Spielberger, C. D., Gorsuch, R. L., & Lushene, R. E. (1970). Manual for the State-Trait Anxiety Inventory. Palo Alto, CA: Consulting Psychologists Press. Spielberger, C. D., Johnson, E. H., Russell, S. F., Crane, R. J., Jacobs, G. A., & Worden, T. J. (1985). The experience and expression of anger: Construction and validation of an anger expression scale. In M. A. Chesney & R. H. Rosenman (Eds.), Anger and hostility in cardiovascular and behavioral disorders (pp. 5–30). New York: Hemisphere/McGraw-Hill. Spielberger, C. D., Krasner, S. S., & Solomon, E. P. (1988). The experience, expression, and control of anger. In M. P. Janisse (Ed.), Health psychology: Individual differences and stress. New York: Springer-Verlag. Taylor, C. B., Sheikh, J., Agras, W. S., Roth, W. T., Margraf, J., Ehlers, A., Maddock, R. J., & Gossard, D. (1986). Self-report of panic attacks: Agreement with heart rate changes. American Journal of Psychiatry, 143, 478–482. Teasdale, J. D. (1985). Psychological treatments for depression: How do they work? Behaviour Research and Therapy, 23, 157–165. Tellegen, A. (1985). Structures of mood and personality and their relevance to assessing anxiety, with an emphasis on self-report. In A. H. Tuma & J. D. Maser (Eds.), Anxiety and the anxiety disorders (pp. 681–706). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Torgersen, S. (1983). Genetic factors in anxiety disorders. Archives of General Psychiatry, 40, 1085–1089. Watson, D., Clark, L. A., & Carey, G. (1988). Positive and negative affectivity and their relation to anxiety and depressive disorders. Journal of Abnormal Psychology, 97, 346–353. Weissman, M. M. (1985). The epidemiology of anxiety disorders: Rates, risks and familial patterns. In A. H. Turna & J. D. Maser (Eds.), Anxiety and the anxiety disorders (pp. 275–296). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Wittchen, H. U. (1986). Epidemiology of panic attacks and panic disorders. In I. Hand & H. U. Wittchen (Eds.), Panic and phobias: Empirical evidence of theoretical models and longterm effects of behavioral treatments (pp. 18–28). New York: Springer-Verlag. Yerkes, R. M., & Dodson, J. D. (1908). The relation of strength of stimulus to rapidity of habit-formation. Journal of Comparative Neurology and Psychology, 18, 459–482. Zillmann, D. (1983). Arousal and aggression. In R. G. Geen & E. Donnerstein (Eds.), Aggression: Theoretical and empirical reviews (Vol. 1). New York: Academic. Zung, W. W. (1965). A self-rating depression scale. Archives of General Psychiatry, 12, 63–70.

Article 14

The Development of Anxiety: The Role of Control in the Early Environment Bruce F. Chorpita University of Hawaii at Manoa

David H. Barlow Boston University

Current developments in cognitive and emotion theory suggest that anxiety plays a rather central role in negative emotions. This article reviews findings in the area of anxiety and depression, helplessness, locus of control, explanatory style, animal learning, biology, parenting, attachment theory, and childhood stress and resilience to articulate a model of the environmental influences on the development of anxiety. Evidence from a variety of sources suggests that early experience with diminished control may foster a cognitive style characterized by an increased probability of interpreting or processing subsequent events as out of one’s control, which may represent a psychological vulnerability for anxiety. Implications for research are discussed. Historically, studies of childhood and adult anxiety and depression have been characterized by a discontinuity between major theoretical frameworks, methodologies, and research paradigms particular to each area. Recently, however, theoretical advances in the understanding of both childhood and adult anxiety and depression are beginning to highlight consistencies and to allow the emergence of a more unified model. For example, in recent adult theories, the dimensional nature of pathological syndromes, the relation of normal to abnormal processes, the multiplicity and interaction of psychosocial and biological influences, and the continuity of anxious and depressive features have received increased emphasis (Alloy, Kelly, Mineka, & Clements, 1990; Barlow, 1991). Findings from outside the clinical literature have also contributed to the integration and extension of childhood and adult models of anxiety. For example, conditioning models (Mineka, 1985; Mineka & Zinbarg, 1996) and biopsychological models (Gray, 1982, 1987; Gray & Mc-Naughton, 1996) have become increasingly relevant to traditional developmental notions of attachment (e.g., Bowlby, 1969), inhibition (e.g., Kagan, 1989), and coping and resilience (Garmezy,

228 Bruce F. Chorpita and David H. Barlow

1986; Hetherington, 1989; Masten, Best, & Garmezy, 1991). What is beginning to emerge is a model of negative emotions that highlights the role of uncontrollability and unpredictability and, while acknowledging the contribution of innate vulnerabilities in the experience of anxiety, also emphasizes the importance of early experience in the development and progression of these vulnerabilities. By reviewing selected findings in these diverse areas, this article is intended to explicate the notion that early experience with reduced control can foster a psychological diathesis that may eventually give rise to increased anxiety (and perhaps depression) in children and adults. We begin with some definitions, A Model of Negative Emotions Contemporary understanding of the relation between anxious and depressed emotion has become increasingly complex. Several lines of evidence now suggest that anxiety may play a rather central role in negative emotions (Barlow, Chorpita, & Turovsky, 1996; Brown, Chorpita, & Barlow, 1998; Chorpita, Albano, & Barlow, 1998; Gray & McNaughton, 1996). Gray (1982; Gray & McNaughton, 1996) defined anxiety as a state of the conceptual or central nervous system characterized by activity of the behavioral inhibition system (BIS). The BIS is, in turn, defined as a functional brain system involving the septal area, the hippocampus, and the Papez circuit (including also the neo-cortical inputs to the septo-hippocampal system, dopaminergic ascending input to the prefrontal cortex, cholinergic ascending input to the septo-hippocampal system, noradrenergic input to the hypothalamus, and the descending noradrenergic fibers of the locus coeruleus [Gray, 1982]). The outputs of this system, then, can be taken as evidence of anxious emotion. Gray defines the primary, short-term outputs as involving narrowing of attention, inhibition of gross motor behavior, increased stimulus analysis, increased exploration of environment (e.g., scanning), and priming of hypothalamic motor systems for possible rapid action that may be required (i.e., possible activation of the fight or flight system [FFS]). Here, anxiety is considered distinct from the emotion of fear and panic, which is functionally related to actual confrontation with danger, not simply the detection of and preparation for danger. In contrast to anxiety, fear is conceptualized as activity of the FFS and is characterized by surges of autonomic arousal and the associated action tendencies of escape, active avoidance, or defensive aggression. Anxiety and Depression

Evidence from a number of sources suggests that, when defined in such a manner, anxiety may actually be a common component of both anxiety disorders and depressive disorders (Barlow et al., 1996; Brown et al., 1998; Chorpita et al., 1998; Clark & Watson, 1991; Gray & McNaughton, 1996). For example, Clark and Watson (1991; Watson & Clark, 1984) articulated a general factor, Negative Affect, common to the self-reported signs and symptoms of both anxious and depressed emotion. In Clark and Watson’s

The Development of Anxiety 229

model, this construct of negative affect bears a striking similarity to Gray’s definition of anxiety (i.e., BIS activity). (A complex issue of semantics arises here in that Clark and Watson consider anxiety to involve both negative affect and physiological hyperarousal, whereas Gray’s [1982] and Barlow’s [1988] models view anxiety as distinct from autonomic arousal, and thus more synonymous with Clark and Watson’s [1991] “negative affect”). Although a thorough explication of the taxonomy and structure of negative emotions is beyond the scope of this article, it is an important premise that a general emotional factor, consistent with Gray’s definition of anxiety, functions as a common component of anxiety disorders and depressive disorders. This has been supported by a number of recent findings concerning the structure of negative emotions (e.g., Brown, Chorpita, Korotitsch, & Barlow, 1997; Chorpita et al., 1998; Joiner, Catanzaro, & Laurent, 1996; Lovibond & Lovibond, 1995). Very recent evidence strongly supports the notion that this general factor does indeed influence anxiety disorders and depressive disorders, and that this factor is characterized by anxious apprehension, narrowed attention, and reduced autonomic reactivity (see Borkovec, 1994). Specifically, Brown et al. (1998) found in a sample of 350 adult outpatients that covariance among dimensions of selected Diagnostic and Statistical Manual of Mental Disorders (DSM-IV; American Psychiatric Association, 1994) anxiety and mood disorders was structurally related to three higher order factors (e.g., Clark & Watson, 1991). One higher order factor demonstrated a uniformly important influence on all of the DSM-IV syndromes, and it was highly consistent with Watson and Clark’s (1984) negative affect as well as Gray’s (1982) and Barlow’s (1988) definitions of anxiety. The syndrome of generalized anxiety disorder was observed to load most highly on this particular factor, suggesting that anxiety (i.e., Gray’s behavioral inhibition) may be a core factor among the disorders of emotion. Related Theoretical Models

Theoretical developments in other areas are consistent with this model. For example, Alloy et al. (1990), approaching some of the same theoretical issues from the perspective of depression, have articulated a comparable point of view (see also Abramson, Metalsky, & Alloy, 1989). Alloy et al. asserted a cognitive model of depression and anxiety that posited a continuum of proximal causality for these syndromes. As such, one dimension on which these disorders or syndromes are said to vary involves the individual’s degree of a sense of control. Specifically, when an individual experiences uncertainty about the ability to control outcomes (i.e., “uncertain helplessness”), the resulting affective state is one of “aroused anxiety.” If this ostensible lack of control increases (i.e., “certain helplessness”), one experiences a state of “mixed anxiety-depression.” Finally, when an individual’s sense of control is entirely diminished (i.e., “hopelessness”) and there is certainty of a negative outcome, one experiences a depressive state (Alloy et al., 1990, pp. 525–526). This cognitive model of helplessness-hopelessness serves to explain many

230 Bruce F. Chorpita and David H. Barlow

of the observed similarities of anxious and depressive syndromes, and it is consistent with the idea of a shared vulnerability for anxiety disorders and depressive disorders. Alloy et al. (1990) reviewed a number of important findings that further support the notion of anxiety as a core component of the emotional disorders. For example, the sequential relation between anxiety and depressive disorders is characterized by a temporal asymmetry, such that anxiety disorders are more likely to precede depressive disorders than to follow (e.g., Angst, Vollrath, Merikangas, & Ernst, 1990). Similar findings have emerged in cross-sectional comorbidity research. That is, cases of anxiety disorders without depression are commonly observed, but cases of depression without anxiety are relatively rare (e.g., Di Nardo & Barlow, 1990; Dobson, 1985). Collectively, these patterns highlight the primacy of anxiety among the disorders of emotion and implicate anxiety as a general risk factor (Barlow et al., 1996). Influences on Anxiety

Let us then consider the events that activate this emotion of anxiety. Gray (1982) states that the BIS responds to signals for punishment, signals for frustrative nonreward, and novel stimuli. These inputs are mediated by what Brooks (1986) and Gray and McNaughton (1996) call the “comparator,” a subsystem involving the Papez loop (subicular area, mammilary bodies, anteroventral hypothalamus, and cingulate cortex). The comparator analyzes information from a number of sources and, based on these analyses, regulates BIS activity. Principally, the sources of compared information are (a) the current observed state of the world, (b) the next planned step in the motor program, (c) stored regularities about the world (stimulus-stimulus associations as determined by Pavlovian conditioning), and (d) stored regularities about the behavior-outcome relations (stimulus-behavior-stimulus associations as determined by instrumental conditioning; see Figure 14.1). According to Gray and McNaughton (1996), the comparator “has the task of predicting the next sensory event to which the animal will be exposed and checking whether it actually does occur; of operating the outputs of the BIS either if there is a mismatch between the actual and predicted events or if the predicted event is aversive; and of testing out alternative strategies (including alternative multidimensional descriptions of stimuli and/or responses) which may overcome the difficulty with which the animal is faced” (p. 75). What is important with respect to this article is that BIS activity (i.e., anxiety) is very much a function of the stored regularities associated with both Pavlovian and instrumental conditioning history, as established during early development. Let us now examine how these may be related to a history of low control. Control and Anxiety Definitions of control, its putative influence on anxiety, and its relation to similar constructs such as predictability and self-efficacy have been argued extensively elsewhere (e.g., Biglan, 1987; Minor, Dess, & Overmier, 1991). For example, Minor et al., reviewing animal research, have argued that lack

The Development of Anxiety 231 Observed State of the World

Stored Regularities (Pavlovian)

Next Step in Motor Program

Stored Regularities (Instrumental)

Comparator Subsystem Novelty or Aversives

Behavioral Inhibition System

Anxiety Inhibition Figure 14.1 Schematic of Gray and McNaughton’s (1996) Model of the Behavioral Inhibition System, Detailing the Function and Operation of the Comparator Subsystem.

of control (i.e., the inability to influence events) is one of a number of pathways to fear and anxiety. For the purposes of this article, control is broadly defined as the ability to personally influence events and outcomes in one’s environment, principally those related to positive or negative reinforcement. This definition overlaps with “prediction” in that control, as such, can implicitly allow prediction of when something will happen, such as the termination of an aversive event. This definition is also consistent with the theorizing of Weisz (1986), who has emphasized that it is not only important that one perceives outcomes as contingent on the behavior of others in general but also whether one perceives outcomes as contingent on one’s own behavior. Given such a definition of control, there exists a diversity of literature supporting the notion that an immediate sense of diminished control is commonly associated with the immediate expression of anxiety (Barlow, 1988, 1991; Beck & Emery, 1985; Lazarus, 1966, 1968; Lazarus, Averill, & Opton, 1970; Mandler, 1972; Sanderson, Rapee, & Barlow, 1989). In terms of Gray’s BIS, lower control over a threatening stimulus would increase the expected probability of danger, given that fewer alternative strategies for avoidance

232 Bruce F. Chorpita and David H. Barlow

would be accessed by the comparator. Hence, evidence of low control would likely increase anxiety (i.e., BIS activity) in the immediate sense. What is more compelling, however, is the idea that a history of lack of control may put individuals at eventual risk to experience chronic anxiety or related negative emotional states through the development of a psychological vulnerability. Specifically, converging evidence suggests that sufficient early experience with uncontrollable events may eventually lead to an increased generalized tendency to perceive or process events as not within one’s control (see Schneewind, 1995). In terms of the effects on anxiety, it seems possible that the conditioning history involving control over (positive and negative) reinforcement would determine the nature of the stored regularities involved in the comparator subsystem. That is, an individual reared with control over these events will have relatively greater access to stored information that predicts the possibility of avoiding punishment or nonreward (mainly through the so-called instrumental regularities; see Figure 14.1). Conversely, experiencing diminished control over events during development may establish stored regularities that more commonly result in the comparator predicting an aversive outcome. Thus, in the face of identical inputs related to signals for punishment or nonreward, stored information related to a history of low control should result in heightened activity of the BIS, hence greater anxiety. In this manner, early experience with uncontrollable events may be thought of as a primary pathway to the development of anxiety in that such experience may foster an increased likelihood to process events as not within one’s control (i.e., a psychological vulnerability). In this way, it appears that early experience can be disproportionately important in that it weights or colors subsequent experience. Of course, this is not to say that experience in adulthood cannot instill or remove a sense of diminished control, only that early experiences contribute most heavily to the formation of this psychological vulnerability (see Rotter, 1966). Control in Animal Models Although one might expect much of the empirical support for these concepts to emanate from research on anxiety in humans, more frequently it is the animal literature that implicates experience with uncontrollable events in subsequently fostering stable anxious responding (Mowrer & Viek, 1948; Weiss, 1971a, 1971b). Much of the original anxiety work has been overshadowed by the study of learned helplessness (e.g., Overmier & Seligman, 1967; Seligman & Maier, 1967). Ironically, although this study of helplessness was originally conceptualized as an analogue for human depression, it has been subsequently described as perhaps the most useful psychological animal model of anxiety in humans (Barlow, 1988; Mineka, 1985; Mineka & Zinbarg, 1996). What has emerged from the collective lines of research is the understanding that the tendency of events to subsequently trigger some analogue of anxiety or negative affect is dependent to some degree on the amount of control the organism experiences over those events.

The Development of Anxiety 233

In a review of this literature, Maier and Seligman (1976) outlined the idea that events uncontrollable to an organism produce subsequent motivational, cognitive, and emotional disturbance. The best known of these experiments involved the classic learned helplessness manipulation used by Overmier and Seligman (1967), during which a dog is repeatedly exposed to inescapable shock and then fails to escape when escape is made possible. It has further been shown that these performance deficits are not related to shock or aversive stimulation, but rather they are a result of the uncontrollability of the shock (e.g., Maier, 1970; Seligman & Maier, 1967). That is, in classic evaluations of the effects of uncontrollable aversive stimulation (in addition to a standard control group), a group with control over aversive stimulation is added, to which the experimental group is yoked, thus equating the total amount of aversive stimulation. Experiments with rats have shown similar evidence of induced helplessness, albeit using more challenging tasks than simply escape (e.g., Maier & Testa, 1975; Seligman, Rosellini, & Kozak, 1975). Subsequent alternatives to learned helplessness theory have emerged (e.g., Minor, Dess, & Overmier, 1991; Overmier, 1988), and although these models emphasize parameters and mechanisms other than response-shock noncontingency (as in learned helplessness), they acknowledge the importance of experience with lack of mastery, predictability, or control as contributory to anxious responding. With respect to learned helplessness, Maier and Seligman (1976) argued that there are profound emotional consequences of lack of control in animals. Some authors have suggested that these effects bear a striking similarity to what is known about chronic anxiety (e.g., Barlow, 1988; Mineka, 1985). In fact, Drugan, Ryan, Minor, and Maier (1984) found that antianxiety drugs administered prior to exposure to uncontrollable stress can prevent subsequent learned helplessness effects in rats. Examining the relation of uncontrollable threat to physiology, Weiss (1971a, 1971b, 1971c) demonstrated the ability of uncontrollable shock (relative to controllable shock) to produce increased cortisol secretion and gastric ulceration in rats. In one rather revealing study, Stroebel (1969) trained a group of rhesus monkeys to control various aversive stimuli in their chamber by means of a lever and, on removing that lever, witnessed dramatic changes in the monkeys’ behavior, including increased weakness, poor grooming, compulsive hand waving, and hair pulling. These behaviors may be an analogue for what humans experience as anxiety. Most recently, Peterson, Maier, and Seligman (1993) reviewed work showing that animals demonstrate heightened anxiety in novel situations following helplessness induction. Mineka, Gunnar, and Champoux (1986) substantially contributed to the understanding of anxiety in humans in their work with primates. Interestingly, their study investigated the effects of control over appetitive, not aversive, events during rearing. Eight of 20 infant rhesus monkeys were raised in conditions allowing their control over delivery of food, water, and treats. Another 8 monkeys received these stimuli noncontingently (yoked control). The remaining 4 monkeys were raised in a “standard rearing” control condition. The monkeys were raised in these environments for up to 12 months,

234 Bruce F. Chorpita and David H. Barlow

and the testing phase began at about the 8th month. During testing, monkeys reared with control were noted to habituate more quickly than the yoked and standard rearing groups when confronted with a mechanical toy robot and also to demonstrate more exploratory behavior in a novel playroom situation. In addition, the master group showed enhanced active coping responses during selected trials of separation from peers. These results seem particularly noteworthy as an analogue for human anxiety. Indeed, the notion that control over appetitive events mitigates a vulnerability for anxiety and expression of fear extends the framework beyond that of stressors causing anxiety to suggest that nearly all aspects of rearing may have an impact on the development of a psychological vulnerability, which can ultimately contribute to the expression of anxiety. Studies from a variety of other sources have suggested the tendency for uncontrollable stimuli to result in submissive or inhibited behavior related to anxiety (e.g., Sapolsky, 1989; Williams & Lierle, 1986; Williams & Scott, 1989). For example, in a series of studies of wild olive baboons, Sapolsky (1989; Sapolsky & Ray, 1989; Ray & Sapolsky, 1992) outlined a possible relation between control and the development of personality characteristics as well as biological functioning. Using male baboons from two troops in the wild, Sapolsky (1989) attempted to ascertain the physiological source of excessive cortisol secretion commonly associated with chronic stress. Glucocorticoids, hormones that are secreted during challenging situations, are found to be at higher basal levels in chronically stressed organisms (i.e., hypercortisolism). Sapolsky hypothesized that the physiological origin of hypercortisolism might involve impaired feedback sensitivity of the pituitary gland or central nervous system (not hypersensitivity of the adrenal glands). This hypothesis was influenced by previous findings concerning dexamethasone nonsuppression1 in anxious or depressed humans (e.g., Carroll, 1985), another indicator of glucocorticoid feedback insensitivity. In the course of identifying a possible central nervous system dysregulation associated with hypercortisolism, Sapolsky (1989) made some important interpretations involving the observed behavioral data from the troops. Specifically, hypercortisolism was found disproportionately among those baboons who were lower in the social rank. These baboons were frequent targets of displaced and unpredictable aggression and more disruption in attempted consortships with females. One might assume that glucocorticoid dysregulation might be the cause of impaired social functioning; however, other data do not support this contention. For example, social ranks commonly change over time in primate groups. It has been found that as animals’ social ranks change over time, changes in psychophysiological profiles of these animals will follow (Rose, Bernstein, Gordon, & Catlin, 1974). Although control over social rank was not directly manipulated, Sapolsky inferred that it was not sub-ordinance per se, but the experience of low control or instability that was responsible for hypercortisolism (Sapolsky & Ray, 1989). Sapolsky (1989) stated: It is not merely the rank, but the context in which the rank occurs. For example, the physiological correlates of dominance tend to be diametrically opposed depending on whether the hierarchy is stable or unstable.

The Development of Anxiety 235

In general, in a stable hierarchy in which dominant males are securely entrenched, the physiological profile of dominance is an adaptive one. . . . Amid an unstable hierarchy, all of those associations with dominance are lost. (p. 1050) In light of these data, Sapolsky (1989, p. 1049) inferred, “for the olive baboon, stressfulness [i.e., the tendency of stimuli to elicit anxious behavior] might best be considered to reflect low degrees of control or predictability about social circumstances.” The implication that experience with control can affect physiological profiles associated with “biological markers” (e.g., hypercortisolism) for human anxiety and related emotions (Fowles, 1995; Schweizer, Swenson, Winokur, Rickels, & Maislin, 1986) highlights the complex interplay of psychological and biological variables. Experience with lack of control may actually engender or exacerbate selected biopsychological risk factors that contribute to stable patterns in behavior. Continuing in this line of research, Ray and Sapolsky (1992) found that control over aversive social threats was not the only factor associated with disrupted glucocorticoid secretion profiles. Data from 41 male baboons suggested that control over appetitive social experience was also related significantly to normal cortisol functioning. Positive social affiliation and degree of sexual contact were associated with lower basal cortisol levels in the male baboons. Ray and Sapolsky inferred that this relation was not so much a function of dominance but rather a function of the degree to which the animal could successfully control access to these events (although, here again, control was never directly manipulated). Similar to the manner in which Mineka et al. (1986) documented the importance of control over appetitive events for influencing habituation to a frightening stimulus, so Ray and Sapolsky have suggested the applicability of control over appetitives to neurophysiological functioning associated with anxiety. The above findings appear to support the relation of uncontrollable events and the expression of anxiety, but the more interesting proposition involves identifying the role of a psychological vulnerability, characterized by chronic perception of events as not in the organism’s control. To that end, other findings by Sapolsky and colleagues do implicate the role of psychological variables as a possible moderator for the effects of stressful stimuli on the organism’s response. For example, male baboons who were better able to distinguish actual rival threats from neutral or mild threat cues also showed normal basal cortisol levels.2 Baboons who could not make the distinction and, therefore, perceived potential threat more commonly, evidenced high basal cortisol levels. This observation suggests that the (accurate or inaccurate) perception of uncontrollability (or unpredictability) may be more important than the degree of actual threat itself (e.g., Raab & Oswald, 1980; Sanderson, Rapee, & Barlow, 1989), emphasizing the importance of psychological factors in modifying the physiological dysregulation associated with stress. It seems that the ubiquitous and equivocal term stress might best be defined in terms of a “top-down” phenomenon, in that its consequences appear to be closely related to its mental construction or interpretation, in lower animals as well as humans.

236 Bruce F. Chorpita and David H. Barlow

Control in Children Locus of Control

Attempts to define or measure a sense of control in humans have a long history (e.g., Rotter, 1954). Rotter (1966) articulated a dimension of control as existing along a continuum from internal to external causality, that is, “locus of control.” According to theory, the degree to which one is reinforced by a stimulus is mediated by the direction of one’s attribution about the response-stimulus relation. Thus, locus of control is generally felt to represent the extent to which an individual perceives personal control over events in one’s environment. The first reliable and valid childhood measure of locus of control was developed by Nowicki and Strickland (Nowicki–Strickland Locus of Control Scale [NSLOC]; 1973). As a model of anxiety that highlights control might predict, external control scores on the NSLOC have been found to correlate with manifest anxiety scores within a clinical sample (r = 31; Finch & Nelson, 1974). More recently, Nunn (1988) observed the same relation in a sample of 267 students in Grades 5–8 (r = .31). Such findings have extended to childhood depression as well. For example, McCauley, Mitchell, Burke, and Moss (1988) noted significant differences between 47 children diagnosed with depression using a structured interview (KSADS; Puig-Antich & Chambers, 1978), 30 children whose depression had remitted within the past year, and 31 nondepressed psychiatric controls, with the current depression group showing the highest and the control group showing the lowest NSLOC scores. In addition, McCauley et al. (1988) found a correlation of .33 between the NSLOC and the Childhood Depression Inventory (CDI; Kovacs, 1981) across the three groups. It is notable that similar findings have emerged from clinically anxious, clinically depressed, and nonclinical children, and that these findings also parallel those in adult populations (e.g., Hoehn-Saric & McLeod, 1985).3 Although these correlations with anxiety may appear to be somewhat modest, recent arguments have emerged that locus of control may be somewhat different from the perceptions of control related to anxiety, in that locus of control is rather general and therefore less representative of the aspects of control that may be directly relevant to negative emotions (e.g., Rapee, Craske, Brown, & Barlow, 1996). For example, some investigations have suggested that Rotter’s construct is multidimensional (e.g., Klockars & Varnum, 1975), and Rapee et al. (1996) demonstrated in an adult sample that a more specific measure of control over threat and one’s reactions to threat (Anxiety Control Questionnaire; ACQ) was more highly correlated with measures of anxiety than was Rotter’s more general measure. This finding, however, is complicated by the fact that items on the ACQ are confounded with degree of exposure to threat; that is, an increase in the overall amount of threat should inflate scores on the ACQ (regardless of control), but a similar increase should have no immediate implications for Rotter’s construct. In that sense, the higher correlations of the ACQ with anxiety measures may partly be a problem with discriminant validity of the ACQ. In general, it appears that the measurement of perceived control as a vulnerability remains a target for empirical investigation, both in adults and in children (e.g., Skinner, Chapman, & Baltes, 1988; Weisz & Stipek, 1982).

The Development of Anxiety 237

Developmental Considerations

One important consideration concerning this idea of helplessness or uncontrollability in children is the timing and manner in which it might develop. Dweck and Leggett (1988) proposed a model of the development of helplessness in children that identified critical antecedents, involving the two dimensions of ability theory and goal orientation. Dweck and Leggett (1988; Burhans & Dweck, 1995) argued that children who perceive their abilities as flexible or variant are likely to be resistant to helplessness, owing to their implicit belief that improvement on a task may eventually be possible. Further, children who approach a goal as an opportunity for learning as opposed to evaluation were also more likely to show more of a mastery orientation. An important aspect of their model is that the more complex function of forming an “ability theory” may not be applicable to younger children, who may not have the capacity to form trait conceptions of their own or others’ abilities. This would suggest that the emergence of a sense of helplessness of control might not substantively emerge until middle childhood, a notion that runs somewhat counter to some of the animal findings that implicate very early development (e.g., Mineka et al., 1986). More recently, however, Burhans and Dweck (1995) have updated this model in the wake of accumulating evidence for helplessness patterns found in children as young as preschoolers (see Dweck, 1991). This expanded model now suggests that an individual need only possess a generalized sense of conditional self-worth. That is, to the extent that a child feels that outcomes reflect on whether she is “good” or “bad,” the capacity exists for helplessness to occur. When viewed together with Dweck and Leggett’s (1988) existing formulation, the expanded model predicts that the greatest likelihood for helplessness should arise when the child views an outcome as an indication of an invariant attribute of global self-worth. This suggests that although important conditions for helplessness may arise early in development (i.e., a sense of contingent self-worth), additional cognitive factors that arise later in development (e.g., conceptions of attribute or trait invariance) may intensify potential for helplessness. In general, research continues to evaluate and refine these conceptions of the transformation of control-related cognitions over development. In terms of our present model, continued investigation will likely inform many of the yet-unanswered questions about the developmental prerequisites that allow uncontrollable events to establish an enduring sense of diminished control. Attributional Style

If the goal is to outline this relation of events, cognitive style, and ultimately psychopathology over the course of development, it may be useful to look at a more well-developed literature in this regard, involving attributional style and depression. Recent work in this area has begun to illuminate the manner in which experience, cognition, and depression interact, and an examination of these findings raises some interesting hypotheses when reconsidered in the context of anxiety. Abramson, Seligman, and Teasdale (1978) introduced the construct of attributional style, arguing that one’s attributions about the causes of positive and

238 Bruce F. Chorpita and David H. Barlow

negative events played a critical role in the development of helplessness and depression. Unlike locus of control, attributional style was advanced at the outset as a multidimensional construct, with the first of those dimensions being internal versus external, the second being global versus specific, and the third being stable versus unstable. The first dimension of attributional style (internal–external) has actually occasioned some interesting points regarding locus of control. For example, Abramson and Sackeim (1977) pointed out that depressives are typically external in locus of control and yet are also characterized by self-blame. This has been described as a theoretical paradox, in that depressives see negative events as both out of their control and a result of their own actions, ideas that may be logically incompatible (see Abramson & Sackeim, 1977, for a review of this issue). One possibility is that depressives may perceive themselves as unable to control a personal tendency to bring about negative outcomes (e.g., an individual feels that by being unattractive or socially unskilled, he or she prevents relationships from arising; here responsibility is implied as well as lack of control). In their theoretical formulation, Abramson et al. (1978) asserted that negative events are not necessarily a risk factor for pervasive learned helplessness deficits unless individuals make global, stable, and internal attributions for these events. Alloy et al. (1990) have since extended this model to include the notion of helplessness and hopelessness as respective risk factors for anxiety and depression. The assessment of attributional style in children was fostered by the development of the Children’s Attributional Style Questionnaire (CASQ; Kaslow, Tanenbaum, & Seligman, 1978; Seligman et al., 1984). The scale assesses explanatory style for these events and is scored on six dimensions: internality, globality, and stability, for both positive and negative events. Some of the controversy involving the correlational study of attributional style concerns the causal direction of attributions and depression. For example, it may be that the depressive attributions are merely consequential to depression itself and may not represent a vulnerability or causal risk factor. Accordingly, the most informative examinations of the cognitive vulnerability model of depression have necessarily been prospective in their design. For example, Nolen-Hoeksema, Girgus, and Seligman (1986) examined attributional style and self-reported depression in a sample of one hundred sixty-eight 8- to 11-year-old school children in a 1-year longitudinal investigation. Composite scores from the CASQ were found to be correlated not only with concurrent depression as measured by the CDI, but also with increases in CDI scores over time. This relation was found to be bidirectional, that is, increases in depression were also predictive of change in attributional style. Thus, the interpretability with respect to cognition as a risk factor is somewhat difficult. Nevertheless, the study represented sound support for a longitudinal relation between childhood attributional style and depression. In a 5-year prospective investigation of five hundred and eight 3rd-grade school children, Nolen-Hoeksema, Girgus, and Seligman (1992) extended their previous work to include influences of negative life events in their model. It was hypothesized that depression would be predicted not only by previous levels of negative attributional style, but also by an interaction of attributional

The Development of Anxiety 239

style and negative life events (Abramson et al., 1978). Moreover, the investigators asserted that the relation of attributional style and depression might become more stable as children became older. Data collected at nine separate epochs over the 5-year period supported these hypotheses. That is, attributional style as well as its interaction with negative life events predicted change in depression scores at the subsequent epoch. In addition, these effects were only observed at Epochs 5, 6, 8, and 9, supporting the adjunct hypothesis that this putative cognitive risk factor may not be fully developed or operative in younger children. Once again, a bidirectional relation was noted between attributional style and depression. Taken together, these results represent support for the relation of attributional style and self-reported depression in a number of ways. They not only represent support for longitudinal relations, but also involve multiple replication of the cross-sectional correlational findings. Although the results are not highly conclusive about the directions of causality, they suggest the compelling idea that early experiences with depression or negative events may contribute to the development of a cognitive style that only appears to demonstrate relatively stable effects on emotion after middle childhood. Structural Relations

Although the work of Nolen-Hoeksema et al. (1992) represents an important contribution to a cognitive-developmental theory for depression, it has been suggested that additional investigation may be required to outline explicitly the structural relations among environmental, cognitive, and clinical variables. For example, Cole and Turner (1993) described a lack of specific attention in the literature to the statistical differentiation of moderational and mediational processes that are an implicit part of a cognitive-developmental theory. For example, a mediational relation would suggest that the effects of negative experience increase negative cognitions, which in turn contribute to increased negative affect (i.e., anxiety or depression). On the other hand, a moderational model would describe the interaction of negative experiences with cognition to effect subsequent negative emotion (see Figure 14.2). The diathesis-stress conceptualizations implicit in most cognitive and cognitive-affective theories of depression and anxiety (e.g., Alloy et al., 1990; Barlow et al., 1996; Beck & Emery, 1985) are best conceptualized as moderational, in that the effects of environmental events are moderated, or amplified, through their interpretation (i.e., cognition). To address this issue, Cole and Turner (1993) comparatively evaluated moderational and mediational attributional models of depression in a nonclinical sample of three hundred fifty-six 4th-, 6th-, and 8th-grade students. Interestingly, the results suggested a mediational cognitive model for children in this age range, despite consistent theoretical support for a moderational cognitive model in adults (e.g., Abramson et al, 1989). Specifically, positive and negative activities (i.e., the frequency of self-reported pleasant and unpleasant events), as assessed by the Children’s Activity Inventory (CAI; Shelton & Garber, 1987), as well as peer ratings of competence, were observed to directly influence attributional style, not simply to interact with it (see mediational model, Figure 14.2). Attributional style, in turn, was found to influence scores on the CDI.

240 Bruce F. Chorpita and David H. Barlow A. Mediational Model (early development)

Event

Emotion

Vulnerability

B. Moderational Model (later development)

Event

Emotion (Amplifier)

Vulnerability

Figure 14.2 Mediational and Moderational Models of the Influence of Psychological Factors on the Expression of Emotion.

In a second examination, Turner and Cole (1994) hypothesized that a moderational model would begin to emerge in children only at later developmental levels. The CAI, the CDI, and a measure of cognitive errors were administered to four hundred nine 4th-, 6th-, and 8th-grade students. The investigators found evidence for Age × Event × Cognition interactions for all events except those related to sports performance. That is, for the oldest children in the sample, the moderational effects of cognition began to have an observable effect, but this effect was not present in the younger children in this sample. The results supported other work that suggested moderational effects of cognition with life events appear only at later developmental levels (e.g., Fincham & Cain, 1986; Rholes, Blackwell, Jordan, & Walters, 1980). It will be useful to extend this line of work with samples spanning a greater developmental period (e.g., into adolescence) to identify more accurately the trend of these moderational effects across development (i.e., the slope of the Event × Cognition term across age levels). In summary, the child attributional style literature has begun to outline a cognitive model of the development of depression characterized by attributional style as a cognitive mediator between events and depression early in development. In light of the present model relating control and anxiety, the general contrast between evidence for a mediational model in early childhood

The Development of Anxiety 241

and a moderational model for late childhood and adulthood (Cole & Turner, 1993; Hammen, Adrian, & Hiroto, 1988; Nolen-Hoeksema et al., 1992) offers a useful conceptual framework. That is, the environment may help to foster a cognitive template, with early uncontrollable experience contributing to the formation of a cognitive vulnerability (i.e., mediational model). Later in development, this vulnerability may then begin to operate as an amplifier for environmental events (i.e., moderational model). Although this developmental structure may be consistent across models of anxiety and depression, it is interesting to note the discrepancy between the emergence of control-related cognitions in young children (Burhans & Dweck, 1995) and the emergence of attributional style middle childhood implied by the cognitive developmental findings (e.g., Nolen-Hoeksema et al., 1992). One possible explanation may be that examinations of attributional style in children have almost exclusively involved the CASQ, and although attributional style could possibly emerge during early development, its reliable self-report might not arise until much later (e.g., Robins & Hinkley, 1989). Familial Influences Before speculating further about mediational structures in the present model of control cognitions and anxiety, it is necessary to review more fundamental relations regarding anxiety, control, and the environment. At present, there is a reasonable body of evidence supporting the notions that (a) a particular set of family characteristics is associated with the development of control-related cognitions in children (e.g., Schneewind, 1995) and (b) a particular set of family characteristics is associated with the development of anxiety and its disorders (e.g., Turner, Beidel, & Costello, 1987). We first review selected evidence for each of these two propositions, before advancing to discussions of whether the respective family characteristics are the same for both processes and ultimately whether family, control cognitions, and anxiety may be linked in a mediational structure in early development (see Figure 14.2). Development of Control-Related Cognitions

The theoretical importance of establishing links between early environment and control-related beliefs has inspired a growing amount of literature examining familial antecedents for uncontrollability in children. The summary of findings describes family characteristics that would provide the child with opportunities for experiencing control over reinforcing events, as the present model would suggest. Family Structure. With regard to family structure, one would predict that children who have the opportunity for undivided attention from parents and who need not compete with siblings for the available reinforcers should be more likely to develop a sense of control over events. A number of studies have documented this pattern, demonstrating that first-born children display more internal locus of control than later-born children (Crandall, Katkovsky, & Crandall, 1965; Hoffman & Teyber, 1979; Krampen, 1982). In addition, family size has also been shown to be related to control cognitions, such that external

242 Bruce F. Chorpita and David H. Barlow

locus of control beliefs increase in later-born children as family size increases (Walter & Ziegler, 1980). Parenting. One would also predict that particular dimensions of parenting style might help foster an increased or diminished sense of control in a child. Specifically, the present model would suggest that two dimensions should be most important. First, parents who are more contingently responsive would provide the child with more occasions to experience the ability to solicit reinforcement, presumably one of the earliest opportunities to experience control. Second, parents who are less intrusive and protective and who provide the child with occasions to develop new skills and to explore and manipulate the environment would help cultivate an enhanced sense of control over events. Such ideas have been supported in a number of studies. Regarding the first dimension noted above, parents who are consistently and contingently responsive to their children have been shown to have children with a more internalized locus of control (Diethelm, 1991; Schneewind & Pfeiffer, 1978). For example, Davis (1969) documented an association between inconsistent parental behavior during a family decision-making task and children’s external locus of control. Similarly, Skinner (1986) used observational methodology to assess parental contingency and found a tendency for high parental contingency to be associated with the child’s internal locus of control. With respect to the second parenting dimension, it has been demonstrated that parents who provide more opportunity for autonomy and independence and who encourage the development of new skills are more likely to foster internal locus of control beliefs in their children (Chandler, Wolf, Cook, & Dugovics, 1980; Gordon, Nowicki, & Wichern, 1981). Carton and Nowicki (1994) reviewed a variety of studies documenting the association between high external locus of control in children and parental dimensions of protectiveness (e.g., Biocca, 1985) and intrusive governing (e.g., Washington, 1974). Parents of children showing internal control expectancies were more likely to reward, value, or encourage independence (e.g., Gordon et al., 1981). With respect to the present model, it appears that both parental dimensions have a common provision of opportunities for the child to experience control over reinforcing events in early development, through social contingency and mastery of the environment. Over time, such experiences can become part of the child’s stored (learned) information and contribute to a generalized sense of control (e.g., Bryant & Trockel, 1976; Carton & Nowicki, 1994). Nolen-Hoeksema, Wolfson, Mumme, and Guskin (1995) recently examined the issue in familial influences on control-related cognition from the perspective of helplessness and depression. Two groups of 5- to 7-year-old children were compared with respect to their ability to demonstrate mastery versus helplessness in a puzzle task, completed jointly with their mothers. Groups differed on the presence or absence of major depression in the mother. Interestingly, maternal diagnosis did not account for differences in children’s display of helplessness, but rather the degree to which mothers were less responsive and able to encourage active problem solving best accounted for the display of helplessness in the children during the task. This highlights the importance of a specific psychosocial link between helplessness in parents and their children,

The Development of Anxiety 243

suggesting that it is not maternal depressed affect per se that influences cognitive development of a sense of diminished control in the child, but rather the specific dimensions that provide and encourage opportunities for control over events. Attachment Theory: A Bridge Between Control and Anxiety?

Given the appearance that particular family characteristics are related to the development of a sense of control in children, the next consideration becomes the relevance of these parenting dimensions to the development of anxiety. However, before proceeding directly to a review of familial antecedents to childhood anxiety, an examination of attachment theory may serve as a useful bridge, in that attachment theory implicitly involves the development of a cognitive style characterized by security (cf. prediction and control; Thompson, 1998) and has been implicated in the development of anxiety and psychopathology more generally (Campbell, 1989). According to classic attachment theory (Bowlby, 1969, 1973, 1980), the parent serves an evolutionary and biological function of a protective and secure base (i.e., attachment object) from which the child operates. Critical to the child’s healthy functioning is a secure and predictable relationship with the caregiver. During a period of threat or surprise, the child can retreat to the parent safely. A child separated from the attachment object may therefore become anxious and protest in a programmed attempt to elicit reunion. If this relationship is disrupted, however, (e.g., through repeated, unexpected separation) one possible outcome may involve the child’s exhibiting “anxious attachment,” becoming more chronically dependent and apprehensive. Further disruption in the relationship will be followed by the gradual dissolution of the anxious response pattern and will predispose a more withdrawn and depressive nature over time (e.g., Rutter, 1980). Empirical investigation of attachment quality has been a challenging task (Sroufe, 1979). This stems in part from the fact that attachment behavior is influenced by numerous transient variables that make reliable assessment difficult. Rutter (1980), for example, has suggested that hunger, pain, tiredness, sickness, fear, rejection, maternal sensitivity, and other state variables may influence attachment behaviors. Attempting to control the effects of transient variables, Ainsworth, Blehar, Waters, and Wall (1978) developed a standardized situational task for assessment of attachment quality. The task, termed the Strange Situation procedure, involved the separation and reunion of the 1-year-old infant and its mother. Ainsworth’s coding system allowed for the classification of infant behavior into three basic categories: (A) avoidant, in which the infant shows little preference for the attachment figure over others, and when separated from the attachment figure avoids the caregiver upon reunion; (B) secure, in which the infant can comfortably explore in the presence of the attachment figure and seeks proximity following reunion; and (C) resistant-ambivalent, in which the infant is withdrawn, shows minimal exploratory behaviors in the presence of the attachment figure, and demonstrates distress or agitation upon reunion. A fourth category, (D) disorganized-disoriented, was later added to this system to characterize

244 Bruce F. Chorpita and David H. Barlow

children who do not fall into the first three classes, are highly atypical, and are thought to be at greatest developmental risk (Main, 1996; Main & Solomon, 1990). This assessment procedure has subsequently fostered a substantial body of research on attachment (e.g., Radke-Yarrow, Cummings, & Kuczynski, 1985; Urban, Carlson, Egeland, & Sroufe, 1992; Waters, Vaughn, Posada, & Kondo-Ikemura, 1995; Waters, Wippman, & Sroufe, 1979). Recent advances in attachment theory emphasize the quality of the infant-child relationship as an antecedent of general sociopersonality development (see Thompson, 1998, for a review). Indeed, some of the proximal sequelae of attachment security appear to be closely related to control cognitions as articulated in our present model. For example, Thompson stated, “Another kind of expectation emerging from. . . . early interactive experiences concerns the infant’s emerging sense of agency or effectance,” and further, “An awareness that [infants’] signals and actions can have predictable effects on others is fostered by the contingency inherent in the adult’s responsiveness” (p. 29). Thus, an important connection can be drawn between the reciprocal social influences inherent in secure attachment and the notions of consistency and autonomy valued in the literature on the development of control cognitions (e.g., Carton & Nowicki, 1994). Rutter (1980) has elaborated on this reciprocal communicative nature of the attachment relationship. He suggested that to the extent that the infant and mother have a well-established repertoire of ways in which to interact, communicate, and influence each other (e.g., smiling, cooing, grabbing), the greater the potential for secure exploratory behavior by the infant. According to Rutter, attachment has specific effects on the infant that are not characteristic of other social interactions. In particular, (a) anxiety will increase attachment or proximity seeking, (b) separation from the attachment figure will increase anxiety and withdrawal, and (c) presence of the attachment figure promotes exploration and lowers inhibition and anxiety. Sroufe (1990) described one example of this process in detail. Specifically, healthy infants who become overstimulated during interaction with the mother can signal the mother to deescalate the interaction using subtle cues (e.g., head turning). Mothers who respond to these cues appropriately allow the child to return to a state of less arousal, and hence prevent crying or other disorganized affect in the infant. At some point, this ability to negotiate the intensity of interaction is thought to become internalized in the infant. Children who are unable to develop this particular skill are believed to be at risk for subsequent anxiety or depression (Sroufe, 1990). These ideas are consistent with the model presented so far that early experience and skill with response-contingent reinforcing outcomes (i.e., control) is one pathway to positive long-term functioning. Outcomes of Attachment. Again, the question remains how and whether these early socialization processes may be related to anxiety and disorders of emotion more generally. Campbell (1989) reviewed a number of empirical investigations on the long-term outcome of attachment patterns. In a high-risk sample of children from impoverished and stressful environments, for example, Sroufe (1983) found anxious attachment (avoidant and resistant) at 1 year of age to be related to behavioral, school-related, and interpersonal problems in preschool. Similarly, in a sample of middle-class children, Lewis,

The Development of Anxiety 245

Feirig, McGuffog, and Jaskir (1984) detected an association between anxious attachment at 1 year and level of psychopathology at age 6 but only for boys. Unfortunately, because of the lack of clinical measures, it is unclear the degree to which these behaviors were related to anxiety. Other findings are less consistent. For example, Bates and Bayles (1988) followed a similar sample over 5 years and found no relation between security of attachment at infancy and subsequent behavior problems as reported by the mother. In general, the evidence is difficult to interpret clearly. As stated earlier, it is frequently difficult to rule out the confounding effects of extensive biological and environmental factors that can act as common influences for attachment and outcome measures (e.g., Lewis, 1990), and Lewis et al. (1984) strongly emphasized the interplay of attachment with other variables (i.e., stressors, demographics) in predicting later child functioning. Continued efforts to evaluate the long-term outcome of attachment in terms of psychopathology and specific syndromes are greatly needed (Thompson, 1998). Attachment Representations. More recent research in the area of attachment has focused on “attachment representations” (i.e., internal working models of the early attachment relationship carried forward into adulthood). The quality of adult attachment has been classified as autonomous-secure, preoccupied, and dismissing (a fourth category of unresolved-disorganized attachment is assigned to individuals not able to be classified into the first three). These groups have been found conceptually to correspond respectively with Ainsworth et al.’s (1978) secure, ambivalent-resistant, avoidant (and disorganized) infant patterns (e.g., Bremerton, 1992; van IJzendoorn & Bakermans-Kranenburg, 1996) and have been shown to be related to pathological functioning (e.g., Fonagy et al., 1996). In this regard, the study of attachment representations in adulthood has in many ways begun to fill some of the conceptual gaps between developmental and cognitive models. In general, this idea that early experiences with attachment figures might not only influence immediate functioning but also contribute to longer standing attachment representations is highly concordant with the recent findings in the literatures on child depression and the socialization of control-related beliefs reviewed above, particularly with respect to the formation of an enduring cognitive risk factor. The links are weakest, particularly in the extant control and attachment models, between these sociocognitive styles (internal control, attachment security) and later psychopathology. Development of Anxiety

Because of this lack of direct support for the link between a sense of diminished control and anxiety, it becomes necessary (at least initially) to look elsewhere to draw such connections. Both the control and the attachment literatures summarized so far appear to implicate the importance of two basic parenting dimensions, which can be abbreviated as follows: (a) sensitivity-consistencycontingency and (b) encouragement of autonomy–lack of intrusion and excessive control. More interesting, the literature relating parenting style and later anxiety implicates two dimensions of parenting demonstrating remarkable overlap with those reviewed above. These dimensions are often labeled

246 Bruce F. Chorpita and David H. Barlow

warmth or sensitivity and overprotection or control. (Here, control as a parenting dimension is the actual intrusive governance and associated constraint imposed on the child’s actions and should not be confused with the cognitive factor of control as discussed above.) In several studies using a retrospective self-report measure designed to assess parenting style from one’s childhood (Egna Minnen Beträffende Uppfostran [EMBU]; “my memories of upbringing”), Perris, Jacobsson, Lindstrom, von Knorring, and Perris (1980) outlined four basic factors of parenting style: Rejection, Emotional Warmth, Overprotection, and Favoring Subject. The Overprotection scale assesses the degree to which parents constrain and intrude on the child’s environment (Gerlsma, Emmelkamp, & Arrindell, 1990). Across a number of studies, the scores on the Overprotection scale have been found to discriminate between clinically anxious samples and controls (e.g., Ehiobuche, 1988) and remitted depressives and controls (e.g., Gotlib, Mount, Cordie, & Whiffen, 1988). (The use of populations in remission is particularly noteworthy, given the state dependent memory bias one might expect for depressed individuals.) These findings complement Rotter’s (1966) assertion that powerful parents can foster the development of external control orientations. The work of Parker (1983; Silove et al., 1991) remains perhaps the most thorough explication of the relation between parental behavior and the development of anxiety. Parker, Tupling, and Brown (1979) developed the Parental Bonding Instrument (PBI) to assess perceived dimensions of warmth and control in parents, which Parker terms care and protection, respectively. Like the EMBU, the PBI is a self-report measure filled out retrospectively by adults to describe the parenting they received as children. The questionnaire is filled out once to rate each parent, yielding four scales total: Maternal Care, Maternal Protection, Paternal Care, and Paternal Protection. The test-retest reliabilities for these four scales range from .87 to .92 in a sample of depressed patients (Parker, 1981) and from .63 to .76 in a nonclinical sample (Parker et al., 1979). Validity for the scales has also been shown to be satisfactory. Perhaps most importantly, the dimensions identified (i.e., care, protection) seem fairly robust in that they are consistent with a diversity of related measures that have yielded theoretically similar two-factor models (Gerlsma et al., 1990), not to mention their correspondence with the control and attachment literatures reviewed above. The dimension of protection (maternal or paternal) is intended to assess what Parker (1983) terms overprotection. He describes this construct as involving excessive parental involvement in controlling the child’s environment to minimize aversive experiences for the child (Parker, 1983). As the term overprotection implies, these anticipated aversive experiences may not represent actual threats to the child, and thus overprotection may not appreciably limit the exposure to aversive situations. However, overprotection is likely to narrow the range of behaviors likely to be exhibited by the child or to constrain the child’s ability to manipulate and engage the environment independently (cf. “coercion”; Hayes & Maley, 1977). The other dimension outlined by Parker (1983), care (maternal or paternal), is related to the notion of warmth or responsiveness of the parent. Again,

The Development of Anxiety 247

this dimension is compatible with both the control and attachment literatures in suggesting that consistently unresponsive parenting may lead to disruption and distress over the course of the child’s development. Low parental care may also serve to teach the child that his or her actions may not control or influence important stimuli in the environment (i.e., reinforcers). An example of this possibility presented by Sroufe (1990) was reviewed above. Parker (1983) has suggested that these two dimensions alone may not be sufficient to predict anxious or depressive outcomes over the course of development but rather that the combination of high protection and low care, what he calls “affectionless control,” is likely to have the most reliable negative influence. For example, a child in a high care and high protection environment may nevertheless have an indirect avenue to control events; for example, crying could elicit a (caring) parental response of investigation or elimination of aversives in the environment. It is when both pathways to control are limited that the child may be at increased risk. Specifically, the effect of low responsivity of parents who, in turn, constrain and narrow behavioral options for their child has visible implications for the child’s development of a sense that events are not under the child’s control. One notable difficulty with the study of care involves the routine association of high attention with contingent attention, creating a confound when one attempts to examine the role of contingency alone. An additional problem with this line of research is that it deals only with the main effects of each parenting dimension and unfortunately does not often examine the statistical interaction of these two types of parenting style. Nevertheless, albeit in a preliminary manner, a number of studies have summarized evidence about parenting style that are consistent with the premise that early experience with lack of control may foster increased risk for anxiety and mood disorders. The PBI has most frequently been used to examine the relation between clinical disturbance and perceptions of parenting. In a study of 125 “neurotic depressives,” 125 matched controls, and 125 “screened” controls (interviewed to ensure no history of depression), the depressive patients were found to have a significantly higher frequency of classifying their parents as low in care and high in protection (Parker, 1979a). A subsequent study (Parker, 1979b) examined the relation of PBI scores to various self-report measures in a nonclinical sample of 236 predominantly Australian participants. Rather than the traditional separation of parents into mother and father groups, parents or other caretakers were classified as to how important they were in the individual’s childhood in terms of parenting. PBI scores were obtained with respect to the most and the second-most important parenting figures for these participants. The protection scores for the primary parenting figure were significantly and moderately correlated with trait depression, the number of depressive episodes in the past year, low self-esteem, trait anxiety, and neuroticism scores from the Eysenck Personality Inventory (EPI; Eysenck & Eysenck, 1964). Care scores for the primary parenting figure were significantly correlated negatively with trait depression, the number of depressive episodes in the past year, low self-esteem, and alienation. Replication of the research involving depression and parenting style has demonstrated that adults with major depression score higher than nonclinical participants on the protection scales and lower on the care scales (Plantes, Prusoff, Brennan, & Parker, 1988).

248 Bruce F. Chorpita and David H. Barlow

Investigations involving bipolar disorder have found that this relation does not generalize to the other mood disorders (Joyce, 1984). Similar investigations have been conducted with anxious samples. Parker (1981) compared 50 outpatients having diagnoses of an anxiety disorder to a matched sample of controls selected from archival normative data (Parker et al., 1979). Anxious participants were found to have a significantly higher rate of classifying parents as high on protection and low on care. In a sample of 289 nonclinical adults, Parker (1979c) found that participants classifying their parents as higher on protection and lower on care demonstrated significantly higher trait anxiety scores than the others in the sample. These results are quite consistent with those from a previous sample and support the notion that perceived parenting style is related to trait anxiety. Attempting to identify how parenting style might discriminate among different anxiety disorders, Silove, Parker, Hadzi-Pavlovic, Manicavasagar, and Blaszczynski (1991) investigated recall of parenting style in patients with DSM-III-R (American Psychiatric Association, 1987) diagnoses of panic disorder (PD; n = 42), generalized anxiety disorder (GAD; n = 36), and matched controls (n = 205). The investigators used scores on the PBI to classify parents into four groups, corresponding to (a) high care–low overprotection, (b) high care–high overprotection, (c) low care–low overprotection, and (d) low care–high overprotection (i.e., “affectionless control”). Conditional logistic regression was used to predict the risk of being assigned to one of these groups based on the diagnostic group membership. Results suggested that insufficient care and excess protection were associated with clinical anxiety. Separate analyses demonstrated that high overprotection scores seemed to be associated with both PD and GAD, and low care scores were associated with PD only. Once again, the results are consistent with the basic notion that a relation exists between high parental control and low parental care and children’s risk for anxiety and depression, although definitive conclusions await stronger empirical support, particularly in terms of the hypothesized interaction of the two parenting dimensions (i.e., care and protection). Dumas, LaFreniere, and Serketich (1995) have identified a similar pattern of coercion and control in mother–child dyads involving anxious children. Preschool children were classified as competent (n = 42), aggressive (n = 42), and anxious (n = 42) using a teacher rating scale. Mother–child interactions during an elaborate, semistructured game were videotaped and coded for affect and behavior. Mothers of anxious children were found to show the highest levels of “aversive control” (i.e., attempts to elicit compliance involving criticism, punishment, intrusion, etc.) and the lowest levels of compliance and responsivity to the child. More interesting, anxious children were found to show the highest degree of noncompliance during these interactions. Dumas et al. (1995) highlighted the role of aversive control attempts in terms of their potential involvement with dysfunction in these dyads and suggested that such interactions may limit the development of prosocial behaviors and adaptive coping styles in anxious children. Recent research by Reiss et al. (1995) represents perhaps the most rigorous cross-sectional évaluation of psychosocial familial influence for negative affect to date. In a study of 708 families, the effects of various dimensions of parenting

The Development of Anxiety 249

were examined on symptoms of depression in adolescents. In a structural equations analysis, multiple indicators (i.e., adolescent and parent self-report, behavioral observation) of parenting variables were examined in terms of their relation to self- or parent-reported symptoms of depression in the adolescent. Parenting was measured through self-report as well as coded videotaped interactions of parents with their adolescents, and was classified into domains of conflict—negativity, warmth-support, and monitoring-control. It was found that paternal and maternal warmth-support (cf. care) had significant path coefficients (−0.26 and −0.37) to adolescent depressive symptoms, suggesting that low warmth was related to increased symptoms in the offspring. In a second analysis, monitoring-control was not found to be related to depressive symptoms; however, one of this domain’s lower order factors, maternal “attempts at control” over the adolescent, was found to have an observable influence on depressive symptoms in that adolescent. Although these data are supportive, some caveats are warranted with respect to the present model. It is unclear, for example, the degree to which the construct of “attempts at control” relates to the notion of Parker’s (1983) “protection” or to the putative cognitive variables associated with a sense of control. In addition, although genetic data were gathered in this study, the results presented are preliminary in that there was no correction for heritability effects in the present analyses. The fact that all of the probands were adolescents also obscures the developmental nature of parental influence, in that parental control may no longer be directly contributory to symptoms of anxiety or depression by these later years. Nevertheless, despite these limitations, Reiss et al. (1995) is clearly a major step toward a comprehensive understanding of the influence of parenting on pathological symptomatology in offspring. The analytical strategies in particular are a likely prototype for the type of work that lies ahead in this area. Of course, longitudinal studies using similar structural and multivariate methodology will likely be the definitive test of these hypotheses. With the exception of Reiss et al. (1995), most studies of parenting style are limited by the fact that assessment has been conducted retrospectively, requiring individuals with anxiety or depression to describe their perceptions of early parenting experiences. Although some evidence exists that these retrospective accounts may not be state dependent or otherwise biased (e.g., Parker, 1979d), there is clearly a need to validate this line of work with an emphasis on cross-sectional or prospective (Messer & Beidel, 1994). Most important, although studies of the relation of parenting style to anxiety or depression implicate high protection and low warmth as significant influences, few studies integrate these findings with the developments in cognitive theory. That is, almost all of the studies, including the work of Reiss et al. (1995), fail to examine the role of a cognitive link (e.g., attributional style, locus of control) between parental behavior and the development of negative affect, a relation that has received preliminary examination in the context of attribution and childhood depression. Because findings from biological, cognitive, and emotion theories support the idea that early experiences with control may play a role in the origin of cognitive vulnerability, it would seem prudent to continue to examine parenting with respect to its influence on or

250 Bruce F. Chorpita and David H. Barlow

through cognitive phenomena (mediational model, Figure 14.2). Research in this area is only in its earliest stages (e.g., Chorpita, in press). Family Influences on Neuroendocrine Responding

Another area of research attention has involved family influences on biological variables, most notably neuroendocrine responding. As mentioned above, a common neuroendocrine correlate of anxiety (and depression) is elevated basal cortisol level. These observations are consistent with Gray’s model of anxiety, in that BIS activity is described as priming the hypothalamic pituitary adrenocortical (HPA) axis, which has cortisol as its end product. Not surprisingly, Kagan, Reznick, and Snidman (1987) found that children classified as behaviorally inhibited were found to have elevated levels of cortisol. Although Kagan et al.’s definition of inhibition differs somewhat from Gray’s, Kagan et al.’s inhibition has been found to be associated with the increased likelihood for the development of anxiety disorders (Hirshfeld et al., 1992) and may represent a very similar underlying process. Regarding the family, it has been demonstrated that features of the parent–child relationship can potentiate or inhibit this HPA stress response. For example, Nachmias, Gunnar, Mangelsdorf, Parritz, and Buss (1996) examined cortisol responding in 77 infants exposed to novel stimuli (e.g., clowns, puppets). Consistent with prediction, only children who had been classified as inhibited showed elevations in salivary cortisol. However, among the inhibited children, salivary cortisol change scores were noted only for those with insecure attachment (avoidant or resistant). The responding of the securely attached inhibited children looked like that of the uninhibited children. These findings suggest that inferences about the developmental elaboration of inhibited temperament may need to consider the influence of parenting. As mentioned above, the theoretical connection between attachment quality and experience with control highlights control as a potentially important pathway for these influences. Gunnar, Larson, Hertsgaard, Harris, and Broderson (1992) looked more closely at the involvement of control by examining caretakers’ ability to influence the HPA cortisol responding. In discussing this study, Gunnar (1994) described separation from parents as involving loss of control: “Not only is the child blocked from access to the mother (loss of control over proximity), but the child also loses the mother’s help in controlling the internal and external environment” (p. 182). Gunnar et al. found that 9-month-old infants showed an elevated cortisol response when separated from mother, but that this effect was eliminated when the infant was accompanied by a highly responsive caretaker versus a less responsive caretaker. These results underscore the possibility that contingent access to appetitive stimulation (i.e., control) may play an important role in reducing HPA responding. Taken together with the findings of Nachmias et al. (1996), they highlight the importance of the interplay between parent–child interaction and temperamental factors related to anxiety. Questions about the degree to which experience with control accounts for the observed effects of responsive parenting clearly suggests an area for continued study.

The Development of Anxiety 251

Stressors and Resilience An Overview

Although parenting plays a major role in shaping the environment of the child, a host of other factors are also important and may also influence the development of a sense of control. For example, the relation between stressors and subsequent maladjustment in children is supported by considerable research (Garmezy, 1986; Garmezy & Rutter, 1983; note also that this relation may be moderated by parenting behavior). However, the effects of stressors are not straightforward, in that their long-term effects on development are not necessarily harmful. In fact, the negative impact of stressors appears to be closely tied to their interpretation (as suggested earlier) and to one’s ability to control these stressors. When control is possible, environmental challenges and the associated distress may potentially facilitate adaptive functioning over time. For example, Izard (e.g., Izard & Beuchler, 1979) described the states of arousal normally associated with stressful childhood experience as the emergence of basic human emotions. Thus, crying, yelling, or other apparently disorganized expressions of emotion in children may not necessarily represent compromised performance, as some models have suggested (see Garmezy, 1986), but rather an integral part of a negotiation or coping process in which an infant or young child must sometimes engage (cf. attachment; Sroufe, 1990). Hence, the degree to which reinforcing events in the environment are contingent on these emotional responses, such stress reactions may actually be desirable and adaptive. This notion resonates with the work reviewed thus far, in that it implies that the ability to manipulate one’s environment effectively, most notably under stress (i.e., coping), leads to healthy development. In a related vein, Masten et al. (1991) reviewed conditions found to serve as protective factors in child development, citing responsive parenting, intellectual skill, social-cognitive ability, and Bandura’s (1986) conception of self-efficacy as factors predictive of healthy development. Although all of these qualities are related to the degree of mastery and control one has over the environment, the construct of self-efficacy may be the most similar to the present notion of control. Self-efficacy, however, is related more to the idea of being able to perform, whereas the idea of control is more broadly related to all aspects of functioning and does not imply a behavioral response. In terms of learning theory, self-efficacy has been challenged as merely being one factor of a more general “outcome expectancy,” and thus it has been argued that the self-efficacy concept has no unique explanatory power (see Biglan, 1987; Eastman & Marzillier, 1984; Lee, 1989). In the present model, control is related to the degree to which an organism has the possibility to influence opportunities for positive or negative reinforcement and is not specific to the organism’s motor behavior. Nevertheless, as a concept related to the behavioral performance aspects of control, self-efficacy has been supported as a predictor of positive development (Masten et al., 1991). In other examinations of protective factors, much has been made of the so-called “steeling effect” (e.g., Garmezy, 1986), that is, the tendency for certain stressors to actually “immunize” or enhance subsequent functioning of the individual. The phenomenon of “required helpfulness” (Rachman, 1979)

252 Bruce F. Chorpita and David H. Barlow

is one such example. According to this principle, if high-risk children are called on to assist in the coping efforts of a larger system (e.g., family, community), this action may have beneficial effects. Elder (1974) described putative psychological benefits experienced by children of the Great Depression, some of whose efforts to contribute to the family welfare may have generated powerful mastery experiences for them. Findings from animal literature also support the notion of a “steeling effect.” Dienstbier (1989) reviewed evidence with rats and mice showing that early exposure to stressful manipulations such as shock or rough handling can actually lead to less fearful or reactive animals when exposed to subsequent threats (e.g., Denenberg, 1967; Levine, 1960). Hannum, Rosellini, and Seligman (1976) showed that although early experience with uncontrollable trauma produced later helplessness effects in rats, controllable trauma actually immunized the animals against later helplessness. Such findings demonstrate the complexity of the effects of stressors on development. Physiological Arousal: Two Responses to Stressors

The understanding of these effects may best be derived from a closer examination of the physiology associated with exposure to stress. Dienstbier (1989) argued that there is evidence for two separate types of short-term physiological response to stress and described these as (a) a catecholamine response, involving the sympathetic nervous system (SNS) and adrenal medulla; and (b) a cortisol response, involving the HPA axis. Reviewing a great number of studies from both human and nonhuman populations, Dienstbier concluded that the long-term ill effects of stress are likely mediated by either catecholamine depletion or by chronic cortisol secretion. He reported that heightened catecholamine capacity is associated with increased performance (e.g., O’Hanlon & Beatty, 1976) and lower neuroticism scores (e.g., Forsman, 1981) and that high levels of basal cortisol are related to depression and anxiety (e.g., Anisman & LaPierre, 1982). What appears to be responsible for the differential long-term effects of stress (“steeling” versus “helplessness” induction) relates back to the idea of prediction and control. Dienstbier (1989) stated that, “in a range of subjects from rats to mice to primates, exposure to stressors with minimal predictability, control, or feedback . . . results in heightened cortisol responses,” and that “when stressful situations are sufficiently extended, they also lead to catecholamine depletion” (pp. 86–87). In describing what he terms toughening, Dienstbier (1989) described how intermittent stressors activate the catecholamine response, and in predictable situations or those allowing coping action, the cortisol response will be minimal (cf. “toughening up”; Miller, 1980). An organism experiencing these types of stressors will actually undergo an increase in catecholamine responding potential, ultimately leading to improvements in performance and functioning under later stress. So long as there are sufficient intervals between stressful events to allow catecholamine restoration and there is enough control or predictability to inhibit excessive HPA activity, the organism will benefit from this exposure. It appears, then, that Dienstbier has identified possible biological substrates of “steeling” and the development of mastery.

The Development of Anxiety 253

From this analysis, it also appears the HPA axis may be more associated with anxiety and that the SNS-adrenal medullar axis may be more associated with fear. This conceptualization helps to unify the diversity of findings relating uncontrollability to inhibition or anxiety and to hypercortisolism (see Gray, Mineka, Kagan, and Sapolsky articles cited above). The fear response, on the other hand, appears to involve another system (i.e., Gray’s [FFS; 1982]), the activation of which may have few deleterious effects, provided that the fear response is not without sufficient reprieve. Again, it is only when fear becomes unpredictable or uncontrollable (e.g., as with PD; Barlow, 1988), that the elicited apprehension leads to long-term impairment in functioning via the HPA system. Understanding fear and anxiety in terms of two separate subsystems of arousal fosters an enhanced understanding of the complexity of these emotions. Conceptual Model and Implications Considered together, the evidence reviewed suggests a number of important points. The idea that experience with lack of control may play an important role in the development of anxiety appears to be suggested by diverse areas of research. In addition, the biological correlates of chronic anxiety appear to be influenced to some degree by psychological variables, using the psychological interpretation of stressful events as either controllable or uncontrollable. In terms of how this sense of control develops, it may be that early experience with lack of control contributes to something of a psychological template, which at some point becomes relatively fixed and diathetic. Stated another way, this psychological dimension of a sense of control is possibly a mediator between stressful experience and anxiety, and over time this sense becomes a somewhat stable moderator of the expression of anxiety (see Figure 14.2). As mentioned earlier, all of these implications require careful and continued testing, as they are just beginning to find empirical support (e.g., Chorpita, Brown, & Barlow, in press). In an effort to integrate these ideas regarding the structure of vulnerability for negative emotion and its associated developmental influences, Chorpita (in press) outlined a preliminary conceptual model (see Figure 14.3). Within this model, early experience with uncontrollable or unpredictable stimuli is described as influencing low perceptions of control, in a manner consistent with the literature on socialization and development of control-related cognitions (e.g., Carton & Nowicki, 1994). In accord with Gray’s (1987) model, the immediate sequelae of these inputs are increased inhibition (i.e., BIS activity; also, Gray’s “anxiety”), followed by the associated somatic outputs first documented by Kagan and colleagues (e.g., Kagan et al., 1987). As one moves across time to a later developmental period, these control-related cognitions may intensify the activity of systems (e.g., Gray’s BIS) underlying inhibition (Kagan et al., 1987) through the events outlined by Dienstbier (1989). Here, Chorpita (in press) offered the term weakening to describe the opposed end of Dienstbier’s phenomenon, such that the long-term influence of perceptions of low control would bring about the ill effects of hypercortisolism for the BIS, increasing the likelihood of phenomena such as glucocorticoid feedback insensitivity and hence further BIS activation.

254 Bruce F. Chorpita and David H. Barlow Early Development

Later Development

Uncontrollable Stimuli

Stimuli

Stored Information Low Perceived Control (i.e., Memory)

Low Perceived Control

(mis)Interpreted as Uncontrollable

Intensification of BIS Activity Inhibition Temperamental Stability of Hypothalamic and Limbic Structures (BIS)

Undifferentiated Somatic Outputs

Negative Affect

Increased CRF and HPA Activity

Figure 14.3 Model of the Development of Vulnerability for Anxiety and Depression. BIS = behavioral inhibition system; CRF = corticotropin releasing factor; HPA = hypothalamic pituitary adrenocortical.

In addition, the model allows for some temperamental stability, as would be expected given the strongly biological nature of these elements. Another pathway by which vulnerability is carried forward concerns the cognitive information related to perceptions of control, which are represented as information stored in memory (e.g., “stored regularities”; Gray & McNaughton, 1996). As Rotter (1966) suggested, this information would become increasingly resistant to new information over time, in that the accumulation of prior experience would have the increasing ability to dilute the effects of new information on generalized expectancies. One could then imagine that, given a history of experience with low control, interpretation or processing of events later in development (especially those for which cues for control are ambiguous) may become biased by prior information. The precise mechanisms and timing by which such biases might advance across development certainly await additional investigation. Indeed, although the role of cognitive biases is ubiquitous in adult models of anxiety and depression (McNally, 1996), little is known about when these control expectancies might become resistant to new information, for which children they occur, and as a result of which experiences they occur. Regarding the output of these elements in later development, one could

The Development of Anxiety 255

speculate (as our reading of Gray’s model would predict) that such processing of stimuli should increase BIS activity, characterized in late development by the experience of negative affect. Further, at this later developmental period, previously undifferentiated somatic outputs of the BIS may become more focused on the HPA system (Dienstbier, 1989), yielding in certain individuals a profile more consistent with that of negative affect as observed in adults. Implications

At present, the role of cognitive and psychological variables in the pathogenesis of anxiety requires clarification. Cole and Turner (1993) remarked that despite well-established theories of anxiety and depression, “surprisingly little empirical or theoretical work has been conducted on the emergence of these cognitive diatheses . . . most theorists point vaguely back to childhood events” (Cole & Turner, 1993, p. 271). They stated further that with regard to anxiety, structural (e.g., mediational vs. moderational) etiological models have simply not been tested. Because preliminary support for a mediational cognitive model exists for childhood depression (Cole & Turner, 1993; Turner & Cole, 1994), it has become an important empirical question to test such a model in the context of a general negative emotional factor (e.g., inhibition; see the four elements in the leftmost column of Figure 14.3). Although some preliminary modeling has been conducted, much of the recent etiological research in this area has also failed to include a clinical validation of the self-reported symptoms (see Chorpita, in press, for an expanded clinical model). In addition, although experimental manipulations of immediate control have been conducted with normal children (e.g., Gunnar, 1980), similar work with clinical samples (e.g., Sanderson et al., 1989) is needed to test directly the immediate effects of low perceived control on the expression of fear and anxiety in children with anxiety disorders. This type of research is only just beginning. Traditionally, much has been said about the difference between childhood and adult anxiety. However, there is need for clarification. Although the structural association among key elements appears to change with development (see Figure 14.2), the similarity of the elements themselves associated with childhood and adult anxiety appears to be strong and noteworthy (see Figure 14.3). Thus, factors influencing anxiety in adults may require increased attention in childhood research and, likewise, a greater understanding of the etiological and developmental aspects of perceived uncontrollability and anxiety may contribute to the understanding of adult disorders of emotion. Although clear and considerable differences exist in the classification, course, and expression between anxiety in childhood and adulthood, some differences may be more phenotypic than absolute. Future theorizing will likely benefit from continuing unification of existing models. Notes 1 The dexamethasone suppression test involves the indirect assessment of pituitary feedback sensitivity to Cortisol. Under normal circumstances, the pituitary gland should respond to the presence of dexamethasone (an agent indistinguishable from cortisol by the pituitary gland) by decreasing (i.e., suppressing) the release of adrenocorticotropic releasing

256 Bruce F. Chorpita and David H. Barlow hormone. It has been shown, however, that some adults with anxiety or depression do not show these normal rates of suppression and, consequently, dexamethasone suppression has been posited as a biological marker (albeit not necessarily causal) for these syndromes (e.g., Carroll, 1985; Schweizer, Swenson, Winokur, Rickels, & Maislin, 1986). 2 Ray and Sapolsky (1992) defined the tendency to differentiate between threatening and nonthreatening interactions as the absolute value of the difference between the probabilities of resumption of prior behavior following threat (as coded by investigators) and following nonthreat. Thus, for example, a baboon who did not return to prior behavior following a nonthreat (as coded by investigators) was considered to have misinterpreted nonthreat; whereas a baboon who resumed prior behavior in those same circumstances correctly interpreted nonthreat. This index was intended to measure accurate perceptions of threat. 3 Although the reviewed studies presented no findings related to the discriminant validity of the NSLOC, Nowicki and Strickland (1973) reported that NSLOC scores were not correlated with a measure of social desirability.

We thank Bob Rosellini for his helpful comments on an early draft of this article. References Abramson, L. Y., Metalsky, G. I., & Alloy, L. B. (1989). Hopelessness and depression: A theory based subtype of depression. Psychological Review, 96, 358–392. Abramson, L. Y., & Sackeim, H. A. (1977). A paradox in depression: Uncontrollability and self-blame. Psychological Bulletin, 84, 838–851. Abramson, L. Y., Seligman, M. E. P., & Teasdale, J. (1978). Learned helplessness in humans: Critique and reformulation. Journal of Abnormal Psychology, 87, 49–74. Ainsworth, M. D. S., Blehar, M., Waters, E., & Wall, S. (1978). Patterns of attachment. Hillsdale, NJ: Erlbaum. Alloy, L. B., Kelly, K. A., Mineka, S., & Clements, C. M. (1990). Comorbidity of anxiety and depressive disorders: A helplessness-hopelessness perspective. In J. D. Maser & C. R. Cloninger (Eds.), Comorbidity of mood and anxiety disorders (pp. 499–543). Washington D.C.: American Psychiatric Press. American Psychiatric Association. (1987). Diagnostic and statistical manual of mental disorders (3rd ed., revised). Washington D.C.: Author. American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington D.C.: Author. Angst, J., Vollrath, M., Merikangas, K. R., & Ernst, C. (1990). Comorbidity of anxiety and depression in the Zurich Cohort Study of Young Adults. In J. D. Maser & C. R. Cloninger (Eds.), Comorbidity of mood and anxiety disorders (pp. 123–137). Washington D.C.: American Psychiatric Press. Anisman, H., & LaPierre, Y. (1982). Neurochemical aspects of stress and depression: Formulations and caveats. In R. W. Neufeld (Ed.), Psychological stress and psychopathology (pp. 179–217). New York: McGraw-Hill. Bandura, A. (1986). Social foundations of thought and action. Englewood Cliffs, NJ: Prentice-Hall. Barlow, D. H. (1988). Anxiety and its disorders: The nature and treatment of anxiety and panic. New York: Guilford. Barlow, D. H. (1991). Disorders of emotion. Psychological inquiry, 2, 58–71. Barlow, D. H., Chorpita, B. E, & Turovsky, J. (1996). Fear, panic, anxiety, and the disorders of emotion. In D. A. Hope (Ed.), Nebraska Symposium on Motivation: Perspectives on anxiety, panic, and fear (Vol. 43, pp. 251–328). Lincoln: University of Nebraska Press.

The Development of Anxiety 257 Bates, J. E., & Bayles, K. (1988). The role of attachment in the development of behavior problems. In J. Belsky & T. Nezworski (Eds.), Clinical implications of early attachment (pp. 253–259). Hillsdale, NJ: Erlbaum. Beck, A. T., & Emery, G. (1985). Anxiety disorders and phobias: A cognitive perspective. New York: Basic. Biglan, A. (1987). A behavior-analytic critique of Bandura’s self-efficacy theory. The Behavior Analyst, JO, 1–15. Biocca, L. J. (1985). The relationship among locus of control, perceived parenting, and sex-role attributes in adolescence. Dissertation Abstracts International, 47, 2193B. Borkovec, T. D. (1994). The nature, functions, and origins of worry. In G. Davey & F. Tallis (Eds.), Worrying: Perspectives on theory, assessment, and treatment (pp. 5–35). Sussex, England: Wiley & Sons. Bowlby, J. (1969). Attachment and loss (Vol. 1): Attachment. New York: Basic. Bowlby, J. (1973). Attachment and loss (Vol. 2): Separation. New York: Basic. Bowlby, J. (1980). Attachment and loss (Vol. 3): Loss, sadness, and depression. New York: Basic. Bretherton, I. (1992). The origins of attachment theory: John Bowlby and Mary Ainsworth. Developmental Psychology, 28, 759–775. Brooks, V. B. (1986). How does the limbic system assist motor learning? A limbic comparator hypothesis. Brain, Behavior, and Evolution, 29, 29–53. Brown, T. A., Chorpita, B. F., & Barlow, D. H. (1998). Structural relationships among dimensions of the DSM-IV anxiety and mood disorders and dimensions of negative affect, positive affect, and autonomic arousal. Journal of Abnormal Psychology 107, 179–192. Brown, T. A., Chorpita, B. E., Korotitsch, W., & Barlow, D. H. (1997). Psychometric properties of the Depression Anxiety Stress Scales (DASS) in clinical samples. Behaviour Research and Therapy, 35, 79–89. Bryant, B. K., & Trockel, J. F. (1976). Personal history of control of psychological stress related to locus of control orientation among college women. Journal of Consulting and Clinical Psychology, 44, 266–271. Burhans, K. K., & Dweck, C. S. (1995). Helplessness in early childhood: The role of contingent worth. Child Development, 66, 1719–1738. Campbell, S. B. (1989). Developmental perspectives. In T. H. Ollendick & M. Hersen (Eds.), Handbook of child psychopathology (2nd ed.). New York: Plenum. Carroll, B. J. (1985). Dexamethasone suppression test: A review of contemporary confusion. Journal of Clinical Psychiatry, 46, 13–24. Carton, J. S., & Nowicki, S. (1994). Antecedents of individual differences in locus of control of reinforcement: A critical review. Genetic, Social, and General Psychology Monographs, 120, 31–81. Chandler, T. A., Wolf, F. M., Cook, B., & Dugovics, D. A. (1980). Parental correlates of locus of control in fifth graders: An attempt at experimentation in the home. Merrill-Palmer Quarterly, 26, 183–195. Chorpita, B. F. (in press). Vulnerability for anxiety and depression: Structure and development. In K. D. Craig, R. J. McMahon, & K. S. Dobson (Eds.), Proceedings of the 29th Banff International Conference on Behavioral Science. New York: Sage. Chorpita, B. F., Albano, A. M., & Barlow, D. H. (1998). The structure of negative emotions in a clinical sample of children and adolescents. Journal of Abnormal Psychology, 107, 74–85. Chorpita, B. E., Brown, T. A., & Barlow, D. H. (in press). Perceived control as a mediator of family environment in etiological models of childhood anxiety. Behavior Therapy.

258 Bruce F. Chorpita and David H. Barlow Clark, L. A., & Watson, D. (1991). Tripartite model of anxiety and depression: Psychometric evidence and taxonomie implications. Journal of Abnormal Psychology, 100, 316–336. Cole, D. A., & Turner, J. E. (1993). Models of cognitive mediation and moderation in child depression. Journal of Abnormal Psychology, 102, 271–281. Crandall, V. C., Katkovsky, W., & Crandall, V.J. (1965). Children’s belief in their own control of reinforcements in intellectual-academic achievement situations. Child Development, 36, 91–109. Davis, W. L., & Phares, E. J. (1969). Parental antecedents of internal-external control of reinforcement. Psychological Reports, 24, 427–436. Denenberg, V. H. (1967). Stimulation in infancy, emotional reactivity, and exploratory behavior. In D. C. Glass (Ed.), Neurophysiology and emotion (pp. 161–190). New York: Rockefeller University Press. Dienstbier, R. A. (1989). Arousal and physiological toughness: Implications for mental and physical health. Psychological Review, 96, 84–100. Diethelm, K. (1991). Mutter-Kind-Interaktion. Entwicklung von ersten Kontrollüberzeugungen [Mother-child interaction. Development of early control beliefs]. Bern, Switzerland: Huber. Di Nardo, P. A., & Barlow, D. H. (1990). Syndrome and symptom cooccurrence in the anxiety disorders. In J. D. Maser & C. R. Cloninger (Eds.), Comorbidity of mood and anxiety disorders (pp. 205–230). Washington D.C.: American Psychiatric Press. Dobson, K. S. (1985). The relationship between anxiety and depression. Clinical Psychology Review, 5, 307–324. Drugan, R. C., Ryan, S. M., Minor, T. R., & Maier, S. F. (1984). Librium prevents the analgesia and shuttlebox escape deficit typically observed following inescapable shock. Pharmacology, Biochemistry, and Behavior, 21, 749–754. Dumas, J. E., LaFreniere, P. J., & Serketich, W. J. (1995). “Balance of power”: A transactional analysis of control in mother–child dyads involving socially competent, aggressive, and anxious children. Journal of Abnormal Psychology, 104, 104–113. Dweck, C. S. (1991). Self-theories and goals: Their role in motivation, personality, and development. In R. Dienstbier (Ed.), Nebraska Symposium on Motivation and Emotion (Vol. 36, pp. 199–235). Lincoln: University of Nebraska Press. Dweck, C. S., & Leggett, E. L. (1988). A social-cognitive approach to motivation and personality. Psychological Review, 95, 256–273. Eastman, C., & Marzillier, J. S. (1984). Theoretical and methodological difficulties in Bandura’s self-efficacy theory. Cognitive Therapy and Research, 8, 213–229. Ehiobuche, I. (1988). Obsessive-compulsive neurosis in relation to parental child rearing patterns amongst Greek, Italian, and Anglo-Australian subjects. Acta Psychiatrics Scandinavica, 78, 115–120. Elder, G. H. (1974). The children of the Great Depression: Social change in life experience. Chicago: University of Chicago Press. Eysenck, H. J., & Eysenck, S. B. G. (1964). Manual of the Eysenck Personality Inventory. London: University of London Press. Finch, A. J., & Nelson, W. M. (1974). Locus of control and anxiety in emotionally disturbed children. Psychological Reports, 35, 469–470. Fincham, F. D., & Cain, K. M. (1986). Learned helplessness in humans: A developmental analysis. Developmental Review, 6, 301–333. Fonagy, P., Leigh, T., Steele, M., Steele, H., Kennedy, R., Mattoon, G., Target, M., & Gerber, A. (1996). The relation of attachment status, psychiatric classification, and response to psychotherapy. Journal of Consulting and Clinical Psychology, 64, 22–31. Forsman, L. (1981). Habitual catecholamine excretion and its relation to habitual distress. Biological Psychiatry, 11, 83–97.

The Development of Anxiety 259 Fowles, D. C. (1995). A motivational theory of psychopathology. In W. Spaulding (Ed.), Nebraska Symposium on Motivation: Integrated views of motivation, cognition, and emotion (Vol. 41, pp. 181–238). Lincoln: University of Nebraska Press. Garmezy, N. (1986). Developmental aspects of children’s responses to the stress of separation and loss. In M. Rutter, C. E. Izard, & P. B. Reid (Eds.), Depression in young people: Developmental and clinical perspectives. New York: Guilford. Garmezy, N., & Rutter, M. (1983). Stress, coping, and development in children. New York: McGraw-Hill. Gerlsma, C., Emmelkamp, P. M. G., & Arrindell, W. A. (1990). Anxiety, depression, and perception of early parenting: A meta-analysis. Clinical Psychology Review, 10, 251–277. Gordon, D. A., Nowicki, S., & Wichern, F. (1981). Observed maternal and child behaviors in a dependency-producing task as a function of children’s locus of control orientation. Merrill-Palmer Quarterly, 27, 43–51. Gotlib, I. H., Mount, J. H., Cordie. N. I., & Whiffen, V. E. (1988). Depression and perception of early parenting: A longitudinal investigation. British Journal of Psychiatry, 152, 24–27. Gray, J. A. (1982). The neuropsychology of anxiety. New York: Oxford University Press. Gray, J. A. (1987). The psychology of fear and stress (2nd ed.). Cambridge, England: Cambridge University Press. Gray, J. A., & McNaughton, N. (1996). The neuropsychology of anxiety: A reprise. In D. A. Hope (Ed.), Nebraska Symposium on Motivation: Perspectives on anxiety, panic, and fear (Vol. 43, pp. 61–134). Lincoln: University of Nebraska Press. Gunnar, M. R. (1980). Control, warning signals, and distress in infancy. Developmental Psychology, 16, 281–289. Gunnar, M. R. (1994). Psychoendocrine studies of temperament and stress in early childhood: Expanding current models, In J. E. Bates & T. D. Wachs (Eds.). Temperament: Individual differences at the interface of biology and behavior (pp. 175–198). Washington D.C.: American Psychological Association. Gunnar, M., Larson, M., Hertsgaard, L., Harris, M. L., & Broderson, L. (1992). The stressfulness of separation among 9-month-old infants: Effects of social context variables and infant temperament. Child Development, 63, 290–303. Hammen, C., Adrian, C., & Hiroto, D. (1988). A longitudinal test of the attributional vulnerability model in children at risk for depression. British Journal of Clinical Psychology, 27, 37–46. Hannum, R. D., Rosellini, R. A., & Seligman, M. E. P. (1976). Learned helplessness in the rat: Retention and immunization. Developmental Psychology, 12, 449–454. Hayes, S. C., & Maley, R. F. (1977). Coercion: Legal and behavioral issues. Behaviorism, 5, 87–95. Hetherington, E. M. (1989). Coping with family transitions: Winners, losers, and survivors. Child Development, 60, 1–14. Hirshfeld, D. R., Rosenbaum, J. F., Biederman, J., Bolduc, E. A., Faraone, S. V., Smidman, N., Reznick, J. S., & Kagan, J. (1992). Stable behavioral inhibition and its association with anxiety disorder. Journal of the American Academy of Child and Adolescent Psychiatry, 31, 103–111. Hoehn-Saric, R., & McLeod, D. R. (1985). Locus of control in chronic anxiety disorders. Acta Psychiatrica Scandinavica, 72, 529–535. Hoffman, J. A., & Teyber, E. C. (1979). Some relationships between sibling age, space, and personality. Merrill-Palmer Quarterly, 25, 77–80. Izard, C. E., & Beuchler, S. (1979). Emotion expressions and personality integration in infancy. In C. E. Izard (Ed.), Emotions in personality and psychopathology (pp. 447–472). New York: Plenum Press.

260 Bruce F. Chorpita and David H. Barlow Joiner, T. E., Jr., Catanzaro, S. J., & Laurent, J. (1996). Tripartite structure of positive and negative affect, depression, and anxiety, in child psychiatric inpatients. Journal of Abnormal Psychology, 105, 401–409. Joyce, P. R. (1984). Parental bonding in bipolar affective disorder. Journal of Affective Disorders, 7, 319–324, Kagan, J. (1989). Temperamental contributions to social behavior. American Psychologist. 44, 668–674. Kagan, J., Reznick, J. S., & Snidman, N. (1987). The physiology and psychology of behavioral inhibition. Child Development, 58, 1459–1473. Kaslow, N. J., Tanenbaum, R. L., & Seligman, M. E. P. (1978), The KASTAN: A children’s attributional style questionnaire. Unpublished manuscript, University of Pennsylvania. Klockars, A. J., & Varnum, S. W. (1975). A test of the dimensionality assumptions of Rotter’s Internal-External scale. Journal of Personality Assessment, 39, 397–404. Kovacs, M. (1981). Rating scales to assess depression in preschool children. Acta Paedopsychiatry, 46, 305–315. Krampen, G. (1982). Schulische und familiäre Entwicklungsbedingungen von Kontrollüberzeugungen [School and familial conditions for developing control beliefs]. Schweizerische Zeitschrift für Psychologie und ihre Andwendungen, 41, 16–35. Lazarus, R. S. (1966). Behaviour reversal vs. non-directive therapy vs. advice in effecting behaviour change. Behaviour Research and Therapy, 4, 209–212. Lazarus, R. S. (1968). Emotions and adaptation: Conceptual and empirical relations. In W. J. Arnold (Ed.), Nebraska Symposium on Motivation (Vol. 16). Lincoln: University of Nebraska Press. Lazarus, R. S., Averill, J. R., & Opton, E. M., Jr. (1970). Towards a cognitive theory of emotion. In M. Arnold (Ed.), Feelings and emotion (pp. 207–229). New York: Academic Press. Lee, C. (1989). Theoretical weaknesses and practical problems: The example of self-efficacy theory. Journal of Behavior Therapy and Experimental Psychiatry, 20, 115–123. Levine, S. (1960). Stimulation in infancy. Scientific American, 202, 80–86. Lewis, M. (1990). Challenges to the study of developmental psychopathology. In M. Lewis & S. M. Miller (Eds.), Handbook of developmental psychopathology (pp. 29–40). New York: Plenum. Lewis, M., Feirig, C., McGuffog, C., & Jaskir, J. (1984). Predicting psychopathology in six-year-olds from early social relations. Child Development. 55, 123–136. Lovibond, P. F., & Lovibond, S. H. (1995). The structure of negative emotional states: Comparison of the Depression Anxiety Stress Scales (DASS) with the Beck Depression and Anxiety Inventories. Behavior Research and Therapy, 33, 335–342. Maier, S. F. (1970). Failure to escape traumatic shock: Incompatible skeletal motor response or learned helplessness? Learning and Motivation, 1, 157–170. Maier, S. F., & Seligman, M. E. P. (1976). Learned helplessness: Theory and evidence. Journal of Experimental Psychology: General, 105, 3–46. Maier, S. F., & Testa, T. J. (1975). Failure to learn escape by rats previously exposed to inescapable shock is partly produced by associative interference. Journal of Comparative and Physiological Psychology, 88, 554–564. Main, M. (1996). Introduction to the special section on attachment and psychopathology: 2. Overview of the field of attachment. Journal of Consulting and Clinical Psychology. 64, 237–243. Main, M., & Solomon, J. (1990). Procedures for identifying infants as disorganized/disoriented during the Ainsworth Strange Situation. In M. Greenberg, D. Cicchetti, &. E. M. Cummings (Eds.), Attachment in the preschool years: Theory, research, and intervention (pp. 121–160). Chicago: University of Chicago Press.

The Development of Anxiety 261 Mandler, G. (1972). Helplessness: Theory and research in anxiety. In C. Spielberger (Ed.), Anxiety: Current trends in theory and research (Vol. 1, pp. 359–374). New York: Academic. Masten, A. S., Best, K. M., & Garmezy, N. (1991). Resilience and development: Contributions from the study of children who overcome adversity. Development and Psychopathology, 2, 425–444. McCauley, E., Mitchell, J. R., Burke, P., & Moss, S. (1988). Cognitive attributes of depression in children and adolescents. Journal of Consulting and Clinical Psychology, 56, 903–908. McNally, R. J., (1996). Cognitive bias in the anxiety disorders. In D. A. Hope (Ed.), Perspectives on anxiety, panic, and fear: Nebraska Symposium on Motivation (Vol. 43, pp. 211–250). Lincoln: University of Nebraska Press. Messer, S. C., & Beidel, D. C. (1994). Psychosocial correlates of childhood anxiety disorders. Journal of the American Academy of Child and Adolescent Psychiatry, 33, 975–983. Miller, N. E. (1980). A perspective on the effects of stress and coping on disease and health. In S. Levine & H. Ursin (Eds.), Coping and health (pp. 323–354). New York: Plenum Press. Mineka, S. (1985). Animal models of anxiety-based disorders: Their usefulness and limitations. In A. H. Tuma & J. D. Maser (Eds.), Anxiety and the anxiety disorders (pp. 199–244). Hillsdale, NJ: Erlbaum. Mineka, S., Gunnar, M., & Champoux, M. (1986). Control and early socioemotional development: Infant rhesus monkeys reared in controllable versus uncontrollable environments. Child Development, 57, 1241–1256. Mineka, S., & Zinbarg, R. (1996). Conditioning and ethological models of anxiety disorders. In D. A. Hope (Ed.), Nebraska Symposium on Motivation: Perspectives on anxiety, panic, and fear (Vol. 43, pp. 135–210). Lincoln: University of Nebraska Press. Minor, T. R., Dess, N. K., & Overmier, J. B. (1991). Inverting the traditional view of “learned helplessness.” In M. R. Denny (Ed.), Fear, avoidance and phobias (pp. 87–133). Hillsdale, NJ: Erlbaum. Mowrer, O. H., & Viek, P. (1948). An experimental analogue of fear from a sense of helplessness. Journal of Abnormal Social Psychology, 83, 193–200. Nachmias, M., Gunnar, M., Marigelsdorf, S., Parritz, R. H., & Buss, K. (1996). Behavioral inhibition and stress reactivity: The moderating role of attachment security. Child Development, 67, 508–522. Nolen-Hoeksema, S., Girgus, J. S., & Seligman, M. E. P. (1986). Learned helplessness in children: A longitudinal study of depression, achievement, and attributional style. Journal of Personality and Social Psychology, 51, 435–442. Nolen-Hoeksema, S., Girgus, J. S., & Seligman, M. E. P. (1992). Predictors and consequences of childhood depressive symptoms: A 5-year longitudinal study. Journal of Abnormal Psychology, 101, 405–422. Nolen-Hoeksema, S., Wolfson, A., Mumme, D., & Guskin, K. (1995). Helplessness in children of depressed and nondepressed mothers. Developmental Psychology, 31, 377–387. Nowicki, S., & Strickland, B. R. (1973). A locus of control scale for children. Journal of Consulting and Clinical Psychology, 40, 148–154. Nunn, G. D. (1988). Concurrent validity between the Nowicki-Strickland Locus of Control Scale and the State-Trait Anxiety Inventory for Children. Educational and Psychological Measurement, 48, 435–438. O’Hanlon, J. F., & Beatty, J. (1976). Catecholamine correlates of radar monitoring performance. Biological Psychology, 4, 293–304.

262 Bruce F. Chorpita and David H. Barlow Overmier, J. B. (1988). Psychological determinants of when stressors stress. In D. Hellhammer, I. Florin, & H. Weiner (Eds.), Neurobiological approaches to human disease (pp. 236–259). Toronto: Hans Huber. Overmier, J. B., & Seligman, M. E. P. (1967). Effects of inescapable shock upon subsequent escape and avoidance behavior, Journal of Comparative and Physiological Psychology, 63, 23–33. Parker, G. (1979a). Parental characteristics in relation to depressive disorders. British Journal of Psychiatry, 134, 138–147. Parker, G. (1979b). Parental deprivation and depression in a non-clinical group. Australian and New Zealand Journal of Psychiatry, 13, 51–56. Parker, G. (1979c). Reported parental characteristics in relation to trait depression and anxiety levels in a non-clinical group. Australian and New Zealand Journal of Psychiatry, 13, 260–264. Parker, G. (1979d). Reported parental characteristics of agoraphobics and social phobics. British Journal of Psychiatry, 135, 555–560. Parker, G. (1981). Parental representation of patients with anxiety neurosis. Acta Psychiatrica Scandinavica, 63, 33–36. Parker, G. (1983). Parental overprotection: A risk factor in psychosocial development. New York: Grune and Stratton. Parker, G., Tupling, H., & Brown, L. B. (1979). A parental bonding instrument. British Journal of Psychiatry, 52, 1–11. Perris, C., Jacobsson, L., Lindstrom, H., von Knorring, L., & Perris, H. (1980). Development of a new inventory for assessing memories of parental rearing behavior. Acta Psychiatrica Scandinavica, 61, 265–274. Peterson, C., Maier, S. F., & Seligman, M. E. P. (1993). Learned helplessness: A theory for the age of personal control. New York: Oxford University Press. Plantes, M. M., Prusoff, B. A., Brennan, J., & Parker, G. (1988). Parental representations of depressed outpatients from a USA sample. Journal of Affective Disorders, 15, 149–155. Puig-Antich, J., & Chambers, W. (1978). The Schedule for Affective Disorders and Schizophrenia for School-Age Children (Kiddie-SADS). New York: New York State Psychiatric Institute. Raab, A., & Oswald, R. (1980). Coping with social conflict: Impact on the activity of tyrosine hydroxylase in the limbic system and in the adrenals. Physiology and Behavior, 24, 387–394. Rachman, S. J. (1979). The concept of required helpfulness. Behaviour Research and Therapy, 17, 1–6. Radke-Yarrow, M., Cummings, E. M., & Kuczynski, L. (1985). Patterns of attachment in two- and three-year-olds in normal families and families with parental depression. Child Development, 56, 884–893. Rapee, R. M., Craske, M. G., Brown, T. A., & Barlow, D. H. (1996). Measurement of perceived control over anxiety related events. Behavior Therapy, 27, 279–293. Ray, J. C., & Sapolsky, R. M. (1992). Styles of male social behavior and their endocrine correlates among high-ranking wild baboons. American Journal of Primatology, 28, 231–250. Reiss, D., Hetherington, E. M., Plomin, R., Howe, G. W., Simmens, S. J., Henderson, S. H., O’Connor, T. J., Bussell, D. A., Anderson, E. R., & Law, T. (1995). Genetic questions for environmental studies: Differential parenting and psychopathology in adolescence. Archives of General Psychiatry, 52, 925–936. Rholes, W. S., Blackwell, J., Jordan, C., & Walters, C. (1980). A developmental study of learned helplessness. Developmental Psychology, 16, 616–624. Robins, C. J, & Hinkley, K. (1989). Social-cognitive processing and depressive symptoms in children: A comparison of measures. Journal of Abnormal Child Psychology, 17, 29–36.

The Development of Anxiety 263 Rose, R., Bernstein, I., Gordon, T., & Catlin, S. (1974), Androgen and aggression: A review and recent findings in primates. In R. Holloway (Ed.), Primate aggression, territoriality, and xenophobia (pp. 271–293). New York: Academic Press. Rotter, J. B. (1954). Social learning and clinical psychology. Englewood Cliffs, NJ: PrenticeHall. Rotter, J. B. (1966). Generalized expectancies for internal versus external control of reinforcement. Psychological Monographs, 80 (No. 609). Rutter, M. (1980). Attachment and the development of social relationships. In M. Rutter (Ed.), Scientific foundation of developmental psychiatry. London: Heinemann. Sanderson, W. C., Rapee, R. M., & Barlow, D. H. (1989). The influence of an illusion of control on panic attacks induced via the inhalation of 5.5% carbon dioxide enriched air. Archives of General Psychiatry, 46, 157–164. Sapolsky, R. M. (1989). Hypercortisolism among socially subordinate wild baboons originates at the CNS level. Archives of General Psychiatry, 46, 1047–1051. Sapolsky, R. M., & Ray, J. C. (1989). Styles of dominance and their endocrine correlates among wild olive baboons (Papio anubis). American Journal of Primatology, 18, 1–13. Schneewind, K. A. (1995). Impact of family processes on control beliefs. In A. Bandura (Ed.), Self-efficacy in changing societies (pp. 114–148). New York: Cambridge University Press. Schneewind, K. A., & Pfeiffer, P. (1978). Elterliches Erziehungsverhalten und kindliche Selbstverantwortlichkeit [Parenting and children’s self-responsibility]. In K. A. Schneewind & H. Lukesch (Eds.), Familiäre Sozialisation [Family Socialization] (pp. 190–205). Stuttgart, Germany: Klett-Cotta. Schweizer, E. E., Swenson, C. M., Winokur, A., Rickels, K., & Maislin, G. (1986). The dexamethasone suppression test in generalized anxiety disorder. Archives of General Psychiatry, 149, 320–322. Seligman, M. E. P., & Maier, S. F. (1967). Failure to escape traumatic shock. Journal of Experimental Psychology, 74, 1–9. Seligman, M. E. P., Peterson, C., Kaslow, N. J., Tanenbaum, R. L., Alloy, L. B., & Abramson, L. Y. (1984). Attributional style and depressive symptoms among children. Journal of Abnormal Psychology, 93, 235–238. Seligman, M. E. P., Rosellini, R. A., & Kozak, M. (1975). Learned helplessness in the rat: Reversibility, time course, and immunization. Journal of Comparative and Physiological Psychology, 88, 542–547. Shelton, M. R., & Garber, J. (1987, August). Development and validation of a children’s pleasant and unpleasant events schedule. Paper presented at the 95th Annual Convention of the American Psychological Association, New York. Silove, D., Parker, G., Hadzi-Pavlovic, D., Manicavasagar, V., & Blaszczynski. A. (1991). Parental representations of patients with panic disorder and generalised anxiety disorder. British Journal of Psychiatry, 159, 835–841. Skinner, E. A. (1986). The origins of young children’s perceived control: Mother contingent and sensitive behavior. International Journal of Behavioral Development, 9, 359–382. Skinner, E. A., Chapman, M., & Baltes, P. B. (1988). Control, means-ends and agency beliefs: A new conceptualization and its measurement during childhood. Journal of Personality and Social Psychology, 54, 117–133. Sroufe, L. A. (1979). The coherence of individual development. American Psychologist, 34, 134–141. Sroufe, L. A. (1983). Infant-caregiver attachment and patterns of adaptation in preschool: The roots of maladaptation and competence. In M. Perlmutter (Ed.), Minnesota symposium on child psychology (Vol. 16, pp. 41–83). Hillsdale, NJ: Erlbaum.

264 Bruce F. Chorpita and David H. Barlow Sroufe, L. A. (1990). Considering the normal and abnormal together: The essence of developmental psychopathology. Development and Psychopathology, 2, 335–347. Stroebel, C. F. (1969). Biologic rhythm correlates of disturbed behavior in the rhesus monkey. Bibliotheca Primatologica, 9, 91–105. Thompson, R. A. (1998). Early sociopersonality development. In W. Damon (Series Ed.) & N. Eisenberg (Vol. Ed.), Handbook of child psychology: Vol. 3. Social, emotional, and personality development (pp. 25–104). New York: Wiley. Turner, J. E., & Cole, D. A. (1994). Developmental differences in cognitive diatheses for child depression. Journal of Abnormal Child Psychology, 22, 15–32. Turner, S. M., Beidel, D. C., & Costello, A. (1987). Psychopathology in the offspring of anxiety disorders patients. Journal of Consulting and Clinical Psychology, 55, 229–235. Urban, J., Carlson, E., Egeland, B., & Sroufe, L. A. (1992). Patterns of individual adaptation across childhood. Development and Psychopathology, 3, 445–460. van IJzendoorn, M. H., & Bakermans-Kranenburg, M. J. (1996). Attachment representations in mothers, fathers, adolescents, and clinical group: A meta-analytic search for normative data. Journal of Consulting and Clinical Psychology, 64, 8–21. Walter, D. A., & Ziegler, D. A., (1980). The effects of birth order on locus of control. Bulletin of the Psychonomie Society, 15, 293–294. Washington, R. A. (1974). The relationship between children’s school performances and parent participation as a function of the movement toward community control of the schools. Dissertation Abstracts International, 35, 4202B. Waters, E., Vaughn, B. E., Posada, G., & Kondo-Ikemura, K. (Eds.). (1995). Caregiving, cultural, and cognitive perspectives on secure-base behavior and working models: New growing points of attachment theory and research. Monographs of the Society for Research in Child Development, 60(2–3, Serial No. 244). Waters, E., Wippman, J., & Sroufe, L. A. (1979). Attachment, positive affect, and competence in the peer group: Two studies in construct validation. Child Development, 50, 821–829. Watson, D., & Clark, L. A. (1984). Negative affectivity: The disposition to experience negative emotional states. Psychological Bulletin, 96, 465–490. Weiss, J. M. (1971a). Effects of coping behavior in different warning signal conditions on stress pathology in rats. Journal of Comparative and Physiological Psychology, 77, 1–13. Weiss, J. M. (1971b). Effects of punishing the coping response (conflict) on stress pathology in rats. Journal of Comparative and Physiological Psychology, 77, 14–21. Weiss, J. M. (1971c). Effects of coping behavior with and without a feedback signal on stress pathology in the rat. Journal of Comparative and Physiological Psychology, 77, 22–30. Weisz, J. R. (1986). Understanding the developing understanding of control. In M. Perlmutter (Ed.), Cognitive perspectives on children’s social and behavioral development: The Minnesota Symposia on Child Psychology (Vol. 18, pp. 219–285). Hillsdale, NJ: Erlbaum. Weisz, J. R., & Stipek, D. J. (1982). Competence, contingency, and the undevelopment of perceived control. Human Development, 25, 250–281. Williams, J. L., & Lierle, D. M. (1986). Effects of stress controllability, immunization, and therapy on the subsequent defeat of colony intruders. Animal Learning and Behavior, 14, 305–314. Williams, J. L., & Scott, D. K. (1989). Influence of conspecific and predatory stressors and their associated odors on defensive burying and freezing responses. Animal Learning and Behavior, 17, 383–393.

Article 15

A Modern Learning Theory Perspective on the Etiology of Panic Disorder Mark E. Bouton University of Vermont

Susan Mineka Northwestern University

David H. Barlow Boston University

Several theories of the development of panic disorder (PD) with or without agoraphobia have emerged in the last 2 decades. Early theories that proposed a role for classical conditioning were criticized on several grounds. However, each criticism can be met and rejected when one considers current perspectives on conditioning and associative learning. The authors propose that PD develops because exposure to panic attacks causes the conditioning of anxiety (and sometimes panic) to exteroceptive and interoceptive cues. This process is reflected in a variety of cognitive and behavioral phenomena but fundamentally involves emotional learning that is best accounted for by conditioning principles. Anxiety, an anticipatory emotional state that functions to prepare the individual for the next panic, is different from panic, an emotional state designed to deal with a traumatic event that is already in progress. However, the presence of conditioned anxiety potentiates the next panic, which begins the individual’s spiral into PD. Several biological and psychological factors create vulnerabilities by influencing the individual’s susceptibility to conditioning. The relationship between the present view and other views, particularly those that emphasize the role of catastrophic misinterpretation of somatic sensations, is discussed. Early learning theorists such as Pavlov (1927), Watson (J. B. Watson & Rayner, 1920), and later Mowrer (1947) and Solomon (e.g., Solomon, Kamin, & Wynne, 1953) were highly interested in the relevance of their work on conditioning and

266 Mark E. Bouton, et al.

learning to understanding the genesis of what was then called neurotic behavior in humans. Other investigators, using a variety of experimental paradigms that induced so-called experimental neurosis in animals (e.g., Pavlov, Gantt, Liddell, and Masserman, to name a few; see Mineka & Kihlstrom, 1978), also assumed that their work would be directly relevant to human neurotic behavior. Unfortunately, enthusiasm for this work on the part of psychopathology researchers began to wane in the 1970s as criticisms of the applicability of this earlier work to human neuroses mounted (see Rachman, 1977, 1990, for reviews).1 At the same time, a virtual revolution in the field of learning was occurring, focusing on the development of new paradigms and new theoretical developments about the nature of the associative learning process. However, most later learning theorists were not as interested as the founding fathers in the relevance of their findings for understanding psychopathology, and the two fields went their separate ways with little cross-fertilization of ideas. Over the past 15 years, a goal of some of our work has been to reinvigorate this cross-fertilization by demonstrating the relevance of newer perspectives on learning theory to understanding anxiety and the anxiety disorders (e.g., Barlow, 1988; Barlow, Chorpita, & Turovsky, 1996; Bouton, 1988, 1991b, 2000; Bouton & Nelson, 1998b; Bouton & Swartzentruber, 1991; Chorpita & Barlow, 1998; Mineka, 1985a, 1985b; Mineka & Zinbarg, 1991, 1995, 1996, 1998). The major goal of the present article is to spell out the relevance of some contemporary work on classical conditioning, and learning theory more broadly, to understanding the etiology and maintenance of anxiety disorders with an emphasis on one of the more common anxiety disorders: panic disorder (PD). The study of anxiety disorders and their treatment expanded steadily in the 1970s and 1980s; the National Institute of Mental Health dubbed the 1980s the “decade of anxiety” (e.g., Rachman & Maser, 1988; Turna & Maser, 1985). The efficacy of treatments for most of these disorders represents one of the great success stories of applied psychological science (e.g., Barlow & Lehmann, 1996). However, although our understanding of basic behavioral and cognitive processes underlying panic and anxiety likewise advanced significantly, the underlying development of theories focused on the etiology of panic disorder has not advanced appreciably since the 1980s. In our view, psychological theories of PD remain in an early developmental stage, as is the case with more biological theories, at least in part because they have not incorporated the latest developments in the basic behavioral, cognitive, and neural sciences or the integration of any of these fields. As noted, this is one purpose of this article. After briefly describing the disorder, as well as current theoretical perspectives on it, we proceed to an articulation of a new contemporary learning theory perspective on PD that does incorporate developments from these other fields. PD is a clinical syndrome that provides a particularly appropriate context in which to examine the relevance of basic principles of conditioning and learning for the development and maintenance of psychopathology. In this anxiety disorder, the fundamental emotional constructs of anxiety, panic, and

A Modern Learning Theory Perspective on the Etiology 267

consequent phobic avoidance come together in an intricate relationship that can devastate the lives of those who develop it in severe form. Some patients are confined to their homes as virtual prisoners for years on end. To meet criteria for PD, a person must experience recurrent unexpected panic attacks that occur without any obvious cues or triggers, such as encounters with a frightening object or experiencing a traumatic event. In other words, the panic attack seems to come out of the blue. A panic attack is an abrupt experience of intense fear or discomfort accompanied by a number of physical and mental symptoms, most usually heart palpitations, chest pain, sensations of shortness of breath, dizzy feelings, and thoughts of going crazy, losing control, or dying. Significantly, the attack must develop abruptly and reach a peak within 10 min (American Psychiatric Association, 1994). The experience of unexpected panic attacks is not enough to meet criteria for PD, however. In addition, the individual must develop substantial anxiety or concern over the possibility of having another attack or about the implications of the attack or its consequences (such as the attack leading to a heart attack, to “going crazy,” or to “losing control”). This is an important criterion, because we now know that numerous individuals experience “nonclinical panic attacks,” in which a susceptible individual under stress may experience a sudden jolt of unexpected panic but fail to develop anxiety about a possible subsequent attack and its consequences, attributing it instead to benign events of the moment. Many people with PD develop the complication of agoraphobia, which literally refers to fear of the marketplace. Agoraphobia, the most severe of all phobias, describes anxiety about being in places or situations from which escape might be difficult or embarrassing or in which help may not be available in the event of an unexpected panic attack (or panic-like symptoms). Typically, agoraphobic fears occur in clusters of situations that include being outside of the home alone, being in a crowded church or shopping mall, or being on public transportation (buses, trains, or airplanes). Agoraphobic avoidance behavior is simply one of the learned consequences of having severe unexpected panic attacks. If you have an unexpected panic attack and are afraid you may have another one, it is not surprising that you want to be in a safe place or at least with a safe person who either can help or knows what you are experiencing in the event another attack occurs. Not everyone develops agoraphobia, and in those who do, agoraphobic avoidance may develop along a continuum from mild to severe. Thus, PD may occur with or without agoraphobic avoidance. For those individuals who do not develop agoraphobic avoidance, it is characteristic to find other associated behavioral coping tendencies such as resorting to the use (and often the abuse) of drugs and/or alcohol to self-medicate anxiety and panic. Avoidance of behaviors or activities that might provoke somatic symptoms similar to those that occur during panic is also common. For example, patients may avoid exercising (which may produce breathlessness or trembling and perspiration), sexual relations (which may produce other similar physical symptoms), or frightening movies (which may produce emotional symptoms that are reminiscent of panic attacks). PD is fairly common; approximately 3.5% to 5.3% of the population meet the criteria for PD at some point during their

268 Mark E. Bouton, et al.

lives (Kessler et al., 1994). Onset usually occurs in early adult life from the midteenage years to about 40 years, although it can occur in children as well (Barlow, in press). Current Theories of PD Currently, there are at least three prominent psychological theories about the origins of PD. Two of them (cognitive theory and anxiety sensitivity theory) emphasize cognitive aspects of the disorder. An earlier conditioning theory was heavily criticized and has stimulated less research in recent years. We now briefly review these approaches and illustrate the criticisms that have been raised for each (see Thorn, Chosak, Baker, & Barlow, 1999, for a more detailed review). Cognitive Theories

Cognitive theories, which are most closely associated with D. M. Clark (1986, 1988, 1996) and Beck (e.g., Beck & Emery, 1985), see an individual’s “catastrophic misinterpretations” of somatic and other sensations as crucial to the development and maintenance of PD (see also Salkovskis, 1988). Specifically, an individual will experience panic when his or her focus on internal bodily sensations (whether produced by anxiety or not) leads to catastrophic thoughts about their imminent meaning (e.g., “I am going to have a heart attack”). Such catastrophic thoughts, which are anxiety-producing themselves, lead to further bodily sensations, which provide more fuel for more catastrophic thoughts, and thus a vicious cycle culminating in a panic attack. The internal focus on somatic and other sensations leads to chronic vigilance and increased sensitivity to otherwise normal physical sensations. Evidence in support of this theory includes the effects of direct manipulations of catastrophic cognitions in the laboratory. If such cognitions are hypothetically stimulated (as, e.g., by reading pairs of words consisting of various combinations of bodily sensations and catastrophes, such as palpitations-die or breathless-suffocate), the probability of panic increases (D. M. Clark et al., 1988). If such cognitions are reduced, the probability of panic in response to panic-provocation agents decreases (Clark, 1996). Further indirect evidence stems from the success of cognitive therapy in alleviating PD (e.g., D. M. Clark et al., 1994). Problems for cognitive theory include the fact that panic attacks can occur in panic patients in the absence of detectable catastrophic cognitions. For example, patients may experience nocturnal panic attacks (usually during the transition from Stage 2 to Stage 3 sleep rather than during rapid eye movement sleep, when most dreaming occurs). Alternatively, they may sometimes have diurnal attacks without antecedent cognitions when the cognitions should have been readily detectable (e.g., Rachman, Lopatka, & Levitt, 1988; Zucker et al., 1989), even in studies using prospective self-monitoring of panic attacks (e.g., Kenardy, Fried, Kraemer, & Taylor, 1992; Kenardy & Taylor, 1999). The phenomenon of “nonfearful” panic, in which patients develop the disorder without having cognitions of danger or threat, also seems problematic for this theory (Barlow, Brown, & Craske, 1994; Kushner & Beitman, 1990). Proponents of the cognitive model might respond by suggesting that catastrophic

A Modern Learning Theory Perspective on the Etiology 269

misinterpretations in these cases may be very quick and below the threshold of awareness, even occurring during the transition between Stage 2 and Stage 3 sleep. However, without an independent measure of catastrophic misinterpretations (other than the panic attacks themselves), this sort of response begins to make the theory appear untestable (McNally, 1994, 1999). It has also been claimed that cognitive models are somewhat vague in specifying or operationalizing terms such as “catastrophic misinterpretations” (Seligman, 1988; Teasdale, 1988). These terms, noted D. M. Clark (1996), “are expressed in everyday language rather than in the more precise technical terms that characterize many models in cognitive psychology (see Teasdale, 1988). . . . the advantage of such a model is the ease with which it suggests specific clinical procedures. The disadvantage is that it can be more difficult to test because it is not always clear how to operationalize key terms in a way that allows them to be precisely measured and manipulated” (p. 319). We suggest that the vagueness is not in the content of the cognitions, but rather how they are acquired, who acquires them, how they can be measured independently of panic itself, and under what conditions they become “catastrophic.” Two other problems with this approach are worth noting. First, as we discuss later, although catastrophic cognitions often may occur in panic patients, it is not necessarily clear that they play a causal role in creating panic attacks. Second, the cognitive model sees little need to distinguish between the emotional states of panic and anxiety, despite growing evidence to the contrary, described below. Anxiety Sensitivity Theory

Anxiety sensitivity (AS) theory posits the existence of a traitlike belief in some individuals, especially including patients with panic disorder. The essence of this belief is that anxiety and its associated symptoms, particularly somatic symptoms, may cause deleterious physical, psychological, or social consequences that extend beyond any immediate physical discomfort during an episode of anxiety or panic itself. Proponents of AS theory, such as Reiss (1991) and McNally (1994), differentiate this theory from other perspectives in several ways. First, they clearly see AS as an enduring traitlike tendency. Second, AS theorists argue that individuals with PD are often fully aware of the causes of their sensations (i.e., they do not misinterpret them) and yet are frightened by them because they still hold an inherent belief that the sensations are harmful to their body or mental state. (Cognitive theorists might claim that it is the sensations and/or the immediate consequences of the sensations, e.g., fainting, heart attack, that are “misinterpreted.”) Moreover, the two theories clearly emphasize different time perspectives regarding the consequences of the sensations, with cognitive theory emphasizing the idea of immediate impending disaster and AS theory emphasizing that harm or danger from the symptoms may accumulate over time. The AS model has generated a widely used measure, the Anxiety Sensitivity Index (ASI; Reiss, Peterson, Gursky, & McNally, 1986), and much thoughtful research. Generally, AS has been found to be normally distributed in the population, suggesting that it is a dimensional construct, and it is seen as a

270 Mark E. Bouton, et al.

vulnerability factor that increases the likelihood of developing an anxiety disorder, especially PD (Reiss, 1991). Yet questions remain about the cause of this trait as well as its precise role in the subsequent etiology of PD (as well as other anxiety disorders). Three studies lend support to anxiety sensitivity as a vulnerability factor for panic attacks. In the first study, Schmidt, Lerew, and Jackson (1997) found that higher initial scores on the ASI (but not trait anxiety scores) in military recruits predicted more anxiety and depressive symptoms and a greater number of unexpected panic attacks after a very stressful course of basic military training. Although the investigators noted that the relationships were relatively weak in this sample of well-adjusted military recruits, accounting for a rather small percentage of the variance, this study helps to confirm AS as reflecting at least one possible vulnerability for later anxiety, panic attacks, and depressive symptoms, triggered by extreme stress. However, this study did not provide evidence of a unique relationship between AS and panic attacks (or PD), in that high AS scores also predicted later anxiety and depressive symptoms more generally. In a similar second study, Schmidt et al. (1999) replicated these findings and found that anxiety sensitivity was more specifically related to anxiety/ panic symptoms compared with depressive symptoms, although the findings accounted for only 2% of the variance in predicting unexpected panic attacks. Finally, Hayward, Killen, Kraemer, and Taylor (2000), in a 4-year prospective study of high school adolescents, found that AS was a significant predictor of onset of panic attacks as defined by the Diagnostic and Statistical Manual of Mental Disorders, 3rd edition, revised (DSM-III-R; American Psychiatric Association, 1994) but not of major depression when changes in panic were measured while controlling for changes in depression. Although AS predicted changes in panic attacks but not major depression in this study, it was not the best predictor of panic attacks; instead, negative affectivity was the single biggest predictor of panic attacks (and major depression). (See also Ehlers, 1995.) Thus, questions remain both about the specificity of what AS predicts and about whether it is a better predictor of unexpected panic than negative affectivity (the Hayward et al. study suggests that it is not). Conditioning Theories

Conditioning theory has a long and distinguished tradition in helping to understand the etiology of anxiety disorders (e.g., Eysenck, 1979; Marks, 1969; Wolpe & Rowan, 1988), and it was one of the first types of theory applied to the cause of PD (e.g., Eysenck, 1960; Eysenck & Rachman, 1965). Generally, conditioning theories suggest that when stimuli, events, or situations (conditioned stimuli [CSs]) are paired with a panic attack (and all of its associated physiological sensations), the learning that may occur can allow the CSs to trigger panic and anxiety when they are encountered again. This sort of theory has taken a number of different forms when applied to PD. Early conditioning theories focused on the role of conditioning in the onset of agoraphobia or situational panic attacks (i.e., conditioning to external or exteroceptive cues). However, perhaps the best known version of conditioning theory applied to PD originated in an important article by Goldstein and Chambless (1978) that

A Modern Learning Theory Perspective on the Etiology 271

described a process they termed “fear of fear.” Citing Razran (1961), Goldstein and Chambless reintroduced the notion of interoceptive conditioning, in which low-level somatic sensations of anxiety or arousal effectively became CSs associated with higher levels of anxiety or arousal. Thus, they posited that early somatic components of the anxiety response can come to elicit significant bursts of anxiety or panic. (These were also expected to generalize to other stimuli.) Thus, the focus of conditioning theory changed from exteroceptive conditioning in explaining agoraphobia and situational panics to interoceptive conditioning in explaining the cause of more “spontaneous” or apparently uncued panic attacks. Criticisms of conditioning theories are presented in more detail under Clarifying the Role of Classic Conditioning in PD. To summarize briefly, interoceptive conditioning has been criticized as being conceptually confusing because anxiety or panic seem to serve somewhat indiscriminately as CS, unconditioned stimulus (US), conditioned response (CR), and unconditioned response (UR; McNally, 1990, 1994; Reiss, 1987). According to Reiss (1987), for example, anxiety seems to become conditioned to itself, but what does that mean? In addition, conditioning theory seems to lead to an overprediction of panic, because a fear or panic response should theoretically occur every time the CS (e.g., a somatic sensation, such as a quickened heart rate) is encountered (e.g., D. M. Clark, 1988). A third criticism stemmed from the observation that the fear or panic response does not seem subject to extinction after numerous natural trials in which arousal and the somatic cues it generates are not followed by panic (Rachman, 1991; Seligman, 1988; Van den Hout, 1988). However, a major point of the present article is that all of these concerns fall away when one considers a more modern perspective on classical conditioning and its many effects on emotion and behavior. Summary and Overview

PD is a common anxiety disorder in which unexpected panic attacks lead to excessive anxiety about future attacks. The approaches sketched previously also contain a number of unmentioned common elements, such as vague allusions to biological vulnerabilities as well as fundamental cognitive or psychological vulnerabilities predating the development of the panic attacks, or PD, but these elements are seldom elaborated. Moreover, each approach tends to highlight one aspect or another of the process, such as interoceptive fear conditioning, catastrophic misinterpretations of physical sensations, or beliefs about the dangers of anxiety, as though they are mutually exclusive. The present article provides an integrative theory of the etiology of PD that uses contemporary learning theory as its base. On the basis of information that is available in the psychometric, ethological, and neurobiological literatures, we distinguish between two aversive motivational states: anxiety and panic. Further, on the basis of the conditioning literature, we expect that a major effect of early experience with panic is the conditioning of anxiety to cues that are associated with the episode. One result is that, in the presence of the interoceptive or exteroceptive cues associated with panic, anxiety now occurs. As is widely recognized, the classical conditioning of anxiety also

272 Mark E. Bouton, et al.

makes it possible for new operant behaviors to be reinforced when they escape or reduce it (e.g., Mowrer, 1947). Conditioning of anxiety can also have other major consequences that are emphasized here. As we show later, the presence of conditioned anxiety may serve to exacerbate or potentiate the next panic attack, beginning the vicious spiral into PD. A second factor that can exacerbate the next panic attack is that panic itself may come to be elicited by a more specific set of cues associated with panic. The conditioning of panic responses to some cues may occur in parallel with the conditioning of anxiety to other cues; conditioning is a multifaceted process in which a range of interacting stimuli can acquire the ability to control a constellation or system of different emotions, cognitions, and behaviors. These themes, coupled with a modern understanding of conditioning processes that will further modulate and influence the course of conditioning, constitute the core of our approach to the development of PD. Other biological and psychological processes that are thought to influence PD (including catastrophic thoughts, AS, and other processes that appear to make certain individuals vulnerable) may be seen as operating through their interaction with this core conditioning process. Introduction to a Modern Learning Theory Perspective on PD Panic and Anxiety

Any theory of PD must acknowledge the strong and growing network of evidence suggesting fundamental differences between the emotional phenomena of panic and anxiety. Panic attacks have been defined as a subjective sense of extreme fear or impending doom accompanied by a massive autonomic surge and strong flight-or-fight behavioral action tendencies (Barlow, Brown, & Craske, 1994; American Psychiatric Association, 1994). Anxiety has been defined as an apprehensive anticipation of future danger, often accompanied by somatic symptoms of tension or feelings of dysphoria. The focus of anticipated danger can be internal or external. Phenomenological as well as neurobiological evidence suggests that panic attacks are descriptively and functionally distinct events when compared with anxiety. Some of this evidence comes from detailed analyses of fear and anxiety in normal populations, in which the findings support these constructs as being partially overlapping and yet partially distinctive at a psychometric level. In early studies, fear emerged as a partially distinct primary (lower order) factor, but it also loaded on the higher order factor of negative affect (L. A. Clark & Watson, 1991; Tellegen, 1985; D. Watson & Clark, 1984). These lower and higher order factors not only seem separable but also relate in rather different ways to other anxiety disorders and depression. For example, more recent structural equation modeling and factor analyses of symptomatology exploring the dimensions of panic, anxiety, and depression in clinically anxious patients have also uncovered two different factors. One, which is characterized by a subjective sense of extreme fear or impending doom, strong autonomic arousal, and strong flight-or-fight behavioral action tendencies, seems best described as panic.

A Modern Learning Theory Perspective on the Etiology 273

The other is characterized by the kinds of apprehension and worry accompanied by tension that are best described as anxiety (T. A. Brown, Chorpita, & Barlow, 1998; Joiner et al., 1999; Mineka, Watson, & Clark, 1998; Zinbarg et al., 1994). Anxiety, a principal component of generalized anxiety disorder, is very closely related to depression. Spikes of strong autonomic arousal characteristic of panic, in contrast, are substantially restricted in patients experiencing more generalized anxiety (T. A. Brown et al., 1998; Borkovec, 1994; Hoehn-Saric, McLeod, & Zimmerli, 1989; Zinbarg et al., 1994). Also, these different emotional experiences may produce differential regional brain activity as measured by electroencephalography (Heller, Nitschke, Etienne, & Miller, 1997). Ethological evidence concerning specific defensive reactions such as the flight-or-fight response points to their survival value and heritability as well as their distinctiveness (and partial separateness at the level of heritability; Kendler, Walters, et al., 1995) compared with more diffuse states analogous to generalized anxiety or depressed mood. Fanselow and Lester (1988), among others, noted the qualitative differences in animal defensive behavior that occur depending on the “imminence” or proximity of threat (see also D.C. Blanchard, 1997; R. J. Blanchard & D. C. Blanchard, 1987). In rodents, freezing may occur after a potential threat has been detected, whereas more active behaviors occur when a predator actually attacks. Ethologists have pointed to hierarchical arrangements between action tendencies of freezing and heightened vigilance, which may represent anxiety (see later discussion), and other more proximal action tendencies that may be associated with fear and panic (e.g., Gallup & Maser, 1977; Gray & McNaughton, 1996). That is, freezing (anxiety) may antedate and potentiate other behaviors that occur as a threatening predator approaches more closely. There is also increasing evidence from neurobiology supporting the existence of at least two negative emotional states mediated by different brain circuits and potentially representing anxiety and panic (Charney, Grillon, & Bremner, 1998; Gray & McNaughton, 1996; Heller et al., 1997; Lang, Davis, & Öhman, 2000; White & Depue, 1999). A great deal of evidence suggests the role of the amygdala and related areas of the brain in fear and defensive behavior (e.g., Davis, 1992; Fanselow, 1994; Kapp, Whalen, Supple, & Pascoe, 1992; LeDoux, 1996; Rosen & Schulkin, 1998). Fanselow (1994) proposed two related but separable brain circuits that are implicated in different defensive emotional reactions. The first circuit, involving the central amygdala and the ventral periaqueductal gray, mediates freezing and opiate-mediated analgesia. (Other output from the amygdala mediates autonomic responses [see Kapp et al., 1992] and potentiated startle responses [see Davis, 1992]). These behaviors and autonomic responses correspond to anxiety in our approach. The second circuit identified by Fanselow, involving the dorsolateral periaqueductal gray and superior colliculus, mediates more active defensive behaviors, such as flight-or-fight and non-opiate-mediated analgesia. These behaviors are associated with what Fanselow called “circa-strike” defense, which has evolved to deal with actual predatory attack and corresponds to panic in our approach. Other investigators, including Davis (e.g., Davis, Walker, & Yee, 1997; see also Rosen & Schulkin, 1998) and Gray and McNaughton (1996; McNaughton & Gray, in press) have also recently distinguished between two aversive

274 Mark E. Bouton, et al.

motivational systems and circuits. Although there are differences in these various approaches, it is now common to distinguish between at least two negative emotional neural systems and the corresponding behavioral systems they control. Thus, many neuroscientists would now agree with psychopathologists that there are important distinctions to be made between emotional states that may correspond to anxiety and panic. True Alarms, False Alarms, and Learned Alarms

Barlow (1988; Barlow et al., 1996) proposed a conditioning theory of clinical anxiety disorders that accepts the distinction between panic and anxiety. This theory, which has come to be known as “alarm theory” (Carter & Barlow, 1995; Forsyth & Eifert, 1996), begins with the observation that the experience of unexpected panic attacks seems to be relatively common in the population at large (e.g., Norton, Cox, & Malan, 1992; Norton, Dorward, & Cox, 1986; Telch, Brouillard, Teich, Agras, & Taylor, 1989; Wittchen & Essau, 1991). Evidence suggests that these attacks seldom progress to PD and, therefore, are referred to as “nonclinical” attacks. Individuals experiencing nonclinical panic attacks show little or no concern over the possibility of experiencing additional attacks. Rather, they seem to dismiss the attacks, for the most part, as associated with some trivial and potentially controllable event. Other evidence suggests that panic attacks may be a nonspecific response to stress, similar to hypertension or headaches, that runs in families (Barlow et al., 1996). Individuals with PD are clearly discriminable from nonclinical panickers in that they develop anxiety focused on the next potential panic attack. These individuals anticipate the next attack apprehensively, perceive the attacks as uncontrollable and unpredictable, and are extremely vigilant for somatic symptoms that might signal the beginning of the next attack. In this formulation, the crucial step in the development of PD is the conditioning of anxiety focused on the next panic attack, occurring most often in those with a preexisting vulnerability to make this association (see later discussion). The model assumes that a panic attack (the flight-or-fight response) is the fundamental emotion of unconditioned fear occurring at the wrong time. When the danger is real, the responses are “true alarms.” Thus, alarms are “false” if they occur when there is nothing to fear (no danger). Because there is nothing to fear, a false alarm often occurs to the great surprise of the individual experiencing one. It has been widely recognized that it is adaptive to learn to associate emotional responses very quickly with both discrete and contextual cues. When these alarms, true or false, are associated with, or conditioned to, external or internal cues, they become “learned alarms.” When the clinical disorder is a specific phobia, it seems clear that fear or anxiety is conditioned to an otherwise harmless object or situation. In contrast, in PD the cues eliciting fear or anxiety are often more diffuse (being away from a safe place or a safe person) or difficult to pinpoint. Thus, following Goldstein and Chambless (1978), Barlow proposed that false alarms could be conditioned to internal physiological stimuli reflecting the process of interoceptive conditioning (Razran, 1961). The occurrence of false alarms and subsequent learned alarms need not be pathological if the alarms are

A Modern Learning Theory Perspective on the Etiology 275

infrequent and anxiety focused on the possible occurrence of future alarms does not develop. It is the development of anxiety that fundamentally produces vigilance for somatic sensations, increased tension and arousal, a resulting increase in somatic cues, and a spiraling of anxiety and panic. Anxiety focused on a possible future panic attack is now part of the defining DSM-IV criteria for PD (American Psychiatric Association, 1994). As noted earlier, the role of conditioning in PD has produced much controversy and confusion, which was not explicitly addressed by alarm theory. Therefore, we now consider some recent developments in our knowledge of conditioning and emotional learning for a fuller explication of this aspect of development of PD and other clinical anxiety disorders. Clarifying the Role of Classical Conditioning in PD Although panic attacks do not inevitably lead to conditioning, we assume that a crucial element in the origin of PD is a conditioning episode (or episodes) involving early panic attacks (usually false alarms as opposed to true alarms reflecting confrontation with a traumatic event), often enabled or potentiated by biological factors and/or the presence of stress (e.g., Barlow, 1988; Barlow et al., 1996). As we review later, a large percentage of patients with PD do point to an initial panic attack at the beginning of the disorder (e.g., Craske, Miller, Rotunda, & Barlow, 1990; Öst & Hugdahl, 1983). Either the panic attack itself (e.g., Forsyth & Eifert, 1996, 1998) or the event that actually triggers it (if indeed there is an identifiable one) then becomes associated with initially neutral cues through the associative learning process known as classical conditioning. Those cues could include proximal or distal cues, such as specific social situations (e.g., eating in restaurants, going to church) or entering general shopping malls or sports arenas, and also interoceptive and exteroceptive stimuli that are more directly involved in panic, such as the bodily sensations arising from the increase in respiration that occurs with hyperventilation. Most likely, both types of cues are involved. The consequences of this conditioning are spelled out in detail in the next sections. As a preliminary to this discussion, we note that classical conditioning has many characteristics and consequences that are not widely understood outside a relatively small community of specialists. Many psychologists now know that modern views of conditioning are “cognitive” in the sense that they use theoretical constructs like attention, surprise, information value, short-term memory, rehearsal, and so forth to explain even simple conditioning (e.g., Mackintosh, 1975; Pearce & Hall, 1980; Rescorla & Wagner, 1972; Wagner, 1976, 1981). We accept and embrace this view. However, we do not accept the idea that a modern view of conditioning implies that conditioning trials inevitably give rise to a kind of propositional, declarative knowledge that is the same as that created by verbal input (e.g., Lovibond, 1993). Conditioning theories themselves are agnostic on the issue. We suspect that, although conditioning and propositional systems may sometimes interact (e.g., Grings, Schell, & Carey, 1973), the extent to which they do will depend on the type of conditioning system, the specific conditioning procedure, and the brain systems they engage (Bechara et al., 1995; R. E. Clark & Squire, 1998; LeDoux, 1996).

276 Mark E. Bouton, et al.

LeDoux (1996) and Öhman, Flykt, and Lundqvist (2000) argued that aversive emotional learning can occur without any conscious representation of the learning. This means that implicit emotional memories can activate the fear system without the person necessarily having any conscious recollection or awareness of why they are aroused. A number of findings are consistent with this possibility, such as the fact that classically conditioned emotional responses are not always influenced by verbal instruction (e.g., Bridger & Mandel, 1964; Hamm & Vaitl, 1996) and that emotional conditioning does not appear to depend on conscious awareness (see LeDoux, 1996; Öhman et al., 2000; Öhman & Mineka, 2001). Data reported by Bechara et al. (1995) provide an especially compelling dissociation between emotional conditioning and verbal, declarative knowledge. When given a classical conditioning experience, a patient with a damaged amygdala did not acquire classical autonomic conditioning but was able to report what CSs predicted the US. Conversely, a patient with a damaged hippocampus acquired the autonomic responding but was unable to report what CSs predicted the US. Such findings encourage the view that conscious, declarative, or propositional knowledge about conditioning contingencies is not necessary or sufficient for emotional conditioning. This is not to claim that cognitions are irrelevant in PD, however. We have already noted that PD may often involve catastrophic thoughts in which the patient believes that the somatic symptoms will lead to a heart attack, death, and so forth. Whether such cognitions have a truly causal role in creating panic attacks is not currently clear. However, the idea that they may sometimes play a causal role (albeit through a mechanism different from the one posited by cognitive theory) would not necessarily be inconsistent with a learning theory perspective on PD. Such thoughts may first arise during a panic attack either because they are an inherent component of a panic attack or because they are instrumental acts (i.e., operants) that have been reinforced in similar situations in the past. (We later review evidence that PD may be correlated with a childhood in which sick role behaviors were modeled and reinforced in the presence of panic symptoms by parents in the home; Ehlers, 1993). Once the thought occurs during a panic attack, it may become a verbal CS associated with the rest of the panic attack. Through a process of verbal conditioning, thought CSs could become capable of eliciting panic themselves; when the thoughts now occur, they might help cause panic, just as other CSs do. It is often productive to think of cognitions as responses or stimuli within the terms of learning theory (e.g., Baldwin & Baldwin, 1998). Thus, there may be nothing particularly unique about their role in PD. As noted earlier, a variety of critiques of conditioning theories have been made in the past. We believe that many stem from the fact that modern issues and findings in conditioning and learning have not been communicated adequately to clinical scientists. One critique is that the conditioning approach lacks conceptual clarity when applied to PD (e.g., McNally, 1990, 1994, 1999; Reiss, 1987). For example, McNally argued that confusion exists about what constitutes the CS, US, CR, and UR, and that defining two arbitrary points on a continuum of arousal “blurs” the distinction between the CS and CR. He also argued that the distinction between the US and the UR is problematic. “Because certain bodily sensations are designated as CSs, by definition,

A Modern Learning Theory Perspective on the Etiology 277

they must have been established as such through association with a US. What, then, is this US and what UR does it elicit?” (1994, p. 107). He concluded, “The interoceptive conditioning hypothesis . . . constitutes little more than a misleading metaphor for the mechanisms underlying panic” (1994, p. 108). In the next sections, we show that modern conditioning research has done much to blur the old distinctions between these different terms. Thus, we argue that this critique of conditioning theory is neither compelling nor valid. A second criticism of conditioning theories is that they seem to “overpredict” panic (e.g., D. M. Clark, 1988). For example, while referring to early simplistic statements (e.g., Wolpe & Rowan, 1988), Clark argued that a conditioning theory should predict that every time a person experienced certain bodily sensations (CSs) that had been associated with panic, he or she should have a panic attack, and yet this is clearly not the case. We show that this argument loses force when one recognizes that the effects of CSs are routinely modulated by other stimuli in the background. A third criticism is that interoceptively conditioned responses should extinguish when the internal CS (such as a pounding heart) is not followed by panic, as might be the case when the patient runs up a flight of stairs (e.g., van den Hout, 1988). As we discuss in detail later, however, modern research on extinction, and particularly on what may be the inherent context specificity of extinction (Bouton, 1991a, 1993), obviates this criticism. Emotional responses extinguished in the context of athletic exertion would not be eliminated in other contexts. In summary, learning theory sheds important light on the development of panic disorder, and earlier critiques of early conditioning theories of PD may be met by an analysis of the current research literature. We believe that the complex state of affairs surrounding the origins of PD is consistent with modern conceptualizations of classical conditioning. Several points seem especially salient as we consider the role of conditioning in the behavioral events surrounding the development of PD. Conditioning Allows the CS to Trigger Constellations of Behavior and Emotion That Are Not Necessarily the Same as the UR

In textbook descriptions of Pavlov’s original experiment, there is a focus on a single response, such as salivation. In fact, Pavlov’s results were undoubtedly more interesting and complex than this. If one could stand back and look at what Pavlov’s dogs were actually doing, one would find the bell eliciting not only drooling but other digestive reflexes (including gastric acid, pancreatic enzymes, and insulin secretion) as well as behavioral responses that are designed to help the animal to get ready for food (Powley, 1977; Woods & Strubbe, 1994). A similar complexity is true of defensive conditioning in animals. Thus, the CR has many components, including freezing and other natural defensive behaviors, changes in respiration, blood pressure, and heart rate, and the release of endogenous opiates that reduce pain (Bolles & Fanselow, 1980; Davis, 1992; Hollis, 1982). The general role of this constellation of behaviors is to prepare the organism to deal with an upcoming dangerous event (e.g., Hollis, 1982, 1997). So it may be with the conditioning involved with PD; conditioned anxiety is a complex set of responses, including ones

278 Mark E. Bouton, et al.

corresponding to vigilance, that essentially help the person prepare for another panic attack. Viewed this way, there is no reason to expect that the CR and the UR will necessarily be identical. One example of the difference between CR and UR is a common result seen in defensive conditioning in rats. Freezing is a common measure of conditioned anxiety (often referred to as conditioned “fear” in this literature, although here we maintain Barlow’s distinction between anxiety and fear [panic]). Notably, freezing does not appear to be a component of the rat’s unconditioned reaction to footshock. If the animal is shocked and then moved immediately to a different location, freezing is not observed; rather, freezing occurs only if the animal is returned to the place where the shock was encountered (Blanchard & Blanchard, 1969; Fanselow, 1980). It is a CR that may involve heightened vigilance (e.g., “risk assessment,” D. C. Blanchard, R. I. Blanchard, & Rodgers, 1991), but it undoubtedly also functions to decrease attack or detection from predators (Hirsch & Bolles, 1980). The unconditioned reaction to shock is a very different burst of activity (Fanselow, 1994). Thus, even in basic aversive learning situations, there is often no identity relation between the UR and the CR. Another classic example of different CRs and URs is some well-known work on drug conditioning that began in the 1970s (e.g., for reviews see Cunningham, 1998; Siegel, 1989). In this literature, there is a difference between the response one observes to the drug US and the response one observes to the CS that signals it. Specifically, the CR often appears to be “opposite” to the UR; it compensates for the upcoming drug effect, another example of a get-ready response. One widely studied example is conditioning with morphine injection as a US. Although the unconditioned effect of morphine itself is to reduce pain, the CR to cues associated with it may be hyperalgesia, an increase in sensitivity to pain (e.g., Siegel, 1975). This and other related drug-conditioning phenomena (e.g., Siegel, 1989) have helped elucidate the complex relationship between the CR and the UR, a complete review of which is outside the scope of this article (for further discussion, see Eikelboom & Stewart, 1982; Hollis, 1982; Ramsay & Woods, 1997). The CR Is Often Determined by the Nature of the CS

In the 1970s, conditioning researchers also discovered that different CSs associated with the same US (and thus the same UR) can come to evoke very different CRs (e.g., Holland, 1977; Timberlake & Grant, 1975). For example, when presentation of a rat is used as a CS to signal food, the subject rat responds not merely with drooling but with social contact behaviors directed at the CS (Timberlake & Grant, 1975). Conditioning is now thought to engage whole behavior systems, or sets of behaviors that are functionally organized to deal with different USs (e.g., Domjan, 1994; Fanselow, 1994; Timberlake, 1994; Timberlake & Silva, 1995). Once a behavior system is engaged, different CSs provide “support” (Tolman, 1932) for particular behaviors in the same sense that a hallway makes it possible to walk and a swimming pool makes it possible to swim. A particularly interesting and potentially relevant example is the sexual conditioning system in Japanese quail (e.g., Domjan, 1994, 1997). In

A Modern Learning Theory Perspective on the Etiology 279

this research, a CS is presented to a male quail before a female US is presented and copulation ensues. The type of CR evoked by the CS depends on the CS’s duration and qualitative nature. For example, when a 30-s presentation of a foam block with feathers signals the female, the male begins to approach the CS; in contrast, when the foam block CS is presented for 20 min, the male instead paces back and forth when it is presented (Akins, Domjan, & Gutierrez, 1994). However, a CS can also come to elicit copulation as the CR (i.e., same as the UR), but only if it is a taxidermically prepared model that contains plumage and other features of the female quail (e.g., Domjan, Huber-McDonald, & Holloway, 1992). Such research suggests that different CSs can support quite different CRs and that only certain cues may actually support a CR that resembles the UR. As mentioned earlier, studies of defensive conditioning in rats with footshock USs have tended to uncover CRs that mostly consist of freezing (e.g., Fanselow, 1989; Fanselow & Lester, 1988); here the freezing CR is very different from the activity-burst UR. Nevertheless, because the form of the CR in other systems depends on both the qualitative nature of the CS as well as its timing (see also Silva, Timberlake, & Koehler, 1996; Timberlake, Wahl, & King, 1982), we should remain open to the possibility that different types of CRs will develop with different types of CSs predicting the US at different delays. In a classic study of defensive conditioning in rabbits, VanDercar and Schneiderman (1967) signaled a shock US delivered near the eye with tone CSs of various durations. In this method, the shock US elicits a closure of the rabbit’s nictitating membrane (which protects the eye) and also an increase in heart rate. Short CSs (e.g., less than 1 s) that ended with the US elicited the nictitating membrane response (i.e., a CR that resembled the UR) and caused a modest decrease in heart rate. In contrast, longer CSs (e.g., 6.75 s) supported no nictitating membrane response at all but elicited a strong decrease in heart rate. Thus, in some examples of aversive conditioning, qualitatively different CRs can emerge with cues associated with the US at different time intervals. One implication for PD is that the nature of the CR elicited by a CS (i.e., anxiety or panic) may depend on the CS’s temporal proximity to the US. The form of the CR in human fear conditioning also depends on the qualitative nature of the CS. For example, if interoceptive cues are “prepared” or “fear relevant” in the same way that evolutionarily based cues for phobias are (cf. Öhman & Mineka, 2001), then there is some evidence in humans that the CR may resemble the UR more so than is the case with unprepared or fear-irrelevant CSs. Cook, Hodes, and Lang (1986) showed that the CR using prepared CSs (slides of snakes) was heart rate acceleration rather than deceleration, as is more typically seen in human autonomic conditioning with fear-irrelevant or unprepared CSs (i.e., the CR is isodirectional to the UR in response to a US of shock). Dimberg (1987) also showed that with fear-relevant angry faces paired with shock, the CR was an accelerated heart rate response and an increase in activity in the corrugator muscle, which controls the eyebrow when frowning; neither of these responses occurred in individuals who received conditioning with fear-irrelevant or happy faces. Finally, Forsyth and Eifert (1998; see also Forsyth, Eifert, & Thompson, 1996) reported related results. One of their fear-relevant stimuli to which heart rate acceleration

280 Mark E. Bouton, et al.

(rather than deceleration) was conditioned was a video clip of a heart beating arrhythmically, an exteroceptive representation of an interoceptive stimulus. Given the variability in the form of the CR, it is easy to imagine that CSs associated with a panic attack will evoke responses that may often differ substantially from panic itself. However, it is also conceivable, given what we know about many conditioning systems, that certain CSs and/or certain CS-US intervals may condition a CR that resembles the UR, a conditioned panic attack. Anxiety and panic reactions are functionally different; a functional perspective on conditioning actually causes us to anticipate that different conditions could support conditioning of different responses. Panic is an immediate response to an insult to the organism; although it may involve cognitions about impending doom, it must have temporal and other properties that help the organism deal with a terrifying event that is already in progress. Anxiety, in contrast, is functionally organized to help the organism prepare for a possible upcoming insult. It is more “forward looking” in this sense (Barlow, 1988, 1991). Given their different functional roles, it is not surprising that panic and anxiety are different. Our perspective thus accepts a qualitative distinction between the two, whereas other approaches to PD either do not make one or certainly do not focus on its importance (cf. Beck & Emery, 1985; D. M. Clark, 1996; McNally, 1994). Interoceptive Conditioning

An important component of the conditioning approach to PD is the idea that interoceptive conditioning is involved (Barlow, 1988; Goldstein & Chambless, 1978). The occurrence of a panic attack is itself a conditioning trial that allows the patient to associate internal bodily sensations that accompany the early onset of the attack with the rest of the attack; the result is that modest changes in heart rate and respiration become signals and can later elicit a full-blown attack. One relevant example is a classic Soviet dog experiment described by Razran (1961, p. 86). Distension of the intestine, an interoceptive CS, was paired with presentation of a mixture of 10% carbon dioxide administered directly to the trachea (the US). When these two events were paired, the intestinal distension (CS) quickly acquired the ability to elicit hypercapnic respiratory changes (CR) that were the same as the UR. The phenomenon might parallel what happens when and if internal sensations (such as intestinal contractions) are paired with hyperventilation, which often occurs during a panic attack; one might expect in the future that the intestinal contractions alone might become sufficient to trigger respiration changes and possibly panic. Interestingly, Razran (1961), in summarizing the Soviet literature, claimed that interoceptive conditioning is especially resistant to extinction, lending further plausibility to its likely role in the etiology of PD. Little interoceptive conditioning research seems to have followed the methods used in the Soviet heyday described in Razran’s (1961) classic monograph (see Dworkin, 1993). However, once again, the literature on classical conditioning with drug stimuli is relevant. Here, interoceptive events clearly do become associated. For example, Siegel (1988) showed that an injection of pentobarbital that signaled an injection of morphine caused a classical CR to

A Modern Learning Theory Perspective on the Etiology 281

pentobarbital: The rats were more tolerant to morphine when pentobarbital was present than when it was not. Evidently, they had learned to associate the morphine US with the pentobarbital CS, two interoceptive events. The reader might recognize this experiment as an example of “state-dependent learning,” the well-known phenomenon in which learning is best when it is tested in the presence of the drug state (or mood state, see Eich, 1995) in which it originally occurred. To the extent that such states are interoceptive, state-dependent learning is an effect indicating that interoceptive events can be associated with their consequences. The idea that internal cues can be associated with interoceptive aversive events is also consistent with some research in humans. Human participants were exposed to three pairings of mental images with a US of 5.5% carbon dioxide (CO2)–enriched air (Stegen, De Bruyne, Rasschaert, Van de Woestijne, & Van den Bergh, 1999). The images were either fear relevant or fear irrelevant (being stuck in an elevator or sauna vs. reading a book or overlooking the sea). Participants exposed to the fear-relevant images (but not those exposed to the fear-irrelevant images) showed significant conditioning of both subjective symptoms of anxiety as well as altered respiratory behavior and cardiac/ warmth symptoms similar to those produced by the US. That fear-relevant imagery can elicit such interoceptive and subjective CRs after only three conditioning trials has implications for our argument regarding the role of interoceptive conditioning in PD. The authors noted that “a little stress-induced hyperventilation in association with a phobia-relevant place or an image thereof may evoke the experience of similar symptoms and anxiety on next confrontations, even without apparent hyperventilation” (p. 150). Moreover, as noted earlier, the nature of the CR in animal conditioning can depend on the qualitative nature of the CS and the interstimulus interval (ISI). Thus, it is possible that a true panic CR might become conditioned with the right CS and the right ISI. Definitive tests of this idea must await future research. A particularly interesting type of interoceptive conditioning is the case in which a low dose of a drug signals a higher dose of the same drug. Greeley, Lê, Poulos, and Cappell (1984) injected a small dose of ethanol in rats before administering a larger dose. Tolerance to the larger dose subsequently depended on the small dose preceding it, but only if the two injections had been paired. Pairings also enabled the small injection to initiate a stronger compensatory response. Thus, a small dose of ethanol acquired the ability to signal more of itself. Comparable results have been reported for morphine-morphine pairings (Cepeda-Benito & Short, 1997). Thus, animals can associate a strong interoceptive event with a weaker version of the same event. More recently, Kim, Siegel, and Patenall (1999) found evidence of a similar kind of learning occurring within single administrations of morphine. They found that rats given long exposures to morphine show a conditioned compensatory response to a short “probe” injection that deliberately mimicked the early-onset properties of the longer injection. Apparently, interoceptive cues corresponding to the onset of a long interoceptive stimulus can indeed become associated with the remainder of the stimulus. It is not difficult to imagine that early-onset properties of a panic attack can, therefore, come to signal the rest of the event as it continues to unfold in time.

282 Mark E. Bouton, et al.

Conditioning in which the onset of an event signals the rest of the event clearly blurs the traditional distinction between CS and US. Nonetheless, we believe this form of conditioning is probably very common. In an extended discussion of the relationship between learning and regulatory physiology, Dworkin (1993) reached a similar conclusion. Dworkin gave the type of conditioning in which early onset is associated with an event’s later aspects a name, the homoreflex, and contrasted it with the better known arrangement in which CS and US are different, the heteroreflex. One example of a homoreflex is a baroreceptor that detects a small increase in blood pressure coming to predict a further increase in blood pressure. Learning this sort of relationship presumably allows other parts of the system to respond and adapt to the blood pressure change more quickly. Because the “CS” and “US” are so similar, there are grounds for expecting this kind of learning to be robust: Similarity between signals and the things they signal can allow especially strong conditioning (e.g., Rescorla & Furrow, 1977; Rescorla & Gillan, 1980; Testa, 1975). Dworkin (1993) noted that if this is correct, then “homoreflexes should turn out to be even more common than heteroreflexes” (Dworkin, 1993, p. 79). Interestingly, there is a sense in which the earliest conditioning experiments in Pavlov’s own laboratory probably also involved the dog associating early and late aspects of a single event. Pavlov’s students first noted that, after introducing sand into the dog’s mouth, which caused salivation, the mere sight of sand itself soon came to cause anticipatory salivation (Domjan, 1998). The dog was associating the sight of sand with its later properties. This type of learning is undoubtedly common as organisms learn about events and objects in their world. Therefore, to understand this learning process better, Pavlov devised a method in which the signal and the signaled event were separated so that they could be manipulated independently. That is, the bell-food experiment we now know so well can be seen as a technique used to break down and analyze learning about dynamic individual events that naturally flow and change over time. We suggest that this type of conditioning is fundamental to understanding PD. That is, the patient with PD has learned to associate weak and early panic symptoms with the remainder of the full-blown panic episode. Mere exposure to a US may inevitably allow cues that correspond to its early onset to become associated with its later aspects, allowing conditioning of anxiety and possibly panic itself. A panic attack can be seen as an opportunity to associate early “warning signs” with the full-blown emotional reaction. Furthermore, because early-onset cues are presumably similar to the US’s later effects, they may be especially easy to condition (see prior discussion); theoretically, they might, therefore, overshadow learning about other perfectly valid predictors of the US, such as other predictive external cues. This scenario suggests that interoceptive conditioning may be a major contributor to the development of PD. Of course, not all of the body’s reactions during early panic onset are necessarily interoceptive. Shortness of breath, for example, may have exteroceptive as well as interoceptive aspects. It is easy to find evidence of related US-US conditioning with USs that have exteroceptive effects. Goddard (e.g., 1996, 1997; Goddard & Jenkins, 1988) reported a number of experiments in which

A Modern Learning Theory Perspective on the Etiology 283

food USs signaled the occurrence of additional USs (see Goddard, 1999, for a review and implications). When food USs are delivered close together in time, the first pellet comes to signal others to follow. Similar learning occurs with aversive stimuli. For example, a weak shock can become more aversive through pairings with a stronger shock (Crowell, 1974). In addition, exposure to long shocks can increase the aversiveness of shorter shocks, apparently because shock onset becomes associated with the later aspects of the long event (Anderson, Crowell, DePaul, & McEachin, 1997). Interestingly, US-US conditioning can be extinguished by presenting the predictive stimulus (e.g., shock-onset cues or the first pellet) on many trials alone (e.g., Anderson et al., 1997; Goddard, 1997). Clearly, with both interoceptive or exteroceptive stimuli, nominal USs can signal other USs; one panic attack, or early aspects of a panic attack, can signal other panic attacks. It is not difficult to imagine this sort of learning contributing to the development and maintenance of PD. CSs Do Not Just Trigger Responses; They Also “Modulate” Other Responses Controlled by Other Events

A case can thus be made for a role for interoceptive conditioning in PD as well as for the idea that the behaviors elicited by a CS will be multiple, complex, and not necessarily the same as the panic reaction itself. However, CSs have additional important means of influencing emotions and behavior; they also modulate other types of behavior. Historically, theorists (e.g., Mowrer, 1947, 1960) emphasized that aversive CSs that elicit anxiety can also modulate or influence the strength of ongoing operant or instrumental behaviors. According to two-process theory (e.g., Rescorla & Solomon, 1967), anxiety CSs should exaggerate avoidance behavior and weaken appetitively motivated behavior that occurs in their presence (see also Overmier & Lawry, 1979; Trapold & Overmier, 1972; see also Colwill, 1994). By this mechanism, conditioned anxiety might increase or potentiate instrumental acts that have been learned through negative reinforcement (e.g., escape or avoidance responses) to potentially keep the organism out of trouble. The tendency to carry “talismans” such as pill bottles, even if they are empty, and other safety signals or “safety behaviors” (Salkovskis, Clark, & Gelder, 1996) are presumably examples of avoidance or escape behaviors that have been reinforced by anxiety reduction or safety. The idea that anxiety CSs can modulate these behaviors means that heightened anxiety may exaggerate them, even if they appear “irrational” (i.e., an empty pill bottle has no objective effect on anxiety). As with all avoidance responses, they may serve to prevent extinction of the CR (which would include preventing disconfirmation of catastrophic thoughts). Interestingly, modulation effects of CSs on instrumental behaviors can sometimes be specific. That is, CSs sometimes seem to arouse expectancies of the specific US with which they are associated and, therefore, mainly modulate instrumental behaviors that are connected with the same US (e.g., Colwill & Motzkin, 1994; Colwill & Rescorla, 1988; Delamater, 1996; Kruse, Overmier, Konz, & Rokke, 1983). In principle, then, a CS may arouse anxiety about a particular US (e.g., fainting and falling to the floor), and thereby selectively increase seemingly bizarre or idiosyncratic behaviors that have been

284 Mark E. Bouton, et al.

reinforced for avoiding that particular US (e.g., clutching door handles, walls). Other behaviors connected with other USs (e.g., taking a spouse or trusted companion along for fear of panicking while driving) might not be affected. Another important modulating effect of CSs is to exaggerate or potentiate the strength of URs or CRs that are elicited in their presence by other stimuli. This point has been emphasized by Wagner, Brandon, and their associates (e.g., Wagner & Brandon, 1989). Following Konorski (1967), they emphasized the fact that CSs enter into associations with both emotive and sensory aspects of the USs with which they are paired. The emotive CR (generated by the emotive association) functions to augment other responses, such as an eyeblink CR elicited by a second CS (Bombace, Brandon, & Wagner, 1991; Brandon, Betts, & Wagner, 1994; Brandon & Wagner, 1991) or an eyeblink or startle reflex evoked by other stimuli (Brandon, Bombace, Falls, & Wagner, 1991; McNish, Betts, Brandon, & Wagner, 1997). The latter effect is consistent with the well-known fact that anxiety CSs also potentiate startle responses elicited by sudden bursts of noise (e.g., J. S. Brown, Kalish, & Farber, 1951; see Davis, 1992, for a review). In the presence of a CS controlling anxiety, we may become more reactive to sudden stimuli that evoke startle responses. These modulating effects of emotions and emotional CSs have been emphasized in discussions of human emotion (e.g., Lang, 1994, 1995; Lang, Bradley, & Cuthbert, 1990) and anxiety disorders (Cook, Hawk, Davis, & Stevenson, 1991; Grillon, Ameli, Goddard, Woods, & Davis, 1994; Grillon, Ameli, Woods, Merikangas, & Davis, 1991; Morgan, Grillon, Southwick, Davis, & Charney, 1995). They may also be especially important to an understanding of PD. Through conditioning, anxiety cues may augment and exacerbate minor panic reactions that are triggered by other stimuli. This may be a key difference between patients who suffer from PD and other individuals who may occasionally experience nonclinical panic episodes without anxiety. Conditioned anxiety may lower the threshold of (or exaggerate) subsequent panic reactions. The modulating effects of anxiety CSs could thus cause, at a level below conscious awareness, the kind of exaggerated panic that cognitive theorists would assume requires catastrophic misinterpretations (e.g., D. M. Clark, 1986, 1988). A “floater,” or floating object observed in the visual field, may mean nothing to a normal individual. However, to a person sensitized by anxiety, it could elicit another panic attack. In other words, classical conditioning may conceivably allow CSs to potentiate panic either through the modulating effect described here or through a more direct, traditional, eliciting effect described earlier in the section on interoceptive conditioning. Thus, by virtue of the modulating function, we can accept and retain the distinction between panic and anxiety but still find the state of anxiety exacerbating panic attacks. The Effects of CSs Are Also Further Modulated by Other Stimuli

We have just emphasized how CRs and URs can be modulated by the presence of another stimulus; the CR to one CS can be potentiated by the emotional effect controlled by another. The general idea that the CR evoked by a CS is quite commonly modulated by the effects of other stimuli has received a great deal of attention in the conditioning literature (e.g., Swartzentruber, 1995). It

A Modern Learning Theory Perspective on the Etiology 285

introduces another layer to a conditioning theory of PD. Because responding is always viewed as the product of a CS as well as other modulating cues in the background, the CR cannot be assumed to be evoked automatically by a CS. This explains why a conditioning theory of PD does not “overpredict” panic in the presence of interoceptive (or exteroceptive) cues associated with panic attack. The conditioning experience creates a potential for a CS to evoke a CR, but the actual strength of the response always further depends on other stimuli in the situation. Learning theory now recognizes several types of CR modulation. One is the potentiation-style effect described previously, in which the emotional properties of a CS can energize CRs, URs, or instrumental actions that are controlled by other stimuli. An even simpler modulation effect is summation: The strength of the CR observed in any situation is determined by the summed value of all the CSs that are currently present. The idea is captured by the Rescorla-Wagner model (Rescorla & Wagner, 1972), which gives excitatory CSs (those that predict USs) positive values and inhibitory CSs (those that, in general, predict no US) negative values. When put together, the strength of the response is determined by the summed value of all the CSs that are present. Thus, when two excitatory CSs are put together, the response is larger than when either is presented alone (e.g., Hendersen, 1975; Reberg, 1972; Van Houten, O’Leary, & Weiss, 1970; S. J. Weiss & Emurian, 1970). Importantly, when an excitatory CS and an inhibitory CS (e.g., a safety signal) are put together, there will be less responding than to the excitor alone (e.g., Rescorla & Holland, 1977).2 Another implication of the Rescorla-Wagner model’s summation mechanism is worth mentioning. According to the model, learning is an adjustment that occurs when there is a discrepancy between the outcome predicted on a conditioning trial (e.g., US or no US) and the actual outcome that occurs. The outcome predicted is determined by the sum of all stimuli present on the trial, as just discussed. This simple idea has several straightforward but remarkable implications. For example, when there is no US, and extinction would ordinarily occur, the presence of an inhibitor along with a to-beextinguished excitor will subtract from and may cancel the exciter’s prediction of a US, eliminating the discrepancy between prediction and actual outcome and thus leaving the excitor unextinguished. The inhibitor is said to “protect” the excitor from extinction (Chorazyna, 1962; Soltysik, Wolfe, Nicholas, Wilson, & Garcia-Sanchez, 1983). To the extent that agoraphobic “safety” behaviors such as carrying pill bottles also operate by canceling an excitatory CS’s prediction of danger, combining the pill bottle with a trip to the shopping mall (an anxiety-evoking CS) may similarly protect the shopping mall CS from extinction (e.g., Soltysik, 1960). The converse of protection from extinction is also possible: During extinction, the addition of extra excitatory CSs that predict the US (instead of predicting no US, as an inhibitor does) would sum to yield an overprediction of the US, increasing the discrepancy between prediction and actual outcome and thus facilitating any associative loss to all the CSs resulting from extinction (Rescorla, 2000; Wagner, 1969). Therefore, extinguishing multiple excitatory CSs together should be more effective than extinguishing any one in isolation. The implication is that, to facilitate extinction of the shopping mall, it would be best to combine that CS with other CSs

286 Mark E. Bouton, et al.

(e.g., elevators, a fast heart rate, shortness of breath) during exposure. Clinically, this is often used as an approach during cognitive-behavioral therapy for agoraphobia, although a different rationale is usually given (e.g., Barlow & Craske, 2000). A complete explication of the Rescorla-Wagner model is beyond the scope of this article. However, our discussion illustrates the extent to which all cues present at any time are important in determining anxiety and learning. A related, and equally important, modulating effect is provided by contextual stimuli that are present but even further in the background when the CS is presented. In the conditioning laboratory, contextual cues are often operationally defined as the room or apparatus in which CSs and USs are presented; however, they can include a variety of other cues, including drug states, mood states, cues correlated with the passage of time, and the memory of recent events (e.g., Bouton & Nelson, 1998b; Bouton & Swartzentruber, 1991). Although such cues are indeed in the background, they can have potent effects on the CR that are relevant to any disorder in which classical conditioning plays a role (e.g., see Bouton & Nelson, 1998b). Bouton and colleagues showed that contextual stimuli are especially important in determining performance after extinction (e.g., Bouton, 1991a) or other situations in which learning in a second phase replaces or interferes with something learned first (Bouton, 1993). For example, anxiety evoked by a CS that has been through extinction is “renewed” if the context is changed after extinction (e.g., Bouton & King, 1983; Bouton & Ricker, 1994). This effect and others related to it have important implications for understanding relapse or return of fear (e.g., Bouton, 1988, 1991b; Bouton & Swartzentruber, 1991; see Mineka, Mystkowski, Hladek, & Rodriguez, 1999, for a preliminary example in humans). Somewhat surprisingly, in contrast to extinction, anxiety itself often generalizes almost perfectly between contexts: If anxiety is conditioned to a CS (by pairing it with shock) in one context and then the CS is tested in a second context, conditioned anxiety there is just as strong (e.g., Bouton & King, 1983; Hall & Honey, 1989; Lovi-bond, Preston, & Mackintosh, 1984). Thus, if a panic attack on a crowded bus conditioned anxiety to the feeling of dizziness, dizziness should elicit anxiety quite well in other contexts, such as the home, an airplane, or a ski lift. However, because extinction does not generalize as well between contexts, if anxiety were extinguished to dizziness in the context of the home, then dizziness would stop eliciting anxiety at home but could still evoke anxiety in other contexts (on the ski lift, the airplane, or the bus). The generalization of anxiety and the lack of generalization of extinction would contribute to the persistence of the disorder. Interestingly, the psychological mechanism through which contexts control performance is different from the other modulating mechanisms described earlier. Instead of working through its direct association with the US, a context often modulates performance to the CS by signaling or retrieving the CS’s own current connection with the US. It is as though the context disambiguates the CS (i.e., gives it its current meaning) in a manner analogous to the way in which contexts determine the meanings of words (e.g., Bouton, 1988, 1994b). This view has a number of implications for the persistence and treatment of anxiety disorders (e.g., Bouton, 1991b; Bouton & Nelson, 1998b; Bouton & Swartzentruber, 1991). For present purposes, what we view as the inherent

A Modern Learning Theory Perspective on the Etiology 287

context specificity of extinction performance is another reason why a conditioning model does not overpredict the occurrence of panic attacks: Anxiety elicited by an interoceptive CS may extinguish in some contexts without eliminating its impact in others. The disambiguating effect of context is theoretically linked to a final form of modulation known as occasion setting (e.g., Holland, 1992; Schmajuk & Holland, 1998; Swartzentruber, 1995). Occasion setters are discrete stimuli (such as tones or lights) that can turn on or turn off responding to other CSs through some mechanism besides their direct association with the US and thus without necessarily evoking behavior on their own. Although their mechanism of action is currently a matter of debate, they are clearly connected with the theoretical issues concerning contextual control (e.g., Bouton & Swartzentruber, 1986; see also Bouton & Nelson, 1998a). Because neither contexts nor occasion setters operate through their direct associations with the US, simple extinction of an occasion setter (or a context) that “turns on” responding to a CS may not influence its ability to modulate responding to the CS (Bouton & Swartzentruber, 1986; Holland, 1989; Rescorla, 1985). Thus, we may find that the reaction of a patient with PD to his or her own racing heart (a CS) is especially problematic during visits to a shopping mall (a possible context or occasion setter). Exposure to the shopping mall alone may not weaken its ability to modulate the response to the heart racing. What is needed is extinction of the CS and the context together (e.g., Rescorla, 1986). This final example should make clear why so-called modulatory mechanisms are relevant to our understanding of PD. This may also be why many cognitive-behavioral therapists find it useful to encourage their patients to bring on frightening bodily sensations (such as by hyperventilating) while engaging in exteroceptive exposure to their agoraphobic situations (Barlow & Craske, 2000). Some Pertinent Clinical Evidence

Evidence That Anxiety Potentiates Panic. From the perspective we are taking here, anxiety often potentiates panic rather than panic attacks truly coming out of the blue, as was originally thought to occur in most cases (Klein, 1981; Sheehan, 1983). As Basoglu, Marks, and Sengiin (1992) argued, the problem with most prior studies suggesting that panic attacks come out of the blue is that they were based on retrospective recall. They noted that patients may not recall episodes of anxiety that preceded their panic episodes. Thus, Basoglu et al. (1992) conducted a prospective study of panic and anxiety in 39 patients who had PD with agoraphobia. Each patient recorded in a diary using an event-sampling technique the duration and intensity of episodes of panic and anxiety over three 24-hr periods, including their anxiety levels before panic episodes. Episodes of anxiety and panic were also classified as “situational/ expected/predictable” versus “spontaneous/unexpected/unpredictable” based on whether the episode was expected and/or clearly linked to a situation that usually triggered such episodes (p. 58). Several results are of primary interest for our purposes. First, of 117 panic episodes recorded, 80% were situational/ expected/predictable, and only 20% were spontaneous/unexpected/unpredictable. Moreover, of the 32 patients who reported at least one panic, 69%

288 Mark E. Bouton, et al.

reported that their panics surged from an already heightened plateau of anxiety. Only 13% of the 32 who panicked had not reported a preceding period of anxiety. Another 18% reported anxiety preceding some panics but not others. In other words, more than 66% reported that anxiety had always preceded their panics on these recorded days, and 87% reported that this happened some or all of the time. Moreover, longer and more intense prepanic baseline anxiety was correlated with more intense panic (including number of episode symptoms). These results are consistent with our view that anxiety may often precede and potentiate the intensity of a panic attack. Two other somewhat different prospective studies using computer-assisted self-monitoring techniques both reported related findings. Kenardy et al. (1992) followed 20 female panic patients for a 1-week period in a study of the psychological precursors of panic attacks (not differentiating between predicted and unpredicted attacks). Patients answered a variety of questions every hour and whenever they believed they were having a panic attack. Anxiety in the hours preceding panic attacks (based on the question “How anxious do you feel?”) was higher than in control hours (preceding no panic), although not significantly so. However, the participants’ estimates of “the likelihood of panic this hour” (a more cognitive but conceptually related concept to anxiety) were significantly elevated in the hour preceding panic attacks relative to control hours. The authors deemed these elevated estimates of the likelihood of panic to support Barlow’s (1988) position that “an underlying apprehension is a precursor to panic” and that such expectancies “can be thought of as developing from a conditioned fear of panic attacks” (Kenardy et al., 1992, p. 672). Another related study (Kenardy & Taylor, 1999) had a somewhat different goal of determining the accuracy of predictions with predicted panic attacks (i.e., the sensitivity and specificity of the predictions) and whether psychological precursors (such as anxiety and perception of physical symptoms) were associated with panic attack prediction versus panic attack occurrence. In this study, 10 female panic patients monitored their panic attacks for 7 days, making predictions every hour about whether they would have a panic attack in that hour as well as their sense of threat or danger, their level of anxiety, number of physical symptoms of panic, and so on at the time of the ratings. Unpredicted panic attacks were only preceded by reports of elevated physical symptoms. Predictions that a panic attack would occur (whether or not it occurred) were associated with both physical symptoms and an elevated sense of threat or danger and anxiety. Thus, elevated physical symptoms and anxiety (only the former for unpredicted attacks) commonly preceded panic attacks, although they also occurred frequently at other times when no panic attacks occurred. (The only related finding reported in the Kenardy et al., 1992, study was that some panic symptoms—about one on average—frequently occurred in hours not associated with panic attacks.) Together, these results are consistent with our learning theory perspective that posits anxiety or physical symptoms as CSs—or potentiating factors—for panic. Other evidence that anxiety precipitates panic comes from laboratory research on panic provocation, in which it has long been recognized that the single best predictor of panic in response to a variety of panic provocation

A Modern Learning Theory Perspective on the Etiology 289

agents is the baseline level of anxiety (e.g., Barlow, 1988; Margraf, Ehlers, & Roth, 1986). For example, Liebowitz et al. (1984) reported that patients who experienced panic attacks during lactate infusions in the laboratory experienced heightened anxiety before the infusion compared with the patients who did not panic. In another study (Liebowitz et al., 1985), heart rate averaged 83.98 beats per minute (bpm) during baseline for patients who went on to panic as opposed to 75.30 bpm during baseline for patients who did not go on to panic. Heart rate in control participants was 62.79 bpm. In fact, baseline differences in levels of anxiety between patients and controls before panic provocation have been reported in almost all studies dating back to the 1940s (e.g., Cohen & White, 1947, 1950). Breggin (1964) suggested current baseline anxiety as one of four principal factors accounting for the production of panic attacks in panic provocation experiments, thus anticipating our contemporary learning theory account by more than 35 years. Interestingly, although anxiety seems to play a role, the specific somatic responses capable of provoking panic may differ markedly from patient to patient. This has never been better illustrated than in the data reported long ago by Lindemann and Finesinger (1938), in which individual patients responded with panic to infusions of either adrenaline or acetylcholine but not both. These substances produce very different, and almost opposing, sets of somatic sensations and have very different underlying neurobiological mechanisms of action. That individuals are specifically sensitive to dissimilar provocation procedures, a finding that characterizes the panic provocation literature (e.g., Barlow, 1988; van den Hout, 1988), suggests that individuals have learned a specific association between panic and certain discrete somatic sensations possibly associated with an earlier major alarm reaction. Evidence That PD Starts with a Panic Attack Conditioning Episode. Most patients report that PD usually starts with an initial panic attack followed by one or more attacks within a matter of weeks or months. It is generally over the course of these initial attacks that anxious apprehension about the possible occurrence of further attacks begins to develop. During this early stage, agoraphobic avoidance of situations that provide a possible setting for the attack may also begin to develop. For example, as noted earlier, Öst and Hugdahl (1983) gave detailed questionnaires to 80 people with PD and found that approximately 91% could recall the initial episode of panic (or vicarious experience of panic) after which the disorder began to develop; only 10% could not recall a conditioning event. Of these 91% with a conditioning mode of onset (as Öst & Hugdahl referred to it), more than 50% said that, after the first panic attack, the initial development of agoraphobia was rapid (i.e., fully developed agoraphobia within 2 weeks), and only 14% described a slow onset. Other investigators reported similar findings (Craske et al., 1990; Merckelbach, de Ruiter, Van den Hout, & Hoeckstra, 1989; Thyer & Himle, 1985; Uhde et al., 1985). Although Öst and Hugdahl (1983) acknowledged that the US (i.e., the specific trigger) for the first panic attack could be identified in only a minority of cases, they shared our opinion, as well as that of Forsyth and Eifert (1996), that identifying a specific US is not essential in determining that conditioning plays a crucial role here. They noted that the attack itself was generally terrifying,

290 Mark E. Bouton, et al.

with thoughts of losing control, fainting, or dying as core symptoms, and that this is sufficient to allow conditioning to occur. It is perhaps worth noting here that critics of this view, such as Menzies and Clarke (1995), have claimed that a US must be identified to ascribe a role to conditioning, but this argument is based on an unnecessarily restrictive view of conditioning. As argued elsewhere (e.g., Barlow, 1988; Carter & Barlow, 1995; see also Forsyth & Eifert, 1996), all that is needed for conditioning to occur is that either a true alarm or a false alarm reaction (activation of the flight-or-fight response) occurs in the presence of some potential interoceptive or exteroceptive CS. The percentage of panic patients who claim to recall an initial panic attack conditioning episode might actually be considered impressive in light of evidence suggesting that emotional conditioning can occur without declarative memory of the conditioning experience. We have already described evidence suggesting that patients without an intact hippocampus can show emotional conditioning even though they cannot report the conditioning contingency (Bechara et al., 1995). Research by Öhman and colleagues has shown that the exteroceptive conditioning sometimes involved in the origins of phobic fears may occur outside of conscious awareness if fear-relevant stimuli are involved (e.g., Öhman, 1996, 1997; Öhman, et al., 2000). In addition, it is worth noting that interoceptive conditioning has historically been thought to occur unconsciously and in the absence of an overt specifiable US because the US is internal (e.g., Razran, 1961). Together, these considerations suggest that accurate self-report data depending on conscious awareness may provide an inherently conservative estimate of the extent to which conditioning processes may be involved in the etiology of anxiety disorders. We should also note, however, that although the extant self-report data are consistent with a conditioning account of PD, there is a clear need for more objective data on the involvement of early conditioning trials. Factors That May Affect the Potency of Panic Attacks and Hence the Strength of Conditioning

Several factors may further affect the potency of panic attacks and thus the likelihood that they will lead to PD through classical conditioning. These factors, which we now review, include whether the attacks are perceived as controllable and predictable, whether they occur in the presence or absence of safety cues, and whether separate experience with panic attacks has increased their emotional impact through some sensitization process. Full-blown panic attacks are themselves experienced by people with PD as unpredictable and uncontrollable aversive (even terrifying) events. They are perceived as uncontrollable in the sense that there is nothing that can be done to abort an ongoing attack. However, they are also sometimes perceived as unpredictable in the sense that they seem to come from out of the blue, even if they may actually be triggered by unconscious interoceptive cues. First, consider the controllability dimension. The animal conditioning literature indicates that uncontrollable shock conditions anxiety to neutral CSs more powerfully than does the same amount of controllable shock (e.g., Mineka, Cook, & Miller, 1984; Mowrer & Viek, 1948). Indeed, Mineka et al. (1984)

A Modern Learning Theory Perspective on the Etiology 291

found that levels of anxiety conditioned with inescapable shock were twice as high as levels of anxiety conditioned with the exact same amount of escapable shock. In fact, there is evidence that initial panic attacks occurring in difficult-to-escape situations (e.g., driving on a highway, flying, being in a formal meeting) condition more anxiety than panics that occur in more escapable locations, such as being home alone or with a significant other (Barlow, 1988; Craske et al., 1990). Moreover, prior experiences with uncontrollable aversive events can later potentiate the conditioning of defensive reactions. Given this, and that initial panic attacks often occur during periods of uncontrollable stressors, we would expect them to be especially powerful sources of conditioning (e.g., see Maier, 1990).3 Sanderson, Rapee, and Barlow (1989) demonstrated that a sense of control may be important in influencing the occurrence of a panic attack itself. Using a 5% CO2 panic provocation procedure, Sanderson et al. divided patients with PD into two groups. Both groups were told to inhale the CO2; they were also told that if the sensations created by it were sufficiently bothersome, if and when a red light came on, they could turn a dial to reduce the rate of infusion of the CO2 if they felt they had to. The two groups differed only in whether the red light actually came on during the CO2 inhalation, affording one group the perception of control. Although no one in the perceived control group attempted to turn the dial, the dial was actually inoperative (it did not affect levels of CO2). Thus, the two groups experienced equal amounts of CO2. Nevertheless, there were substantial group differences in physiological reactivity to the CO2; the perceived control group showed lower levels of physiological responding to the CO2 than the no perceived control group. Moreover, 8 of 10 patients in the no perceived control group reported experiencing a panic attack compared with only 2 of 10 in the perceived control group. Turning to unpredictability, there is evidence in the animal literature that unpredictable aversive events are perceived as more stressful than predictable aversive events (and that when given a choice, humans and animals generally prefer predictable to unpredictable aversive events; cf. Mineka & Hendersen, 1985). There is also evidence that unpredicted panics lead to more anxiety in a clinical population. On the basis of hypotheses developed from the animal conditioning literature, Craske, Glover, and Decola (1995) hypothesized that patients with PD would experience greater levels of anxiety on days that happened to follow an unpredicted panic than on days after a predicted panic. Although there were no differences between the group who only experienced predicted panic attacks and the group who only experienced unpredicted attacks, among the group who experienced a combination of the two this hypothesis was upheld. That is, patients with PD who experienced a mixture of the two kinds of attacks showed higher levels of anxiety and worry on days after an unpredicted attack than on days after a predicted attack. The authors also noted that these findings might be related to findings from the animal literature on experimental neurosis suggesting that having a history of predictability may exacerbate the effects of lack of predictability (Mineka & Kihlstrom, 1978; see also Mineka & Zinbarg, 1996). This is because the prediction was only upheld among those who had some history of predicted attacks intermixed with unpredicted attacks.

292 Mark E. Bouton, et al.

Another factor affecting the potency of a panic attack may be the anxiety-inhibiting presence of having a safe person present. Clinical observations have long suggested that patients with PD show a far greater range of activity and are less prone to panic when a safe person (often a spouse but sometimes a trusted companion of another sort) is with them than when they are alone. This phenomenon has been studied and documented in a laboratory study. Patients with PD showed fewer panic symptoms (distress symptoms, catastrophic cognitions, and physiological arousal) in response to CO2 if a safe person was with them than in the absence of a safe person (Carter, Hollon, Carson, & Shelton, 1995). The potency of panic attacks might also change as a function of previous experience with panic. Such experience might increase the intensity of later panics by engaging sensitization processes, including the neurobiological process recently described by Rosen and Schulkin (1998). Rosen and Schulkin argued that animal laboratory experiments on sensitization (in which exposure to uncontrollable stressors increases the organism’s reactions to later stressors) serve as a model for similar observations in humans, showing that either distal or proximal exposure to uncontrollable stressors sensitizes them to the effects of subsequent stress, possibly including panic itself. They emphasized nonassociative (unlearned) neurobiological processes that might occur in the amygdala as underlying these effects. We would note that associative processes might also be involved. For example, as noted earlier, through repeated exposure to the stressor, the organism might learn to associate onset of the event with the rest of the event (e.g., Anderson et al., 1997). Event onset (including panic attack onset) could thus acquire a stronger and stronger emotional impact. Thus, either nonassociative or associative mechanisms could allow early panic attacks or other stressors to make later attacks more intense. Thus, in turn, sensitization to panic attacks could increase the strength of conditioning that results when a panic attack is subsequently associated with new CSs. Sensitization of panic attacks might also influence the strength of anxiety reactions that have been conditioned previously. For example, inflation effects, first discovered by Rescorla (1974) and later replicated (at least through clinical case studies) in humans by Davey, de Jong, and Tallis (1993), are said to occur when a mild fear that is first conditioned to some cue using a mild unconditioned stressor later grows in magnitude when the animal or person is exposed noncontingently to a more powerful unconditioned stressor (not paired with the CS). Thus, a rat first experiencing a tone paired with a mild shock shows a weak conditioned fear response, but if the rat is later exposed to a noncontingent strong shock on its own, the rat’s level of fear of the tone is increased. By analogy, if a person experienced several mild panic attacks that were preceded by an upset or growling stomach, some conditioned excitatory strength might accrue to the stomach cues. However, this excitatory strength might later be inflated through the experience of other more severe panic attacks that were not accompanied by the same stomach cues. Bouton (1984) showed that such inflation effects are not context specific; that is, the inflation events need not occur in the same context in which the original conditioning took place, or where fear is tested, for inflation to occur. In addition, Hendersen (1985)

A Modern Learning Theory Perspective on the Etiology 293

also showed that inflation effects are often larger when the strong stressors are experienced a long time after the initial conditioning experience. Summary

PD begins in individuals with certain psychological and biological vulnerabilities (to be discussed next) when early panic attacks occur and condition anxiety, a functional, forward-looking constellation of responses that prepares the person for the next panic attack. Several psychological factors that influence the perceived intensity of panic (e.g., its perceived controllability or predictability) may influence the potency of the US and thus the degree of conditioning. Panic may be associated with the setting or environment in which it occurs, and with interoceptive and/or exteroceptive correlates of early stages of panic itself. Anxiety can have a number of consequences. We have emphasized the possibility that it can potentiate or exaggerate the effects of other triggering events (other CSs, USs, or perhaps endogenous events) that may stimulate the next panic. Importantly, these potentiating effects are further modulated by other CSs and contextual cues that may be present in the background. It is plausible to suppose that conditioned anxiety is an essential process that allows early panics to spiral into PD. The idea is further consistent with clinical evidence suggesting that anxiety often precedes panic attacks and that anxiety seems to contribute to the evocation of panic in the laboratory. It is also consistent with evidence suggesting that PD often begins with panic experiences that can provide potent conditioning trials. We have tended to emphasize the conditioning of anxiety CRs over panic CRs. This emphasis is consistent with the conditioning literature in animals such as rats, in which defensive CRs (e.g., freezing and endogenous analgesia) are not the same as the UR to footshock (activity bursting and pain). However, we believe it is likely that panic itself can also become a CR when certain fear-relevant interoceptive or exteroceptive cues are associated with panic attacks. From a functional perspective, panic reactions, unlike anxiety reactions, are designed to deal with an aversive US that is already in progress. Therefore, cues that are especially proximal in time to panic, such as the interoceptive correlates of panic onset, may be more likely than other kinds of cues to evoke this kind of CR. Support for this idea is largely indirect at this point, coming from a few animal and human conditioning experiments in which the CR is the same as the UR with certain CSs or certain interstimulus intervals (e.g., Cook et al., 1986; Dimberg, 1987; Domjan et al., 1992; Forsyth & Eiffert, 1996, 1998; Stegen et al., 1999; VanDercar & Schneiderman, 1967). Nonetheless, the idea that panic itself can be a CR is also an implication of what we know about conditioning in other behavior systems (e.g., Domjan, 1994, 1997). Further research is necessary before it can be determined what kinds of conditions allow panic itself to be a CR. Importantly, conditioning processes and reactions like those we are describing here may often occur without conscious awareness, perhaps reflecting the operation of neurobiological emotional systems that are dissociable from declarative knowledge systems (see Bechara et al., 1995; R. E. Clark & Squire,

294 Mark E. Bouton, et al.

1998; see also LeDoux, 1996; Öhman, 1996, 1997; Öhman et al., 2000). As noted earlier, evidence suggests that emotional conditioning may be independent of declarative memory and conscious awareness. This aspect of emotional conditioning may obviate the well-known problem that cognitive perspectives have in explaining why panics can occur in the absence of catastrophic thoughts or ideas (e.g., Kenardy et al., 1992; Rachman et al., 1988). We believe that conditioning processes may go a considerable distance in explaining the major features of PD. Vulnerabilities for the Development of PD

As noted earlier, not everyone who experiences occasional stress-related false alarms goes on to develop PD. Indeed, a majority do not. What makes some people who experience panic more vulnerable to developing PD than others? We have already noted that certain psychological concomitants of early panic attacks, such as their perceived controllability and predictability, can influence their perceived intensity and hence their ability to initiate conditioning. Moreover, multiple genetic, temperamental, and experiential factors have some empirical support as contributing to vulnerability to PD, and we now review what we consider to be three of the most prominent sets of factors. Two of these vulnerability factors (one biological and one psychological) are rather nonspecific and may cause vulnerability to many different anxiety, mood, and related disorders. One additional set of psychosocial (experiential) factors is more specific for PD. Each may have an impact by influencing the conditioning process, described previously, that we view as central to the development of PD. Nonspecific Biological (Genetic) Factors

There is clear evidence of the heritability of the trait variously referred to as “trait anxiety,” “neuroticism,” or “negative affect” (e.g., Eysenck, 1967; Gray & McNaughton, 1996; McGuffin & Reich, 1984; Plomin, DeFries, McClearn, & Rutter, 1997). It is unlikely that a single gene will be identified that relates to this heritability, but behavior genetic methods are at least beginning to identify the role of genetic contributions to anxiety and mood disorders. Nonetheless, although there is clear evidence for the heritability of each of these disorders, it should be emphasized that it is only modest in magnitude. For example, Kendler, Neale, Kessler, Heath, and Eaves, (1992), in a large female twin study, estimated that 35% to 39% of the variance in liability to agoraphobia and PD was due to genetic factors. There is also evidence that the genetic vulnerability factors for PD and specific phobias (among other related disorders) may overlap (Kendler, Walters, et al., 1995). The possible genetic overlap between panic and phobias discussed by Kendler, Kessler, et al. (1995) is interesting in light of our learning theory perspective on the etiology of PD. Classical conditioning has long been implicated in the etiology of many specific phobias (e.g., Mineka, 1985a, 1985b; Mineka & Zinbarg, 1996; Öst & Hugdahl, 1981; J. B. Watson & Rayner, 1920), and we are making a parallel argument for PD here. In both specific phobias

A Modern Learning Theory Perspective on the Etiology 295

and PD, anxiety is conditioned to either exteroceptive cues or interoceptive cues, and conditioning of the flight-or-fight response may also occur. There is a long history of documenting the contribution of genetic and temperamental variables such as neuroticism or trait anxiety to classically conditioned aversive emotional responses (e.g., Brush, 1985; Levey & Martin, 1981; Pavlov, 1927), and this may be how the partially overlapping genetic diatheses between phobias and panic disorder may operate. Alternatively, another possible explanation for this partially shared genetic vulnerability may be for the frequency or intensity of experiencing panic attacks themselves, which sets the stage for conditioning of anxiety and/or panic. Some evidence also suggests that genetic contributions to panic and generalized anxiety may differ, at least to some degree (Barlow, 1988; Kendler, Walters, et al., 1995). Elsewhere (Barlow, 1988) we have articulated how separate but perhaps overlapping biological vulnerabilities to anxiety and panic may increase the synergy between anxiety and panic (in which the presence of anxiety increases the probability of the occurrence of panic), long noted by ethologists (Maser & Gallup, 1974) and now emphasized here. Thus, although having a genetically based vulnerability does not cause either panic or anxiety directly, it may well create the appropriate conditions for the occurrence of anxiety or panic or both in people undergoing stress. (Similarly, the tendency to react to stress with specific psychophysiological responses other than panic, e.g., headaches or irritable bowel syndrome, also seems to run in families and may have a somewhat heritable component; Barlow, 1991.) These overlapping genetic vulnerabilities could conceivably influence the onset of PD in one or more of three different ways. First, as already noted, they might influence the potency of panic attacks, making some people more prone to experiencing especially terrifying panic episodes. Second, they might influence the salience of fear-relevant CSs. For example, patients with PD are known to differentially attend to bodily sensations that occur when aroused (e.g., Ehlers & Breuer, 1992, 1996; see Craske, 1999, for a review). This heightened awareness of, or attention to, somatic sensations of arousal could increase their salience and, in turn, increase the probability of developing conditioned anxiety to subsequent panic attacks (see Mackintosh, 1974, for a review of the effects of CS salience on conditioning). Finally, as already noted, genetic vulnerabilities might influence the conditionability of panic and anxiety just as genetic and temperamental variables are known to influence other forms of conditioning (see Mineka & Zinbarg, 1991, 1995, for reviews). With regard to temperamental or personality variables influencing conditioning, it is interesting to speculate that observed sex differences in neuroticism and trait anxiety (cf. Feingold, 1994, for a meta-analysis) could at least partially mediate the corresponding sex differences in PD (approximately 2:1 female:male) and agoraphobia (4:1 for severe agoraphobia). For example, in a meta-analysis, the average effect size (d) for sex differences in trait anxiety across 28 studies was −.30 (Feingold, 1994). Trait anxiety (or neuroticism) is the major personality variable known to be a risk factor for both anxiety and mood disorders (L. A. Clark, Watson, & Mineka, 1994), including PD (Hayward et al., 2000). Furthermore, as discussed earlier, trait anxiety has been consistently shown to increase the conditionability of aversive emotional reactions

296 Mark E. Bouton, et al.

(e.g., Levey & Martin, 1981; Spence & Spence, 1966; Zinbarg & Mohlman, 1998). If higher trait anxiety in females were to lead to greater vulnerability to emotional conditioning, then PD would be more likely to develop in females even if females and males have the same probability of having initial panic attacks, as a number of studies indicate (e.g., King, Gullone, Tonge, & Ollendick, 1993; Telch et al., 1989). Nonspecific Psychological Factors

The genetic and temperamental vulnerabilities just discussed do not operate in isolation, but must combine with psychological vulnerabilities emanating from early experiences to create a diathesis for the development of an anxiety disorder. Two of these early experiential factors seem to be nonspecific, serving as vulnerabilities for most anxiety and mood disorders; another seems to be more specific to PD. Prior Experience with Control and Mastery. There is reason to believe that people who have grown up with a sense of mastery or control over their environments (including their emotional lives) may be less likely to develop anxiety (and subsequently PD) when and if they have an unexpected panic attack than people who have grown up with a relatively impoverished sense of control and mastery over their environment. Developmental psychologists have long argued that an infant’s experience with control over important aspects of his or her environment promotes exploration of novel events and less fearful reactions to strange or arousing stimuli. Infants and young children can gain such a sense of mastery if they have parents who respond to their needs, requests, and initiatives in a contingent way; this is in contrast to the impoverished sense of mastery (or helplessness) that occurs in infants and young children with unresponsive parents who respond to their child in a relatively noncontingent manner. Chorpita, Brown, and Barlow (1998) investigated this general idea in a retrospective study. They operationalized the degree of experience with control or mastery that school-age children had received by using a measure of parenting style that assessed the degree to which the parent discourages autonomy and shows high protection of the child. Previous work had shown that this parenting style may influence the child’s locus of control (i.e., high overprotectiveness is associated with external locus of control in the child; Schneewind, 1995). This parenting style is also associated with anxious and depressive symptoms in the child (e.g., Parker, 1983). In a cross-sectional study, Chorpita et al. found that parental overprotectiveness predicted external locus of control in the children (replicating Schneewind, 1995) and that the locus of control variable mediated the effects of parental overprotectiveness on clinical symptoms of anxiety and depression. Unfortunately, these ideas regarding the role of mastery in reducing susceptibility to panic and anxiety are difficult to study experimentally in human infants and children because it is unethical to manipulate directly the controllability of a child’s environment for significant periods of time. However, experimental evidence generally supports this idea in a study conducted in infant monkeys that were reared in controllable versus uncontrollable environments for the first year of life (Mineka, Gunnar, & Champoux, 1986). In the controllable environments, the master monkeys had levers to press and

A Modern Learning Theory Perspective on the Etiology 297

chains to pull to deliver themselves food, water, and treats. In the uncontrollable environments, the yoked monkeys received access to the same food, water, and treats, but these were delivered uncontrollably whenever a master monkey earned a reinforcer. When tested in several frightening and novel situations between 7 and 11 months of age, the master monkeys reared with control adapted more quickly in several different fear-provoking situations compared with the yoked monkeys reared without control. Thus, early experience with control and mastery over positive reinforcers appears to affect the level of fear that novel and frightening events evoke, paralleling what is thought to occur in early human development. By decreasing the intensity of reactions to frightening events or by increasing the rate of habituation to them, having a sense of control or mastery may thereby decrease the conditioning of panic or anxiety. One thing that makes this example especially noteworthy is that the sense of mastery generalized across domains (in this case, from appetitive to aversive). Related findings have also been found in several other animal studies in the learned helplessness tradition. For example, Joffe, Rawson, and Mulick (1973) reported that rats raised in environments in which they had control over access to food, water, and visual stimulation later showed less emotionality and more exploratory behavior in a novel situation (known to evoke some anxiety) than did rats reared in yoked-uncontrollable environments. Hannum, Rosellini, and Seligman (1976) also showed that immunization effects (at least within an aversive domain) could be demonstrated over a substantial time frame. They found that immunization (or mastery) experiences with controllable shock given to young weanling rats had a protective effect when rats were later exposed as adults to inescapable shocks. Finally, J. Williams and Maier (1977) found immunization effects even when different kinds of aversive stimuli were used in the immunization (mastery) and helplessness induction phases (e.g., experiencing escaping from cold water immunized rats against the effects of subsequent exposure to uncontrollable footshocks). In combination, these studies suggest that learning a sense of mastery or control in one or more areas of life (but not necessarily related to control over aversive stimuli or emotions) could generalize to situations in which aversive stimuli or emotions are involved, such as coping with a few unexpected panic attacks. Thus far, we have emphasized the ways in which an enhanced sense of mastery and control may immunize against later anxiety and possibly depression relative to some normative level of controllability. Conversely, high levels of prior experience with uncontrollable stressors may also enhance vulnerability to anxiety and depression relative to some normative baseline of experience with stress. Relevant to this latter line of research, stressful life events have been found to play two somewhat distinctive roles in the etiology of PD. First, stressful life events such as early parental death, separation, or divorce have been found in many, but not all, studies to enhance vulnerability to development of some of these disorders, most notably PD and agoraphobia (e.g., Kendler et al., 1992; Tweed, Schoenback, George, & Blazer, 1989). Second, for PD (and major depression), higher than normal levels of stressful life events have been found to precede and possibly precipitate the onset of the disorder in vulnerable individuals (e.g., see reviews by Monroe & Simons, 1991, for depression; Craske, 1999, for anxiety disorders).

298 Mark E. Bouton, et al.

Neurobiological mechanisms might also be involved in these effects. One possible mechanism is the one proposed by Rosen and Schulkin (1998), who argued that both distal (e.g., early childhood) and proximate (e.g., past few months) stressful life events, both physical and psychological, may serve to sensitize fear circuits in the brain (primarily the amygdala), making them “hyperexcitable” or easier to trigger for a long time. Such hyperexcitability might also occur simply as a function of the experience of panic attacks themselves, which are often perceived as terrifying life-threatening events. In addition, building on the pioneering efforts of Ader and Denenberg, Nemeroff et al. (e.g., Heim & Nemeroff, 1999; Ladd et al., 2000) noted permanent effects on brain function of early stressful experiences (separation) in rat pups. Specifically, early stressful experiences exert a significant impact on the developing hypothalamic-pituitary-adrenal (HPA) axis, causing an increase in the organism’s response to psychologically stressful events as adults (e.g., exposure to a novel environment, restraint) but not physical stress (e.g., hemorrhage). This very specific heightened responsiveness to psychological stress in adulthood seems to be a function of hyperreactive HPA axis responding to these stressors as indexed by markedly elevated corticosterone and adrenocorticotropic hormone. Thus, early experience with uncontrollable stress may create a nonspecific diathesis for later life events perceived as unpredictable and/or uncontrollable. Prior Experience with Unpredictability. Lack of control over one’s environment often implies lack of predictability as well (i.e., if one cannot control an event, one often does not know when it will occur or terminate). As noted earlier, a good deal of research shows that the predictability versus unpredictability of uncontrollable events (possibly including panic attacks, cf. Craske et al., 1995) has a large impact on the amount of stress generated (e.g., Overmier, 1985; Seligman & Binik, 1977; J. Weiss, 1971; see Mineka & Hendersen, 1985, for a review). Unfortunately, there is very little in the way of longitudinal research during early development on the effects of being raised in a highly unpredictable versus predictable environment. However, one important study in monkeys investigated this issue. Coplan et al. (1996) found that infant monkeys whose mothers experienced unpredictable foraging conditions (food sometimes scarce and sometimes abundant) showed higher levels of corticotropin-releasing factor (one of the major stress hormones) in cerebrospinal fluid as adults than did infants whose mothers had either a predictable overabundance of food or chronically (i.e., predictably) scarce food. In summary, these findings regarding the effects of early experience with uncontrollable or unpredictable events, in conjunction with a larger web of related findings from developmental psychology, may have considerable relevance for understanding an individual’s vulnerability to PD (Barlow, 1988; Barlow et al., 1996; Chorpita & Barlow, 1998; Mineka, 1985a; Mineka & Kelly, 1989; Mineka & Zinbarg, 1991, 1995, 1996). The mechanisms for these effects may be consistent with a traditional learned helplessness account (e.g., Maier & Seligman, 1976; Peterson, Maier, & Seligman, 1993). Specifically, learning that one does not have control over important life events at one point in time can produce associative and motivational deficits that result in failure to learn control at future points at which control is indeed possible. Alternatively or

A Modern Learning Theory Perspective on the Etiology 299

additionally, a nonassociative neurobiological sensitization mechanism may also operate (e.g., Heim & Nemeroff, 1999; Rosen & Schulkin, 1998). Once again, this vulnerability is viewed as nonspecific to PD and most likely undergirds the development of all or most anxiety and mood disorders as well as related disorders. Specific Psychological Factors: Vicarious and Instrumental Learning

We have discussed both nonspecific biological and nonspecific psychological vulnerabilities for anxiety and mood disorders. In addition, there also seem to be some specific early vulnerability factors that may predispose only some people who experience a panic attack to develop PD as opposed to other anxiety disorders or no disorder. One rather specific set of psychological factors not yet reviewed concerns early learning experiences regarding the potential dangers of unexplained bodily sensations and how to respond to them based on observations of one’s parents’ behavior (see Levy, 1998, for a review). In an early study of these influences in the realm of medical illness, Turkat (1982) studied 27 diabetics whose parents had not been chronically ill while they were growing up. Two thirds of these individuals reported that their parents had engaged in sick role behavior (such as not going to work, canceling activities, or receiving special attention) when temporarily ill. Of the individuals with diabetes whose parents had shown sick role behavior when ill, 66% reported illness-related avoidance as adults themselves. Another 33% reported that their parents had not engaged in sick role behavior when they were ill while they were growing up; only 22% of these people with diabetes reported illness-related avoidance themselves. Thus, diabetics whose parents showed sick role behavior appeared much more likely to demonstrate similar illness behavior themselves as well as work and responsibility avoidance, leading Turkat to conclude that “parental reactions to illness may be transmitted to their offspring as well” (1982, p. 522). In addition, the individuals with diabetes whose parents had engaged in sick role behavior made more visits to their doctors and had more hospital admissions and more days ill than the other group, even though there was no evidence that they were more seriously ill based on physiological measures. In another large-scale study, Whitehead, Winget, Fedoravicius, Wooley, and Blackwell (1982) found that adults were more likely to miss school or work as a result of illness and to seek medical help if their parents had reinforced them (e.g., with toys or special food) when they were ill as children. Moreover, in a later study of women, Whitehead, Bush, Heller, and Costa (1986) showed some correlational specificity to such relationships. For example, if the women had been encouraged to be cautious as children when they had colds, they were more likely to seek help for nongynecological problems. However, if they had been reinforced for sick role behavior for menstrual symptoms while growing up, they were more likely to miss school or work and seek medical attention for menstrual symptoms as adults. Turning to PD, Ehlers (1993) suggested that any learning experience encouraging sick role behavior and/or negative evaluations of somatic symptoms associated with panic attacks may create a potential specific vulnerability factor for

300 Mark E. Bouton, et al.

PD. In a retrospective study, Ehlers (1993) assessed 121 panic patients (including 24 in remission for at least 6 months), 86 infrequent panickers, 38 patients with other anxiety disorders (mostly specific phobias), and 61 normal controls for learning experiences when they were children and adolescents with respect to somatic symptoms. All individuals were asked about parental encouragement of sick role behavior when they were experiencing panic symptoms (as well as their frequency of occurrence), observation of parental sick role behavior and of frequency of parental panic symptoms, parental encouragement of sick role behavior when sick with colds, number of chronically ill family members (where chronic illness was defined as six months in duration or longer), and frequency of uncontrolled behavior of household members (because of rage or being drunk). Ehlers reported that the frequency of uncontrolled behavior was assessed because of the clinical observation that patients with PD who report fear of loss of control often have parents who abuse substances. No differences were found between patients with PD and infrequent panickers on most variables, and the patterns of correlations between the groups were identical. All three anxiety groups reported greater frequency of uncontrolled behavior in their parents compared with controls. In addition, patients with PD and infrequent panickers reported having observed parents experiencing panic symptoms more frequently than anxiety disorder or normal controls; all three anxious groups also reported more panic symptoms in themselves while growing up than controls. The four groups had comparable parental encouragement of sick role behavior during the experience of panic-like symptoms (e.g., special attention and instructions to take special care of themselves and to avoid strenuous activities or social engagements). However, because of the differences in frequency with which actual symptoms were experienced across the four groups, there was more parental reinforcement of panic symptoms in the participants in the three anxiety groups relative to controls and more parental engagement in sick role behavior when the parents had panic symptoms in the two panic groups relative to the other groups. In contrast, there were no differences between anxiety disorder groups and controls in reported parental encouragement of sick role behavior in the event of colds. Moreover, among the two panic groups combined, there were some modest but significant correlations between a combined index of this encouragement of sick role behavior and responses on two widely used measures of fear of bodily sensations: the Body Sensations Questionnaire and the Agoraphobic Cognitions Questionnaire (Chambless, Caputo, Bright, & Gallagher, 1984; Ehlers, Margraf, & Chambless, 1992). Overall, the results suggest that both clinical and nonclinical panickers had a history of sick role encouragement when experiencing panic symptoms (but not cold symptoms); controls reported significantly fewer symptoms in themselves and their parents, although when such symptoms occurred sick role behavior was also reinforced. Ehlers (1993) also found that patients with PD and infrequent panickers reported a higher number of chronic illnesses in their households while growing up compared with those with other anxiety disorders or controls. She noted, “Observing physical suffering can also contribute to the evaluation that somatic symptoms are dangerous and that special care is needed” (p. 276). Whether this learning should be interpreted as an instance of vicarious

A Modern Learning Theory Perspective on the Etiology 301

classical conditioning (cf. Mineka & Cook, 1993), in which the child associates his or her own distress at watching the parent’s distress with symptoms they are showing, or instrumental learning, in which the child observes the parent’s reinforcement for doing certain things in response to certain symptoms, awaits further analysis (although these two possibilities are not mutually exclusive). In the former, illness cues might become associated with negative affect through evaluation of somatic symptoms as dangerous. In the latter, family members might have reinforced attributions about being sick and sick role behavior more directly. Finally, there is some evidence that prior experience causes individuals to focus their anxiety on specific constellations of responses within a panic attack, such as respiratory symptoms (e.g., breathlessness), vestibular symptoms (e.g., dizziness), cardiovascular symptoms (e.g., increased heart rate), or symptoms of dissociation (e.g., depersonalization). Reports of a preexisting and presumably learned sensitivity to suffocation cues differentially predicts panic attacks to respiratory challenges such as breathing through a straw (e.g., Taylor & Rachman, 1994) or breathing into a paper bag (McNally & Eke, 1996). In addition, Craske (1999) reviewed some evidence that early experience with specific chronic illness in family members (e.g., chronic obstructive pulmonary disease) may lead to enhanced sensitivity to specific constellations of sensations such as respiratory symptoms. Also, Barrett, Rapee, Dadds, and Ryan (1996) observed parents of socially anxious children unwittingly reinforcing specific avoidant behavior associated with hypothetically threatening social situations, and this behavior increased after family discussions of ambiguously threatening social situations. Barrett et al. also found that parents of children with specific phobias reinforced their children for avoidance of hypothetically physically threatening situations. These results suggest that parents convey specific fear information to children including, perhaps, information about specific somatic sensations. Of course, many of the studies we have discussed in this section have major methodological limitations in that they are retrospective in nature, and recall of early learning experiences may be colored by the participant’s current emotional tendencies and symptomatology. Moreover, given the genetic vulnerabilities contributing to PD, one must consider the possible role of genetic factors in some of the associations described here. Nevertheless, they do suggest that early learning factors (including vicarious ones) may contribute to a specific vulnerability for PD by sensitizing individuals to the potential danger of somatic sensations. Thus, when a stress-related panic attack occurs in someone with such specific vulnerabilities, certain somatic sensations may be especially salient for that person, leading to especially robust conditioning of anxiety and/or panic to those cues. Summary

Vulnerability factors for PD seem to include a variety of general and specific factors; that is, some vulnerability factors may predispose to other anxiety and related disorders, and some may be rather specific for PD. Genetic evidence on the specificity issue is somewhat inconsistent, but Kendler, Walters et al.’s

302 Mark E. Bouton, et al.

large twin study (e.g., 1995) suggests that there may be some shared vulnerability between PD and specific phobias, which would be consistent with the theory proposed here. This is because the shared vulnerability could either be for experiencing especially intense or frequent panic-fear episodes, which set the stage for conditioning of specific phobias or for the onset of PD, or because of enhanced conditionability to exteroceptive or interoceptive cues. In addition, early experience with uncontrollable and unpredictable events is likely to serve as a psychosocial vulnerability factor for many anxiety disorders (Barlow, 1988; Barlow et al., 1996; Chorpita & Barlow, 1998; Mineka, 1985a, 1985b; Mineka & Zinbarg, 1995, 1996). Finally, vicarious learning of anxiety focused on certain bodily sensations and/or reinforcement of illness behavior while growing up may serve as a more specific vulnerability factor for PD but not other anxiety disorders (Ehlers, 1993). Relationships Between a Learning Theory Perspective and Other Accounts of PD A contemporary learning theory perspective appears to be consistent with much that is known about PD. It is also broadly consistent with what we view as the positive aspects of the other approaches and theories reviewed at the beginning of this article. For example, the data supporting the role of AS in the development of PD fit with the notion of a specific psychological vulnerability based, perhaps, on vicarious learning encouraging sick role behavior or negative evaluation of somatic symptoms such as those occurring during panic attacks. In our approach, individuals who go on to develop PD would learn AS or, more specifically, that somatic symptoms are potentially dangerous. In support of this idea, a reanalysis of data from the Barlow laboratory (Zinbarg, Brown, Barlow, & Rapee, in press) suggests that the physical harm factor of the ASI (anxiety focused on somatic sensations), in contrast to other factors derived from this scale, accounts for almost all of the variance in predicting whether panic attacks will occur in patients with PD who are provoked with CO2 inhalations. In addition, Hayward et al. (2000) also found that the physical harm factor of the ASI was the only significant predictor from the ASI of naturally occurring panic attacks, when controlling for depression, in their 4-year prospective study of adolescents. However, as noted earlier, high scores on the ASI do not invariably predict the development of panic attacks (and in the Hayward et al., 2000, study, high negative affectivity was a substantially better predictor). Moreover, no study has yet shown them to predict onset of full-blown PD in an unselected population. Thus, in our view AS plays an important role in the development of PD but at the level of only one of our three hypothetical vulnerabilities. We also believe that an approach based on contemporary learning theory is consistent with data that seem to support cognitive theories emphasizing a role for catastrophic misinterpretation (e.g., Beck & Emery, 1985; D. M. Clark, 1986, 1988, 1996). As we noted earlier, catastrophic misinterpretations may indeed accompany many panic attacks. This may be because such thoughts are a natural part of the constellation of responses involved in panic or because they might have been encouraged and reinforced in a manner analogous to the

A Modern Learning Theory Perspective on the Etiology 303

sick role behaviors and negative evaluations identified by Ehlers (1993) during earlier experience of panic-like symptoms during childhood and adolescence. If they actually play a causal role in generating or exacerbating panic, they may do so because they serve as CSs that have been associated with panic. Through classical conditioning, they may come to elicit anxiety and panic when they occur again. Although a causal role for catastrophic cognition is thus not outside the scope of a learning theory analysis, we are less convinced than other theorists that catastrophic thoughts are necessary to generate panic attacks. Whether they are sufficient remains to be determined. It is worth noting that the modal observation during panic attacks or extreme fear during confrontation with a phobic situation (as opposed to the period of anxiety preceding the confrontation) is an absence, or substantial diminishment, of cognitive activity such as catastrophic cognitions or other conscious appraisals of danger (Craske, 1999; Last, O’Brien, & Barlow, 1985; S. L. Williams, Kinney, Harap, & Liebmann, 1997). D. M. Clark (e.g., 1996) and others (Beck & Emery, 1985; Salkovskis, 1988) have, of course, emphasized the role of catastrophic cognitions in causing panic attacks as well as several threads of evidence that seem to justify their importance. However, few of the threads truly force one to accept a causal role. For example, D. M. Clark (e.g., 1996) reviewed evidence from a variety of studies supporting the idea that panic patients are more likely to choose negative interpretations of ambiguous internal events than are normals and other anxiety-disordered controls. Although such results are consistent with cognitive theory, he acknowledged that they do not show that catastrophic cognitions play a causal role in creating panic but rather may be epiphenomenal. For example, these cognitions may be one manifestation of a preexisting psychological vulnerability that facilitates the conditioning of anxiety to CSs signaling subsequent panic attacks. D. M. Clark (1996) also cited a study by Ehlers, Margraf, Roth, Taylor, and Birbaumer (1988) in which panic patients were given false auditory feedback indicating a sudden increase in heart rate to determine whether activating patients’ catastrophic cognitions would increase their anxiety more than that in normal controls. Panic patients did show greater increases in heart rate, blood pressure, skin conductance, and self-reported anxiety than normals. However, there is no need to invoke catastrophic misinterpretations to explain these findings, even though they may well occur. That is, one can easily construe the false heart rate feedback as an exteroceptive CS similar to a conditioned interoceptive CS (increased heart rate); conditioned anxiety may merely generalize to it. Moreover, one would expect signs of an elevated heart rate to be an effective stimulus for anxiety in patients with PD, but not in normal controls who do not have the same conditioning history. Cognitive theorists (e.g., D. M. Clark, 1996) have also cited a study from Clark’s laboratory (D. M. Clark, Salkovskis, & Anastasiades, 1990) in which panic patients, recovered panic patients, and normal controls read pairs of words that involved combinations of bodily sensations and catastrophes (e.g., palpitations-dying, breathless-suffocate, numbness-stroke) and were asked to rate their anxiety and occurrence of any DSM-III panic symptoms. The results indicated that 83% of the panic patients (but none of the recovered controls

304 Mark E. Bouton, et al.

or normal controls) had a DSM-defined panic attack while reading the cards with these pairs of words. Although these results were seen as evidence that catastrophic misinterpretations were sufficient to provoke panic attacks, a role for catastrophic misinterpretations per se was not directly established. For example, the word pairs might alternatively evoke thoughts and images that have been associated with panic and thus have become CSs capable of evoking panic. A learning theory perspective accepts the role of other cognitive factors. It is undoubtedly true that certain kinds of cognitive processes other than catastrophic cognitions can also influence the perceived intensity of panic and thus influence its impact as a US on the development of conditioning. The learning literature has long emphasized predictability and controllability, which may be mediated by cognitions that modulate the perceived intensity of aversive events. Two experiments that have been interpreted to suggest the role of catastrophic cognitions may merely emphasize the importance of the perception of predictability and controllability. Panic patients were given virtually no explanation of what to expect before inhaling 5% CO2 (Rapee, Mattick, & Murrell, 1986) or infusing sodium lactate (D. M. Clark et al., 1990), or they were given detailed information about what sensations to expect and that these sensations were due to the experimental agent. In both studies, the patients given the detailed explanations were significantly less likely to report panicking than those not given much explanation of what to expect. As discussed earlier, predictable aversive events are generally perceived as less stressful than are unpredictable aversive events. One way of making an event predictable is by providing extensive information about it ahead of time, as was done in these experiments (see Leventhal, 1982; Leventhal, Brown, Shacham, & Engquist, 1979; Mineka & Hendersen, 1985). Thus, panic patients, like normal controls, find predictable stressors to be far less stressful (and, therefore, less likely to provoke panic) than they find unpredictable stressors. The effect is cognitively mediated, but prior researchers have not found it necessary to invoke catastrophic misinterpretations to explain it. The Sanderson et al. (1989) study, already discussed, is also consistent with this point: The effects of perceived control (rather than predictability) over the rate of CO2 infusion underscore the importance of a sense of control in reducing vulnerability to anxiety. Controllability and predictability are cognitive constructs that are relatively easy to operationalize and are connected with a long tradition of experimental research (e.g., Mineka & Henderson, 1985; Minor, Dress, & Overmier, 1991; Peterson et al., 1993; Seligman, Maier, & Solomon, 1971). We should also acknowledge that, as mentioned earlier, classical conditioning itself may sometimes give rise to cognitive processes that may permit input from other cognitive sources. Although we have emphasized the idea that conditioning processes can be engaged without consciousness or awareness, many forms of human classical conditioning may involve the acquisition of explicit expectancies of the US. In such cases, it might be possible to influence conditioned responding with verbal information or cognition. Consistent with this idea, Lovibond (e.g., 1993) argued that verbal information about the nature of the US can be sufficient to cause changes in conditioned electrodermal responses elicited by conditioned stimuli (e.g., see Grings et al., 1973). We

A Modern Learning Theory Perspective on the Etiology 305

believe that this sort of verbal influence on conditioning is probably not universal and depends on the type of conditioning and the brain system involved. There is a need for more research on the interaction between verbal input and conditioning processes. Nonetheless, at this point we should expect some overlap between conditioning and these overtly “cognitive” processes (see also Öhman & Mineka, 2001). A modern view of conditioning may also suggest reinterpretation of other evidence sometimes cited in favor of the catastrophic misinterpretation view. In an important early study suggesting the role of cognitive factors in determining panic reactions, D. M. Clark and Hemsley (1982) had normal participants hyperventilate. There was variability in their responses to this event. Participants who recalled experiencing the sensation during sex or while they were high on a pleasant drug rated the sensations as positive; those who recalled an unpleasant experience (e.g., fainting) rated the sensations as negative. To us, this result implies a clear role for the participant’s associative learning history, the subject matter of conditioning. Memory retrieval is part of what the associative learning process represented by conditioning is all about (e.g., Bouton, 1994a). Like classical conditioning itself, the Clark and Hemsley result looks to us like another interesting example of associative learning. Finally, there has also been an emphasis on the idea that the success of any treatment will depend on cognitive changes having occurred during the course of therapy (D. M. Clark, 1996, p. 322). In a treatment study, D. M. Clark et al. (1994) compared the effects of cognitive therapy for panic with the effects of applied relaxation treatment and imipramine. Collapsing across the three groups, misinterpretation of bodily sensations at the end of treatment was a significant predictor of a composite measure of panic/anxiety at follow-up even when partialing out the level of panic-anxiety at the end of treatment. Moreover, among those patients who were panic free at the end of treatment, misinterpretations of bodily sensations at the end point predicted subsequent relapse. On this point, we note that catastrophic cognitions seem to be part of a “context” or constellation of cues that have been associated with panic or are even part of the response itself, and may signal or mark residual anxiety conditioned specifically to panic attacks. If they are not extinguished, PD could well return. In addition, several treatment studies have shown that extensive exteroceptive and interoceptive exposure therapy may be equally effective as cognitive therapy in reducing panic disorder, in both the short term and long term (Margraf & Schneider, 1995; Telch, 1995). A contemporary learning theory perspective may uniquely accommodate these different forms and mechanisms of therapy. In summary, many findings that seem uniquely interpretable from the perspective of cognitive theory (e.g., Beck & Emery, 1985; D. M. Clark, 1986, 1988, 1996) are not incompatible with contemporary learning theory. The evidence favoring the idea that catastrophic cognitions cause panic is weak. A learning theory perspective has advantages, moreover, in that it can account for panic attacks that occur apparently without identifiable catastrophic misinterpretations, including nocturnal panic (e.g., Craske, 1999; Kenardy & Taylor, 1999; Rachman et al., 1988). This perspective also recognizes the distinction between anxiety and panic, which seem central to current psychometric (e.g.,

306 Mark E. Bouton, et al.

T. A. Brown et al., 1998) and neurobiological work (e.g., Fanselow, 1994; Gray & McNaughton, 1996). In addition, it allows prediction of return of fear and anxiety based on a thorough analysis of context and conditioning (e.g., Bouton, 1991b) and a more explicit analysis of factors modulating the acquisition of fear, panic, and PD based on well-established paradigms in the laboratories of experimental psychology. Implications and Conclusion We are in a good position at this point to summarize our argument and mention some of its major implications. First, in keeping with earlier approaches to PD, our perspective emphasizes a fundamental role for early conditioning episodes in the etiology of the disorder. This idea can be seen as a strength of a learning perspective, because it constitutes a testable explanation of PD etiology. Prospective data on early panic attacks and their associative and psychological consequences are needed. We would expect, for example, that after an initial panic attack, people—especially those with one or more of the general or specific vulnerabilities discussed earlier—would show more anxiety to cues associated with the first attack. Such results could theoretically be obtained from prospective diary monitoring studies in vulnerable adolescents or young adults (similar to those reported by Basoglu et al., 1992; Kenardy et al., 1992; Kenardy & Taylor, 1999, although with additional measures). However, we also emphasize the fact that current learning theory does not regard conditioning as an inevitable consequence of CS-US pairings. Instead, the extent to which conditioning develops depends on many additional factors, including the person’s previous experience with the CS and the US and with the “informativeness” of the various CSs present on the conditioning trial and on other modulating factors in the background, and so on. As we have emphasized throughout this article, laboratory research on conditioning and learning has progressed considerably since the 1960s, and it suggests a surprisingly nuanced perspective on the associative learning involved in panic and other anxiety disorders (see also Mineka, 1985a; Mineka & Zinbarg, 1996). A second important aspect of our approach is that anxiety and panic are seen as separable aspects of PD. Anxiety is not merely a weak version of panic, and panic is not merely a strong form of anxiety. We see each state, and the constellation of behaviors and physiological responses connected with each, as serving different functions. Anxiety prepares the system for an anticipated trauma, whereas panic deals with one that is already in progress. Anxiety and panic are thus different. This perspective is consistent with data addressed early in this article suggesting that anxiety and panic seem at least in good part phenomenologically, psychometrically, ethologically, and neurobiologically distinct. Anxiety and panic do interact, however, and our approach assumes that their interaction is central to the development of PD. We propose that anxiety potentiates panic, and the development and presence of conditioned anxiety, therefore, serve to exacerbate subsequent panic attacks. A third crucial feature of our perspective, then, is the idea that the conditioned anxiety that comes to be elicited by interoceptive and exteroceptive cues associated with panic

A Modern Learning Theory Perspective on the Etiology 307

serves to augment future panic reactions. Anxiety thus becomes a precursor of panic. The approach fits conditioning research, which suggests that anxiety is perhaps the major response learned in aversive conditioning situations, and that anxiety can function to exacerbate CRs and URs elicited while in that state (e.g., Lang, 1994, 1995; Lang et al., 1990; Wagner & Brandon, 1989). It is also consistent with available data that suggest that panic attacks are very often preceded by anxiety in patients with PD (e.g., Barlow, 1988; Basoglu et al., 1992; Kenardy & Taylor, 1999). We expect that prospective data sets would show that, once conditioned anxiety develops as a consequence of the first panic attack, it would potentiate and exacerbate subsequent panics and thus begin the spiral into PD. It should also be noted that the conditioned elicitation of anxiety does not rely on conscious processing (e.g., LeDoux, 1996; Öhman et al., 2000). Therefore, we predict that panic can be potentiated in the absence of conscious thought or reflection, as is more consistent with functioning in largely subcortical emotional networks connected to defensive motivational systems (Barlow, in press; Lang, 1995). Although we believe that the conditioning of anxiety is a major consequence of the conditioning made possible by panic attacks, we also suspect that panic itself may become conditioned directly to certain kinds of cues. That is, although anxiety is a prevalent CR, panic itself, instead of anxiety, may also emerge as a CR to some of the available CSs. As we reviewed earlier, laboratory research suggests that the nature of the CR depends on a number of factors, including the qualitative nature of the CS and its temporal proximity to the US. We predict that proximal, fear-relevant cues, which might include early somatic aspects of panic itself, are especially likely to have the ability to elicit panic CRs. However, this issue needs more research; we need a more complete understanding of the kinds of cues or circumstances that allow a CS associated with an aversive US to control a CR that resembles the UR. We also predict that those CSs that do elicit panic should do so more strongly when they are presented after the evocation of anxiety. That is, anxiety should potentiate panic whether it is a CR or UR. A fourth implication of our analysis concerns treatment. Like other conditioning perspectives, we expect that treatments that involve extinction or counterconditioning exposure to the interoceptive and exteroceptive CSs influencing the disorder are the most likely to yield success. However, we explicitly accept many different kinds of events and cues—interoceptive, exteroceptive, verbal, and cognitive—as potential CSs involved in the disorder. Therefore, we predict that approaches that entail extinction or counterconditioning exposure to all of these kinds of events will be most successful, particularly if they are conducted in a way that recognizes the crucial role of context in controlling extinction and other retroactive interference effects (e.g., Bouton, 1991b; Bouton & Nelson, 1998b; Bouton & Swartzentruber, 1991). Moreover, we also predict that treatments designed in part to extinguish the anxiety-reducing properties of safety behaviors (such as carrying a pill bottle or an umbrella) will be very useful (for preliminary evidence see Salkovskis et al., 1996). In our view, this is because such safety behaviors serve to protect fear of various exteroceptive cues (and potentially interoceptive cues as well) from extinction.

308 Mark E. Bouton, et al.

If therapists endeavor to expose the anxiety-providing cues without allowing safety behaviors, fear of those cues should extinguish more fully. In the last analysis, a good theory is a useful tool to guide future scientific exploration of a given topic. Unfortunately, early learning theory approaches to the etiology of panic and other anxiety disorders were overly simplistic, leading to some of the confusion regarding the usefulness of these approaches (e.g., McNally, 1990, 1994). However, the underlying science has developed rapidly over the decades and has been enriched by increasingly important developments in the biological and cognitive bases of learning reviewed here. In our view, a modern learning theory approach will provide the soundest base for future theoretical and empirical developments in the study of PD. Thus, we look forward to precise experimental tests of the various components of our approach, some of which have been suggested here. It is this kind of activity that will ultimately best advance our understanding. We also believe that contemporary learning theory, perhaps integrated with related research on neurobiology, will continue to provide an essential framework for studying the development and maintenance of other anxiety and emotional disorders. Many of the processes described here in the context of PD are equally applicable to other anxiety and emotional disorders, although each has distinctive features (Barlow, 1988, in press; T. A. Brown et al., 1998; Mineka, 1985a; Mineka & Zinbarg, 1995, 1996, 1998). Although we have touched on evidence supporting the existence of both biological and psychological vulnerabilities, it is the etiological process that occurs in the context of these vulnerabilities on which we have focused most intently (i.e., what happens during and after the first panic attack). Finally, efforts at prevention, one ultimate goal of the study of psychopathology, will benefit from a greater understanding not only of etiological processes but also of the development of various vulnerabilities. Thus, we have attempted to point out how modern learning theory conceptualizes psychosocial vulnerabilities that, when combined with biological vulnerabilities, set the stage for the development of PD. In so doing, we have attempted to reflect the complexity of modern learning theory and of the psychopathology of PD as well as the enormous task, both theoretical and empirical, that remains before us before we fully understand the genesis of PD. Notes 1

Preparation of this article began while the authors were Fellows at the Center for Advanced Study in the Behavioral Sciences (CASBS) during 1997–1998 (supported by John D. and Catherine T. MacArthur Foundation Grant 95–32005–0). David H. Barlow was also supported by the CASBS Foundations Fund for Research in Psychiatry as a Fritz Redlich Fellow. The final phases of article preparation were supported by National Science Foundation Grant IBN 9727992. We thank the Center and its staff for providing an ideal environment that makes even unplanned collaborations attractive and possible. We also thank David M. Clark, Anke Ehlers, Michael Fanselow, Chris Hayward, Peter Lovibond, Colin MacLeod, Richard McNally, C. Barr Taylor, Richard Zinbarg, and two anonymous reviewers for their comments on drafts of the article at various stages.

A Modern Learning Theory Perspective on the Etiology 309 2

3

It is worth noting that summation might not occur in all Pavlovian methods or response systems (e.g., Bouton, 1984; Bouton & King, 1986; Rescorla & Coldwell, 1995). One factor (among others) that appears to facilitate summation is similarity between the separate CSs and the compound (Pearce, Aydin, & Redhead, 1997). As discussed extensively elsewhere (e.g., Mineka et al., 1984; Mineka & Hendersen, 1985), the mechanisms through which controllability operates may functionally be mediated by the added predictability over offset of a US that controllability affords (namely, when the US will end). However, for our purposes, the precise mechanisms through which controllability exerts its effects are not important and are not detailed here.

References Akins, C. K., Domjan, M., & Gutierrez, G. (1994). Topography of sexually conditioned behavior in male Japanese quail (Cotumix japonica) depends on the CS-US interval. Journal of Experimental Psychology: Animal Behavior Processes, 20, 199–209. American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington D.C.: Author. Anderson, D. C., Crowell, C. R., DePaul, M., & McEachin, J. (1997). Intensification of punishment effects through exposure to prolonged, fixed-duration shocks: The role of shock cues as a stimulus for fear. Animal Learning & Behavior, 25, 68–83. Baldwin, J. D., & Baldwin, J. I. (1998). Behavior principles in everyday life (3rd ed.). Englewood Cliffs, NJ: Prentice Hall. Barlow, D. H. (1988). Anxiety and its disorders: The nature and treatment of anxiety and panic. New York: Guilford Press. Barlow, D. H. (1991). The nature of anxiety: Anxiety, depression, and emotional disorders. In R. M. Rapee & D. H. Barlow (Eds.), Chronic anxiety: Generalized anxiety disorder and mixed anxiety-depression (pp. 1–28). New York: Guilford Press. Barlow, D. H. (in press). Anxiety and its disorders: The nature and treatment of anxiety and panic (2nd ed.). New York: Guilford Press. Barlow, D. H., Brown, T. A., & Craske, M. G. (1994). Definitions of panic attacks and panic disorder in the DSM-IV: Implications for research. Journal of Abnormal Psychology, 103, 553–564. Barlow, D. H., Chorpita, B. F., & Turovsky, J. (1996). Fear, panic, anxiety, and disorders of emotion. In D. A. Hope (Ed.), Perspectives on anxiety, panic, and fear (The 43rd Annual Nebraska Symposium on Motivation) (pp. 251–328). Lincoln: Nebraska University Press. Barlow, D. H., & Craske, M. G. (2000). Mastery of your anxiety and panic: Client workbook for anxiety and panic (MAP–3). San Antonio, TX: Graywind Publications. Barlow, D. H., & Lehman, C. L. (1996). Advances in the psychosocial treatment of anxiety disorders: Implications for national health care. Archives of General Psychiatry, 53, 727–735. Barrett, P. M., Rapee, R. M., Dadds, M. M., & Ryan, S. M. (1996). Family enhancement of cognitive style in anxious and aggressive children. Journal of Abnormal Child Psychology, 24, 187–203. Basoglu, M., Marks, I., & Sengiin, S. (1992). A prospective study of panic and anxiety in agoraphobia with panic disorder. British Journal of Psychiatry, 160, 57–64. Bechara, A., Tranel, D., Damasio, H., Adolphs, R., Rockland, C., & Damasio, A. R. (1995). Double dissociation of conditioning and declarative knowledge relative to the amygdala and hippocampus in humans. Science, 269, 1115–1118.

310 Mark E. Bouton, et al. Beck, A. T., & Emery, T. (1985). Anxiety disorders and phobias. New York: Basic Books. Blanchard, D. C. (1997). Stimulus, environmental, and pharmacological control of defensive behaviors. In M. E. Bouton & M. S. Fanselow (Eds.), Learning, motivation, and cognition: The functional behaviorism of Robert C. Bolles (pp. 283–303). Washington D.C.: American Psychological Association. Blanchard, D. C., Blanchard, R. J., & Rodgers, R. J. (1991). Risk assessment and animal models of anxiety. In B. Olivier, J. Mos, & J. L. Slangen (Eds.), Animal models in psychopharmacology (pp. 117–134). Basel, Switzerland: Birkhäuser. Blanchard, R. J., & Blanchard, D. C. (1969). Crouching as an index of fear. Journal of Comparative & Physiological Psychology, 67, 370–375. Blanchard, R. J., & Blanchard, D. C. (1987). An ethoexperimental approach to the study of fear. The Psychological Record, 37, 305–316. Bolles, R. C., & Fanselow, M. S. (1980). A perceptual-defensive-recuperative model of fear and pain. Behavioral & Brain Sciences, 3, 291–323. Bombace, J. C., Brandon, S. E., & Wagner, A. R. (1991). Modulation of a conditioned eyeblink response by a putative emotive stimulus conditioned with hindleg shock. Journal of Experimental Psychology: Animal Behavior Processes, 17, 323–333. Borkovec, T. D. (1994). The nature, functions, and origins of worry. In G. Davey & F. Tallis (Eds.), Worrying: Perspectives on theory, assessment, and treatment (pp. 5–35). Sussex, England: Wiley. Bouton, M. E. (1984). Differential control by context in the inflation and reinstatement paradigms. Journal of Experimental Psychology: Animal Behavior Processes, 10, 56–74. Bouton, M. E. (1988). Context and ambiguity in the extinction of emotional learning: Implications for exposure therapy. Behaviour Research & Therapy, 26, 137–149. Bouton, M. E. (1991a). Context and retrieval in extinction and in other examples of interference in simple associative learning. In L. Dachowski & C. F. Flaherty (Eds.), Current topics in animal learning: Brain, emotion, and cognition (pp. 25–53). Hillsdale, NJ: Erlbaum. Bouton, M. E. (1991b). A contextual analysis of fear extinction. In P. R. Martin (Ed.), Handbook of behavior therapy and psychological science: An integrative approach (pp. 435–453). New York: Pergamon Press. Bouton, M. E. (1993). Context, time, and memory retrieval in the interference paradigms of Pavlovian learning. Psychological Bulletin, 114, 80–99. Bouton, M. E. (1994a). Conditioning, remembering, and forgetting. Journal of Experimental Psychology: Animal Behavior Processes, 20, 219–231. Bouton, M. E. (1994b). Context, ambiguity, and classical conditioning. Current Directions in Psychological Science, 3, 49–53. Bouton, M. E. (2000). A learning theory perspective on lapse, relapse, and the maintenance of behavior change. Health Psychology, 19 (Suppl.), 57–63. Bouton, M. E., & King, D. A. (1983). Contextual control of the extinction of conditioned fear: Tests for the associative value of the context. Journal of Experimental Psychology: Animal Behavior Processes, 9, 248–265. Bouton, M. E., & King, D. A. (1986). Effect of context on performance to conditioned stimuli with mixed histories of reinforcement and nonreinforcement. Journal of Experimental Psychology: Animal Behavior Processes, 12, 4–15. Bouton, M. E., & Nelson, J. B. (1998a). Mechanisms of feature-positive and feature-negative discrimination learning in an appetitive conditioning paradigm. In N. A. Schmajuk & P. C. Holland (Eds.), Occasion setting: Associative learning and cognition in animals (pp. 69–112). Washington D.C.: American Psychological Association. Bouton, M. E., & Nelson, J. B. (1998b). The role of context in classical conditioning: Some implications for cognitive behavior therapy. In W. T. O’Donohue (Ed.), Learning and behavior therapy (pp. 59–84). Boston: Allyn & Bacon.

A Modern Learning Theory Perspective on the Etiology 311 Bouton, M. E., & Ricker, S. T. (1994). Renewal of extinguished responding in a second context. Animal Learning & Behavior, 22, 317–324. Bouton, M. E., & Swartzentruber, D. (1986). Analysis of the associative and occasion-setting properties of contexts participating in a Pavlovian discrimination. Journal of Experimental Psychology: Animal Behavior Processes, 12, 333–350. Bouton, M. E., & Swartzentruber, D. (1991). Sources of relapse after extinction in Pavlovian and instrumental learning. Clinical Psychology Review, 11, 123–140. Brandon, S. E., Betts, S. L., & Wagner, A. R. (1994). Discriminated lateralized eyeblink conditioning in the rabbit: An experimental context for separating specific and general associative influences. Journal of Experimental Psychology: Animal Behavior Processes, 20, 292–307. Brandon, S. E., Bombace, J. C., Falls, W. A., & Wagner, A. R. (1991). Modulation of unconditioned defensive reflexes by a putative emotive Pavlovian conditioned stimulus. Journal of Experimental Psychology: Animal Behavior Processes, 17, 312–322. Brandon, S. E., & Wagner, A. R. (1991). Modulation of a discrete Pavlovian conditioned reflex by a putative emotive Pavlovian conditioned stimulus. Journal of Experimental Psychology: Animal Behavior Processes, 17, 299–311. Breggin, P. R. (1964). The psychophysiology of anxiety with a review of the literature concerning adrenaline. Journal of Nervous and Mental Disease, 139, 558–568. Bridger, W. H., & Mandel, I. J. (1964). A comparison of GSR fear responses produced by threat and electric shock. Journal of Psychiatric Research, 2, 31–40. Brown, J. S., Kalish, H. I., & Farber, I. E. (1951). Conditioned fear as revealed by magnitude of startle response to an auditory stimulus. Journal of Experimental Psychology, 41, 317–327. Brown, T. A., Chorpita, B. F., & Barlow, D. H. (1998). Structural relationships among dimensions of the DSM-IV anxiety and mood disorders and dimensions of negative affect, positive affect, and autonomic arousal. Journal of Abnormal Psychology, 107, 179–192. Brush, F. R. (1985). Genetic determinants of avoidance learning: Mediation by emotionality? In F. R. Brush & J. B. Overmier (Eds.), Affect, conditioning, and cognition: Essays on the determinants of behavior (pp. 27–42). Hillsdale, NJ: Erlbaum. Carter, M. M., & Barlow, D. H. (1995). Learned alarms: The origins of panic. In W. O’Donohue & L. Krasner (Eds.), Theories of behavior therapy: Exploring behavior change (pp. 209–228). New York: Guilford Press. Carter, M. M., Hollon, S., Carson, R., & Shelton, R. (1995). Effects of a safe person on induced distress following a biological challenge in panic disorder with agoraphobia. Journal of Abnormal Psychology, 104, 156–163. Cepeda-Benito, A., & Short, P. (1997). Morphine’s interoceptive stimuli as cues for the development of associative morphine tolerance in the rat. Psychobiology, 25, 236–240. Chambless, D., Caputo, G., Bright, P., & Gallagher, R. (1984). Assessment of fear of fear in agoraphobics: The Body Sensations Questionnaire and the Agoraphobic Cognitions Questionnaire. Journal of Consulting and Clinical Psychology, 52, 1090–1097. Charney, D. S., Grillon, C., Bremner, J. D. (1998). The neurobiological basis of anxiety and fear: Circuits, mechanisms, and neurochemical interaction, part 1. The Neuroscientist, 4, 35–44. Chorazyna, H. (1962). Some properties of conditioned inhibition. Acta Biologiae Experimentalis, 22, 5–13. Chorpita, B. F., & Barlow, D. H. (1998). The development of anxiety: The role of control in the early environment. Psychological Bulletin, 124, 3–21. Chorpita, B. F., Brown, T., & Barlow, D. (1998). Perceived control as a mediator of family environment in etiological models of childhood anxiety. Behavior Therapy, 29, 457–476. Clark, D. M. (1986). A cognitive approach to panic. Behaviour Research and Therapy, 24, 461–470.

312 Mark E. Bouton, et al. Clark, D. M. (1988). A cognitive model of panic attacks. In S. Rachman & J. D. Maser (Eds.), Panic: Psychological perspectives (pp. 71–89). Hillside, NJ: Erlbaum. Clark, D. M. (1996). Panic disorder: From theory to therapy. In P. M. Salkovskis (Ed.), Frontiers of cognitive therapy (pp. 318–344). New York: Guilford Press. Clark, D. M., & Hemsley, D. (1982). The effects of hyperventilation: Individual variability and its relation to personality. Journal of Behavior Therapy and Experimental Psychiatry, 13, 41–47. Clark, D. M., Salkovskis, P., & Anastasiades, P. (1990, November). Cognitive mediation of lactate induced panic. In R. Rapee (Chair), Experimental investigations of panic disorder. Symposium of the Association for Advancement of Behavior Therapy, San Francisco, CA. Clark, D. M., Salkovskis, P. M., Gelder, M. G., Koehler, C., Martin, M., Anastasiades, P., Hackman, A., Middleton, H., & Jeavons, A. (1988). Tests of a cognitive theory of panic. In I. Hand & H. U. Wittchen (Eds.), Panic and phobias II (pp. 71–90). New York: Springer-Verlag. Clark, D. M., Salkovskis, P. M., Hackmann, A., Middleton, H., Anastasiades, P., & Gelder, M. (1994). A comparison of cognitive therapy, applied relaxation and imipramine in the treatment of panic disorder. British Journal of Psychiatry, 164, 759–769. Clark, L. A., & Watson, D. (1991). Tripartite model of anxiety and depression: Psychometric evidence and taxonomic implications. Special issue: Diagnoses, dimensions, and DSM-IV: The science of classification. Journal of Abnormal Psychology, 100, 316–336. Clark, L. A., Watson, D., & Mineka, S. (1994). Temperament and personality in mood and anxiety disorders. Journal of Abnormal Psychology, 103, 103–116. Clark, R. E., & Squire, L. R. (1998). Classical conditioning and brain systems: The role of awareness. Science, 280, 77–81. Cohen, M. E., & White, P. D. (1947). Studies of breathing, pulmonary ventilation, and subjective awareness of shortness of breath (dyspnea) in neurocirculatory asthenia, effort syndrome, anxiety neurosis. Journal of Clinical Investigation, 26, 520. Cohen, M. E., & White, P. D. (1950). Life situations, emotions, and neurocirculatory asthenia (anxiety neurosis, neurasthenia, effort syndrome). In H. G. Wolff (Ed.), Life stress and bodily disease (Nervous and Mental Disease, Research Publication No. 29). Baltimore, MD: Williams & Wilkins. Colwill, R. M. (1994). Associative representations of instrumental contingencies. In D. L. Medin (Ed.), The psychology of learning and motivation: Advances in research and theory. (Vol. 31, pp. 1–72). San Diego, CA: Academic Press. Colwill, R. M., & Motzkin, D. K. (1994). Encoding of the unconditioned stimulus in Pavlovian conditioning. Animal Learning & Behavior, 22, 384–394. Colwill, R. M., & Rescorla, R. A. (1988). Associations between the discriminative stimulus and the reinforcer in instrumental learning. Journal of Experimental Psychology: Animal Behavior Processes, 14, 155–164. Cook, E. W., Hawk, L. W., Davis, T. L., & Stevenson, V. E. (1991). Affective individual differences and startle reflex modulation. Journal of Abnormal Psychology, 100, 5–13. Cook, E. W., Hodes, R. L., & Lang, P. J. (1986). Preparedness and phobia: Effects of stimulus content on human visceral conditioning. Journal of Abnormal Psychology, 95, 195–207. Copian, J. D., Andrews, M. W., Rosenblum, L. A., Owens, M. J., Friedman, S., Gorman, J. M., & Nemeroff, C. B. (1996). Persistent elevations of cerebrospinal fluid concentrations of corticotropin-releasing factor in adult nonhuman primates exposed to early-life stressors: Implications for the pathophysiology of mood and anxiety disorders. Proceedings of the National Academy of Sciences, 93, 1619–1623. Craske, M. G. (1999). Anxiety disorders: Psychological approaches to theory and treatment. Boulder, CO: Westview Press.

A Modern Learning Theory Perspective on the Etiology 313 Craske, M. G., Glover, D., & DeCola, J. (1995). Predicted versus unpredicted panic attacks: Acute versus general distress. Journal of Abnormal Psychology, 104, 214–223. Craske, M. G., Miller, P. P., Rotunda, R., & Barlow, D. H. (1990). A descriptive report of features of initial unexpected panic attacks in minimal and extensive avoiders. Behaviour Research & Therapy, 28, 395–400. Crowell, C. R. (1974). Conditioned-aversive aspects of electric shock. Learning and Motivation, 5, 209–220. Cunningham, C. L. (1998). Drug conditioning and drug-seeking behavior. In W. T. O’Donohue (Ed.), Learning and behavior therapy (pp. 518–544). Boston: Allyn & Bacon. Davey, G., de Jong, P., & Tallis, F. (1993). UCS inflation in the etiology of a variety of anxiety disorders: Some case histories. Behaviour Research and Therapy, 31, 495–498. Davis, M. (1992). The role of the amygdala in conditioned fear. In J. P. Aggleton (Ed.), The amygdala: Neurobiological aspects of emotion, memory, and mental dysfunction (pp. 255–305). New York: Wiley-Liss. Davis, M., Walker, D. L., & Yee, Y. (1997). Amygdala and bed nucleus of the stria terminalis: Differential roles in fear and anxiety measured with the acoustic startle reflex. In L. Squire & D. Schacter (Eds.), Biological and psychological perspectives on memory and memory disorders (pp. 305–331). Washington D.C.: American Psychiatric Association. Delamater, A. R. (1996). Effects of several extinction treatments upon the integrity of Pavlovian stimulus-outcome associations. Animal Learning & Behavior, 24, 437–449. Dimberg, U. (1987). Facial reactions, autonomic activity, and experienced emotion: A three-component model of emotional conditioning. Biological Psychology, 24, 105–122. Domjan, M. (1994). Formulation of a behavior system for sexual conditioning. Psychonomic Bulletin & Review, 1, 421–428. Domjan, M. (1997). Behavior systems and the demise of equipotentiality: Historical antecedents and evidence from sexual conditioning. In M. E. Bouton & M. S. Fanselow (Eds.), Learning, motivation, and cognition: The functional behaviorism of Robert C. Bolles (pp. 31–51). Washington D.C.: American Psychological Association. Domjan, M. (1998). The principles of learning and behavior (4th ed.). Pacific Grove, CA: Brooks/Cole. Domjan, M., Huber-McDonald, M., & Holloway, K. S. (1992). Conditioning copulatory behavior to an artificial object: Efficacy of stimulus fading. Animal Learning & Behavior, 20, 350–362. Dworkin, B. R. (1993). Learning and physiological regulation. Chicago: University of Chicago Press. Ehlers, A. (1993). Somatic symptoms and panic attacks: A retrospective study of learning experiences. Behaviour Research and Therapy, 31, 269–278. Ehlers, A. (1995). A 1-year prospective study of panic attacks: Clinical course and factors associated with maintenance. Journal of Abnormal Psychology, 104, 164–172. Ehlers, A., & Breuer, P. (1992). Increased cardiac awareness in panic disorder. Journal of Abnormal Psychology, 101, 371–382. Ehlers, A., & Breuer, P. (1996). How good are patients with panic disorder at perceiving their heartbeats? Biological Psychology, 42, 165–182. Ehlers, A., Margraf, J., & Chambless, D. (1992). Fragebögen zu körperbezogenen Ängsten, Kognitionen and Vermeidung. Weinheim: Beltz. Ehlers, A., Margraf, J., Roth, W. T., Taylor, C. B., & Birbaumer, N. (1988). Anxiety induced by false heart rate feedback in patients with panic disorder. Behaviour Research and Therapy, 26, 1–11. Eich, E. (1995). Searching for mood dependent memory. Psychological Science, 6, 67–75.

314 Mark E. Bouton, et al. Eikelboom, R., & Stewart, J. (1982). Conditioning of drug-induced physiological responses. Psychological Review, 89, 507–528. Eysenck, H. J. (Ed.). (1960). Behavior therapy and the neuroses. Oxford, England: Pergamon Press. Eysenck, H. J. (1967). The biological basis of personality. Springfield, IL: Charles C. Thomas. Eysenck, H. J. (1979). The conditioning model of neurosis. Behavioral and Brain Sciences, 2, 155–199. Eysenck, H. J., & Rachman, S. (1965). The causes and cures of neurosis. London: Routledge & Kegan Paul. Fanselow, M. S. (1980). Conditional and unconditional components of postshock freezing. Pavlovian Journal of Biological Science, 15, 177–182. Fanselow, M. S. (1989). The adaptive function of conditioned defensive behavior: An ecological approach to Pavlovian stimulus-substitution theory. In R. J. Blanchard, P. F. Brain, D. C. Blanchard, & S. Parmigiani (Eds.), Ethoexperimental approaches to the study of behavior (pp. 151–166). Dordrecht, the Netherlands: Kluwer Academic. Fanselow, M. S. (1994). Neural organization of the defensive behavior system responsible for fear. Psychonomic Bulletin & Review, 1, 429–438. Fanselow, M. S., & Lester, L. S. (1988). A functional behavioristic approach to aversively motivated behavior: Predatory imminence as a determinant of the topography of defensive behavior. In R. C. Bolles & M. D. Beecher (Eds.), Evolution and learning (pp. 185–212). Hillsdale, NJ: Erlbaum. Feingold, A. (1994). Gender differences in personality: A meta-analysis. Psychological Bulletin, 116, 429–456. Forsyth, J. P., & Eifert, G. H. (1996). Systemic alarms in fear conditioning I: A reappraisal of what is being conditioned. Behavior Therapy, 27, 441–462. Forsyth, J. P., & Eifert, G. H. (1998). Response intensity in content-specific fear conditioning comparing 20% versus 13% CO2-enriched air as unconditioned stimuli. Journal of Abnormal Psychology, 107, 291–304. Forsyth, J. P., Eifert, G. H., & Thompson, R. N. (1996). Systemic alarms in fear conditioning: II. An experimental methodology using 20% carbon dioxide inhalation as an unconditioned stimulus. Behavior Therapy, 27, 391–415. Gallup, G. G., Jr., & Maser, J. D. (1977). Tonic immobility: Evolutionary underpinnings of human catalepsy and catatonia. In J. D. Maser & M. E. P. Seligman (Eds.), Psychopathology: Experimental models (pp. 334–357). San Francisco: Freeman. Goddard, M. J. (1996). Effect of US signal value on blocking of a CS-US association. Journal of Experimental Psychology: Animal Behavior Processes, 22, 258–264. Goddard, M. J. (1997). Spontaneous recovery in US extinction. Learning & Motivation, 28, 118–128. Goddard, M. J. (1999). The role of US signal value in contingency, drug conditioning, and learned helplessness. Psychonomic Bulletin & Review, 6, 412–423. Goddard, M. J., & Jenkins, H. M. (1988). Blocking of a CS-US association by a US-US association. Journal of Experimental Psychology: Animal Behavior Processes, 14, 177–186. Goldstein, A. J., & Chambless, D. L. (1978). A reanalysis of agoraphobia. Behavior Therapy, 9, 47–59. Gray, J. A., & McNaughton, N. (1996). The neuropsychology of anxiety. Reprise. In D. A. Hope (Ed.), Perspectives on anxiety, panic, and fear (The 43rd Annual Nebraska Symposium on Motivation) (pp. 61–134). Lincoln: Nebraska University Press. Greeley, J., Lê, D. A., Poulos, C. X., & Cappell, H. (1984). Alcohol is an effective cue in the conditional control of tolerance to alcohol. Psychopharmacology, 83, 159–162. Grillon, C., Ameli, R., Goddard, A., Woods, S. W., & Davis, M. (1994). Baseline and fear-potentiated startle in panic disorder patients. Biological Psychiatry, 35, 431–439.

A Modern Learning Theory Perspective on the Etiology 315 Grillon, C., Ameli, R., Woods, S. W., Merikangas, K., & Davis, M. (1991). Fear-potentiated startle in humans: Effects of anticipatory anxiety on the acoustic blink reflex. Psychophysiology, 28, 588–595. Grings, W. W., Schell, A. M., & Carey, C. A. (1973). Verbal control of an autonomic response in a cue reversal situation. Journal of Experimental Psychology, 99, 215–221. Hall, G., & Honey, R. C. (1989). Contextual effects in conditioning, latent inhibition, and habituation: Associative and retrieval functions of contextual cues. Journal of Experimental Psychology: Animal Behavior Processes, 15, 232–241. Hamm, A. O., & Vaiti, D. (1996). Affective learning: Awareness and aversion. Psychophysiology, 33, 698–710. Hannum, R., Rosellini, R., & Seligman, M. (1976). Retention of learned helplessness and immunization in the rat from weaning to adulthood. Developmental Psychology, 12, 449–454. Hayward, C., Killen, J. D., Kraemer, H. C., & Taylor, C. B. (2000). Predictors of panic attacks in adolescence. Journal of the American Academy of Child and Adolescent Psychiatry, 39, 207–214. Heim, C., & Nemeroff, C. B. (1999). The impact of early adverse experiences on brain systems involved in the pathophysiology of anxiety and affective disorders. Biological Psychiatry, 46, 1509–1522. Heller, W., Nitschke, J. B., Etienne, M. A., & Miller, G. A. (1997). Patterns of regional brain activity differentiate types of anxiety. Journal of Abnormal Psychology, 106, 376–385. Hendersen, R. W. (1975). Compounds of conditioned fear stimuli. Learning & Motivation, 6, 28–42. Hendersen, R. W. (1985). Fearful memories: The motivational significance of forgetting. In F. R. Brush & J. B. Overmier (Eds.), Affect, conditioning and cognition: Essays on the determinants of behavior (pp. 43–52). Hillsdale, NJ: Erlbaum. Hirsch, S. M., & Bolles, R. C. (1980). On the ability of prey to recognize predators. Zeitschrift für Tierpsychologie, 54, 71–84. Hoehn-Saric, R., McLeod, D. R., & Zimmerli, W. D. (1989). Somatic manifestations in women with generalized anxiety disorder: Psycho-physiological responses to psychological stress. Archives of General Psychiatry, 46, 1113–1119. Holland, P. C. (1977). Conditioned stimulus as a determinant of the form of the Pavlovian conditioned response. Journal of Experimental Psychology: Animal Behavior Processes, 3, 77–104. Holland, P. C. (1989). Feature extinction enhances transfer of occasion setting. Animal Learning & Behavior, 17, 269–279. Holland, P. C. (1992). Occasion setting in Pavlovian conditioning. In G. Bower (Ed.), The psychology of learning and motivation (Vol. 28, pp. 69–125). Orlando, FL: Academic Press. Hollis, K. L. (1982). Pavlovian conditioning of signal-centered action patterns and autonomic behavior: A biological analysis of function. Advances in the Study of Behavior, 12, 131–142. Hollis, K. L. (1997). Contemporary research on Pavlovian conditioning: A “new” functional analysis. American Psychologist, 52, 956–965. Joffe, J., Rawson, R., & Mulick, J. (1973). Control of their environment reduces emotionality in rats. Science, 180, 1383–1384. Joiner, T. E., Jr., Steer, R. A., Beck, A. T., Schmidt, N. B., Rudd, M. D., & Catanzaro, S. J. (1999). Physiological hyperarousal: Construct validity of a central aspect of the tripartite model of depression and anxiety. Journal of Abnormal Psychology, 108, 290–298. Kapp, B. S., Whalen, P. J., Supple, W. F., Jr., & Pascoe, J. P. (1992). Amygdaloid contributions to conditioned arousal and sensory information processing. In J. P. Aggleton (Ed.), The amygdala: Neurobiological aspects of emotion, memory, and mental dysfunction (pp. 229–254). New York: Wiley.

316 Mark E. Bouton, et al. Kenardy, J., Fried, L., Kraemer, H. C., & Taylor, C. B. (1992). Psychological precursors of panic attacks. British Journal of Psychiatry, 160, 668–673. Kenardy, J., & Taylor, C. B. (1999). Expected versus unexpected panic attacks: A naturalistic prospective study. Journal of Anxiety Disorders, 13, 435–445. Kendler, K. S., Kessler, R. C., Walters, E. D., MacLean, C., Neale, M. C., Heath, A. C., & Eaves, L. J. (1995). Stressful life events, genetic liability, and onset of an episode of major depression in women. American Journal of Psychiatry, 152, 833–842. Kendler, K. S., Neale, M., Kessler, R., Heath, A., & Eaves, L. (1992). The genetic epidemiology of phobias in women: The interrelationship of agoraphobia, social phobia, situational phobia, and simple phobia. Archives of General Psychiatry, 49, 273–281. Kendler, K. S., Walters, E. E., Neale, M. C., Kessler, R. C., Heath, A. C., & Eaves, L. J. (1995). The structure of the genetic and environmental risk factors for six major psychiatric disorders in women: Phobia, generalized anxiety disorder, panic disorder, bulimia, major depression, and alcoholism. Archives of General Psychiatry, 52, 374–382. Kessler, R. C., McGonagle, K. A., Zhao, S., Nelson, C. B., Hughes, M., Eshleman, S., Wittchen, H.-U., & Kendler, K. S. (1994). Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States: Results from the National Comorbidity Survey. Archives of General Psychiatry, 51, 8–19. Kim, J. A., Siegel, S., & Patenall, V. R. A. (1999). Drug-onset cues as signals: Intraadministration associations and tolerance. Journal of Experimental Psychology: Animal Behavior Processes, 25, 491–504. King, N. J., Gullone, E., Tonge, B. J., & Ollendick, T. H. (1993). Self-reports of panic attacks and manifest anxiety in adolescents. Behaviour Research and Therapy, 31, 111–116. Klein, D. F. (1981). Anxiety reconceptualized. In D. F. Klein & J. Rabkin (Eds.), Anxiety: New research and changing concepts (pp. 235–264). New York: Raven Press. Konorski, J. (1967). Integrative activity of the brain. Chicago: University of Chicago Press. Kruse, J. M., Overmier, J. B., Konz, W. A., & Rokke, E. (1983). Pavlovian conditioned stimulus effects upon instrumental choice behavior are reinforcer specific. Learning and Motivation, 14, 165–181. Kushner, M. G., & Beitman, B. D. (1990). Panic attacks without fear: An overview. Behaviour Research and Therapy, 28, 469–479. Ladd, C. O., Huot, R. L., Thrivikraman, K. V., Nemeroff, C. B., Meaney, M. J., & Plotsky, P. M. (2000). Long-term behavioral and neuroendocrine adaptations to adverse early experience. In E. A. Mayer & C. B. Saper (Eds.), Progress in brain research: The biological basis for mind body interactions (Vol. 122, pp. 81–103). Amsterdam: Elsevier. Lang, P. J. (1994). The varieties of emotional experience: A meditation on James-Lange theory. Psychological Review, 101, 211–221. Lang, P. J. (1995). The emotion probe: Studies of motivation and attention. American Psychologist, 50, 372–385. Lang, P. J., Bradley, M. M., & Cuthbert, B. N. (1990). Emotion, attention, and the startle reflex. Psychological Review, 97, 377–395. Lang, P. J., Davis, M., & Öhman, A. (2000). Fear and anxiety: Animal models and human cognitive psychophysiology. Journal of Affective Disorders, 61, 137–159. Last, C. G., O’Brien, G. T., & Barlow, D. H. (1985). The relationship between cognitions and anxiety. Behavior Modification, 9, 235–241. LeDoux, J. E. (1996). The emotional brain: The mysterious underpinnings of emotional life. New York: Simon & Schuster. Leventhal, H. (1982). The integration of emotion and cognition: A view from the perceptual-motor theory of emotion. In M. Clark & S. Fiske (Eds.), Affect and Cognition: The 17th Annual Carnegie Symposium on Cognition. Hilldale, NJ: Erlbaum.

A Modern Learning Theory Perspective on the Etiology 317 Leventhal, H., Brown, D., Shacham, S., & Engquist, G. (1979). Effects of preparatory information about sensations, threat of pain, and attention on cold pressor distress. Journal of Personality and Social Psychology, 37, 688–714. Levey, A., & Martin, I. (1981). Personality and conditioning. In H. Eysenck (Ed.), A model for personality (pp. 123–168). Berlin: Springer-Verlag. Levy, E. (1998). Learning experiences and vulnerability to panic disorder: Possible origins of high anxiety sensitivity and acquisition of catastrophic misinterpretation of sensations. Unpublished doctoral dissertation, Northwestern University. Liebowitz, M. R., Fyer, A. J., Gorman, J. M., Dillon, D., Appleby, I. L., Levy, G., Anderson, S., Levitt, M., Palij, M., Davies, S. O., & Klein, D. F. (1984). Lactate provocation of panic. Archives of General Psychiatry, 41, 764–770. Liebowitz, M. R., Gorman, J. M., Fyer, A. J., Levitt, M., Dillon, D., Levy, G., Appleby, I. L., Anderson, S., Palij, M., Davies, S. O., & Klein, D. F. (1985). Lactate provocation of panic attacks: II. Biochemical and physiological findings. Archives of General Psychiatry, 42, 709–719. Lindemann, E., & Finesinger, J. E. (1938). The effect of adrenaline and mecholyl in states of anxiety in psychoneurotic patients. American Journal of Psychiatry, 95, 353–370. Lovibond, P. F. (1993). Conditioning and cognitive-behaviour therapy. Behaviour Change, 10, 119–130. Lovibond, P. F., Preston, G. C., & Mackintosh, N. J. (1984). Context specificity of conditioning, extinction, and latent inhibition. Journal of Experimental Psychology: Animal Behavior Processes, 10, 360–375. Mackintosh, N. J. (1974). The psychology of animal learning. San Diego, CA: Academic Press. Mackintosh, N. J. (1975). A theory of attention: Variations in the associability of stimuli with reinforcement. Psychological Review, 82, 276–298. Maier, S. F. (1990). Role of fear in mediating shuttle escape learning deficit produced by inescapable shock. Journal of Experimental Psychology: Animal Behavior Processes, 16, 137–149. Maier, S. F., & Seligman, M. E. P. (1976). Learned helplessness: Theory and evidence. Journal of Experimental Psychology: General, 105, 3–46. Margraf, J., Ehlers, A., & Roth, W. T. (1986). Sodium lactate infusions and panic attacks: A review and critique. Psychosomatic Medicine, 48, 23–51. Margraf, J., & Schneider, S. (1995, July). Psychological treatment of panic: What works in the long run? Paper presented at the World Congress of Behavioural and Cognitive Therapies, Copenhagen, Denmark. Marks, I. (1969). Fears and phobias. London: Heineman Medical Books. Maser, J. D., & Gallup, G. G. (1974). Tonic immobility in the chicken: Catalepsy potentiation by uncontrollable shock and alleviation by imipramine. Psychosomatic Medicine, 36, 199–205. McGuffin, P., & Reich, T. (1984). Psychopathology and genetics. In H. Adams & P. Sutker (Eds.), Comprehensive handbook of psychopathology. New York: Plenum. McNally, R. J. (1990). Psychological approaches to panic disorder: A review. Psychological Bulletin, 108, 403–419. McNally, R. J. (1994). Panic disorder: A critical analysis. New York: Guilford Press. McNally, R. J. (1999). Theoretical approaches to the fear of anxiety. In S. Taylor (Ed.), Anxiety sensitivity: Theory, research, and treatment of the fear of anxiety. Hillsdale, NJ: Erlbaum. McNally, R. J., & Eke, M. (1996). Anxiety sensitivity, suffocation fear, and breath-holding duration as predictors of response to carbon dioxide challenge. Journal of Abnormal Psychology, 105, 146–149.

318 Mark E. Bouton, et al. McNaughton, N., & Gray, J. A. (in press). Anxiolytic action on the behavioural inhibition system implies multiple types of arousal contribute to anxiety. Journal of Affective Disorders, 61, 161–176. McNish, K. A., Betts, S. L., Brandon, S. E., & Wagner, A. R. (1997). Divergence of conditioned eyeblink and conditioned fear in backward Pavlovian training. Animal Learning & Behavior, 25, 43–52. Menzies, R., & Clarke, J. (1995). The etiology of phobias: A nonassociative account. Clinical Psychology Review, 15, 23–48. Merckelbach, H., de Ruiter, C., van den Hout, M., & Hoekstra, R. (1989). Conditioning experiences and phobias. Behaviour Research and Therapy, 27, 657–662. Mineka, S. (1985a). Animal models of anxiety-based disorders: Their usefulness and limitations. In A. H. Tuma & J. Maser (Eds.), Anxiety and the anxiety disorders (pp. 199–244). Hillsdale, NJ: Erlbaum. Mineka, S. (1985b). The frightful complexity of the origins of fears. In F. R. Brush & J. B. Overmier (Eds.), Affect, conditioning, and cognition: Essays on the determinants of behavior (pp. 55–73). Hillsdale, NJ: Erlbaum. Mineka, S., & Cook, M. (1993). Mechanisms underlying observational conditioning of fear in monkeys. Journal of Experimental Psychology: General, 122, 23–38. Mineka, S., Cook, M., & Miller, S. (1984). Fear conditioned with escapable and inescapable shock: Effects of a feedback stimulus. Journal of Experimental Psychology: Animal Behavior Processes, 10, 307–323. Mineka, S., Gunnar, M., & Champoux, M. (1986). Control and early socioemotional development: Infant rhesus monkeys reared in controllable versus uncontrollable environments. Child Development, 57, 1241–1256. Mineka, S., & Hendersen, R. (1985). Controllability and predictability in acquired motivation. Annual Review of Psychology, 36, 495–530. Mineka, S., & Kelly, K. (1989). The relationship between anxiety, lack of control, and loss of control. In A. Steptoe (Ed.), Stress, personal control and health (pp. 163–191). New York: Wiley. Mineka, S., & Kihlstrom, J. F. (1978). Unpredictable and uncontrollable events: A new perspective on experimental neurosis. Journal of Abnormal Psychology, 87, 256–271. Mineka, S., Mystkowski, J., Hladek, D., & Rodriguez, B. (1999). The effects of context on return of fear following successful exposure treatment for spider phobia. Journal of Consulting and Clinical Psychology, 67, 599–604. Mineka, S., Watson, D., & Clark, L. A. (1998). Comorbidity of anxiety and unipolar mood disorders. Annual Review of Psychology, 49, 377–412. Mineka, S., & Zinbarg, R. (1991). Animal models of psychopathology. In C. E. Walker (Ed.), Clinical psychology: Historical and research foundations (pp. 51–86). New York: Plenum. Mineka, S., & Zinbarg, R. (1995). Conditioning and ethological models of social phobia. In R. Heimberg, M. Liebowitz, D. Hope, & F. Schneier (Eds.), Social phobia: Diagnosis, assessment and treatment (pp. 1134–1162). New York: Guilford Press. Mineka, S., & Zinbarg, R. (1996). Conditioning and ethological models of anxiety disorders: Stress-in-dynamic-context anxiety models. In D. A. Hope (Ed.), Perspectives on anxiety, panic, and fear. Nebraska Symposium on Motivation (Vol. 43, pp. 135–210). Lincoln: Nebraska University Press. Mineka, S., & Zinbarg, R. (1998). Experimental approaches to understanding the mood and anxiety disorders. In J. Adair (Ed.), Advances in psychological research, volume 2 (pp. 429–454). Hove, England: Psychology Press. Minor, T., Dess, N., & Overmier, J. B. (1991). Inverting the traditional view of “learned helplessness.” In R. Denny (Ed.), Fear, avoidance, and phobias (pp. 87–133). Hillsdale, NJ: Erlbaum.

A Modern Learning Theory Perspective on the Etiology 319 Monroe, S. M., & Simons, A. E. (1991). Diathesis-stress theories in the context. of life stress research: Implications for the depressive disorders. Psychological Bulletin, 110, 406–425. Morgan, C. A., Grillon, C., Southwick, S. M., Davis, M., & Charney, D. (1995). Fear-potentiated startle in posttraumatic stress disorder. Biological Psychiatry, 36, 378–385. Mowrer, O. H. (1947). On the dual nature of learning—A re-interpretation of “conditioning” and “problem-solving.” Harvard Educational Review, 17, 102–148. Mowrer, O. H. (1960). Learning theory and behavior. New York: Wiley. Mowrer, O. H., & Viek, P. (1948). An experimental analogue of fear from a sense of helplessness. Journal of Abnormal Social Psychology, 83, 193–200. Norton, G. R., Cox, B. J., & Malan, J. (1992). Nonclinical panickers: A critical review. Clinical Psychology Review, 12, 121–139. Norton, G. R., Dorward, J., & Cox, B. J. (1986). Factors associated with panic attack in non-clinical subjects. Behavior Therapy, 17, 239–252. Öhman, A. (1996). Preferential pre-attentive processing of threat in anxiety: Preparedness and attentional biases. In R. Rapee (Ed.), Current controversies in the anxiety disorders (pp. 253–290). New York: Guilford Press. Öhman, A. (1997). Unconscious pre-attentive mechanisms in the activation of phobic fear. In G. C. L. Davey (Ed.), Phobias: A handbook of theory, research, and treatment (pp. 349–374). Chichester: Wiley. Öhman, A., Flykt, A., & Lundqvist, D. (2000). Unconscious emotion: Evolutionary perspectives, psychophysiological data, and neuropsychological mechanisms. In R. Lane & L. Nadel (Eds.), The Cognitive neuroscience of emotion (pp. 296–327). New York: Oxford University Press. Öhman, A., & Mineka, S. (2001). Revisiting preparedness: Toward an evolved module of fear learning. Psychological Review. Manuscript submitted for publication. Öst, L.-G., & Hugdahl, K. (1981). Acquisition of phobias and anxiety response patterns in clinical patients. Behaviour Research and Therapy, 19, 439–447. Öst, L.-G., & Hugdahl, K. (1983). Acquisition of agoraphobia, mode of onset, and anxiety response patterns. Behaviour Research and Therapy, 19, 439–447. Overmier, J. B. (1985). Toward a reanalysis of the causal structure of the learned helplessness syndrome. In F. R. Brush & J. B. Overmier (Eds.). Affect, conditioning, and cognition: Essays on the determinants of behavior (pp. 211–227). Hillsdale, NJ: Erlbaum. Overmier, J. B., & Lawry, J. A. (1979). Pavlovian conditioning and the mediation of behavior. In G. H. Bower (Ed.). The psychology of learning and motivation (Vol. 13, pp. 1–55). New York: Academic Press. Parker, G. (1983). Parental overprotection: A risk factor in psychosocial development. New York: Grune & Stratton. Pavlov, I. P. (1927). Conditioned reflexes. London: Oxford University Press. Pearce, J. M., Aydin, A., & Redhead, E. S. (1997). Configural analysis of summation in autoshaping. Journal of Experimental Psychology: Animal Behavior Processes, 23, 84–94. Pearce, J. M., & Hall, G. (1980). A model for Pavlovian learning: Variations in the effectiveness of conditioned but not of unconditioned stimuli. Psychological Review, 87, 532–552. Peterson, C., Maier, S. F., & Seligman, M. E. P. (1993). Learned helplessness: A theory for the age of personal control. New York: Oxford University Press. Plomin, R., DeFries, J., McClearn, G., & Rutter, M. (1997). Behavioral genetics, 3rd edition. New York: Freeman. Powley, T. L. (1977). The ventromedial hypothalamic syndrome, satiety, and a cephalic phase hypothesis. Psychological Review, 84, 89–126. Rachman, S. (1977). The conditioning theory of fear-acquisition: A critical examination. Behaviour Research and Therapy, 15, 375–387. Rachman, S. (1990). Fear and courage. New York: Freeman.

320 Mark E. Bouton, et al. Rachman, S. (1991). Neo-conditioning and the classical theory of fear acquisition. Clinical Psychological Review, 11, 155–173. Rachman, S., Lopatka, C., & Levitt, K. (1988). Experimental Analyses of panic: II. Panic patients. Behaviour Research and Therapy, 26, 33–40. Rachman, S. J., & Maser, J. (1988). Panic and cognition. Hillsdale, NJ: Erlbaum. Ramsay, D. S., & Woods, S. C. (1997). Biological consequences of drug administration: Implications for acute and chronic tolerance. Psychological Review, 104, 170–193. Rapee, R., Mattick, R., & Murrell, E. (1986). Cognitive mediation in the affective component of spontaneous panic attacks. Journal of Behavior Therapy and Experimental Psychiatry, 17, 245–253. Razran, G. (1961). The observable unconscious and the inferable conscious in current Soviet psychophysiology: Interoceptive conditioning, semantic conditioning, and the orienting reflex. Psychological Review, 68, 81–147. Reberg, D. (1972). Compound tests for excitation in early acquisition and after prolonged extinction of conditioned suppression. Learning & Motivation, 3, 246–258. Reiss, S. (1987). Theoretical perspectives on the fear of anxiety. Clinical Psychology Review, 7, 585–596. Reiss, S. (1991). Expectancy model of fear, anxiety, and panic. Clinical Psychology Review, 11, 141–153. Reiss, S., Peterson, R. A., Gursky, D. M., & McNally, R. J. (1986). Anxiety sensitivity, anxiety frequency, and the prediction of fearfulness. Behaviour Research and Therapy, 24, 1–8. Rescorla, R. A. (1974). Effect of inflation of the unconditioned stimulus value following conditioning. Journal of Comparative and Physiological Psychology, 86, 101–106. Rescorla, R. A. (1985). Conditioned inhibition and facilitation. In R. R. Miller & N. E. Spear (Eds.), Information processing in animals: Conditioned inhibition (pp. 299–326). Hillsdale, NJ: Erlbaum. Rescorla, R. A. (1986). Extinction of facilitation. Journal of Experimental Psychology: Animal Behavior Processes, 12, 16–24. Rescorla, R. A. (2000). Extinction can be enhanced by a concurrent excitor. Journal of Experimental Psychology: Animal Behavior Processes, 25, 251–260. Rescorla, R. A., & Coldwell, S. E. (1995). Summation in autoshaping. Animal Learning & Behavior, 23, 314–326. Rescorla, R. A., & Furrow, D. R. (1977). Stimulus similarity as a determinant of Pavlovian conditioning. Journal of Experimental Psychology: Animal Behavior Processes, 3, 203–215. Rescorla, R. A., & Gillan, D. J. (1980). An analysis of the facilitative effect of similarity on second-order conditioning. Journal of Experimental Psychology: Animal Behavior Processes, 6, 339–351. Rescorla, R. A., & Holland, P. C. (1977). Associations in Pavlovian conditioned inhibition. Learning & Motivation, 8, 429–447. Rescorla, R. A., & Solomon, R. L. (1967). Two-process learning theory: Relationships between Pavlovian conditioning and instrumental learning. Psychological Review, 74, 151–182. Rescorla, R. A., & Wagner, A. R. (1972). A theory of Pavlovian conditioning: Variations in the effectiveness of reinforcement and nonreinforcement. In A. H. Black & W. F. Prokasy (Eds.), Classical conditioning: II. Current research and theory (pp. 64–99). New York: Appleton-Century-Crofts. Rosen, J. B., & Schulkin, J. (1998). From normal fear to pathological anxiety. Psychological Review, 105, 325–350. Salkovskis, P.M. (1988). Phenomenology, assessment, and the cognitive model of panic. In S. M. J. D. Rachman (Ed.), Panic: Psychological perspectives (pp. 111–136). Hillsdale, NJ: Erlbaum.

A Modern Learning Theory Perspective on the Etiology 321 Salkovskis, P.M., Clark, D. M., & Gelder, M. G. (1996). Cognition-behaviour links in the persistence of panic. Behaviour Research & Therapy, 34, 453–458. Sanderson, W., Rapee, R., & Barlow, D. (1989). The influence of an illusion of control on panic attacks induced via inhalation of 5.5% carbon dioxide-enriched air. Archives of General Psychiatry, 46, 157–162. Schmajuk, N. A., & Holland, P. C. (Eds.). (1998). Occasion setting: Associative learning and cognition in animals. Washington D.C.: American Psychological Association. Schmidt, N. B., Lerew, D. R., & Jackson, R. J. (1997). The role of anxiety sensitivity in the pathogenesis of panic: Projective evaluation of spontaneous panic attacks during acute stress. Journal of Abnormal Psychology, 106, 355–364. Schmidt, N. B., Lerew, D. R., & Jackson, R. J. (1999). Prospective evaluation of anxiety sensitivity in the pathogenesis of panic: Replication and extension. Journal of Abnormal Psychology, 108, 532–537. Schneewind, K. (1995). Impact of family processes on control beliefs. In A. Bandura (Ed.), Self-efficacy in changing societies (pp. 114–148). New York: Cambridge Press. Seligman, M. E. P. (1988). Competing theories of panic. In S. Rachman & J. D. Maser (Eds.), Panic: Psychological perspectives (pp. 321–329). Hillsdale, NJ: Erlbaum. Seligman, M. E. P., & Binik, Y. (1977). The safety signal hypothesis. In H. Davis & H. M. B. Hurwitz (Eds.), Operant-Pavlovian interactions (pp. 165–188). Hillsdale, NJ: Erlbaum. Seligman, M. E. P., Maier, S., & Solomon, R. (1971). Unpredictable and uncontrollable aversive events. In F. Brush (Ed.), Aversive conditioning and learning (pp. 347–400). New York: Academic Press. Sheehan, D. (1983). The anxiety disease. New York: Bantam Books. Siegel, S. (1975). Evidence from rats that morphine tolerance is a learned response. Journal of Comparative and Physiological Psychology, 89, 498–506. Siegel, S. (1988). State dependent learning and morphine tolerance. Behavioral Neuroscience, 102, 228–232. Siegel, S. (1989). Pharmacological conditioning and drug effects. In A. J. Goudie & M. W. Emmett-Oglesby (Eds.), Psychoactive drugs: Tolerance and sensitization (pp. 115–180). Clifton, NJ: Humana Press. Silva, F. J., Timberlake, W., & Koehler, T. L. (1996). A behavior systems approach to bidirectional excitatory serial conditioning. Learning and Motivation, 27, 130–150. Solomon, R. L., Kamin, L. J., & Wynne, L. C. (1953). Traumatic avoidance learning: The outcomes of several extinction procedures with dogs. Journal of Abnormal and Social Psychology, 48, 291–302. Soltysik, S. S. (1960). Studies on avoidance conditioning: III. Alimentary conditioned reflex model of the avoidance reflex. Acta Biologiae Experimentalis, 20, 183–191. Soltysik, S. S., Wolfe, G. E., Nicholas, T., Wilson, W. J., & Garcia-Sanchez, L. (1983). Blocking of inhibitory conditioning within a serial conditioned stimulus-conditioned inhibitor compound: Maintenance of acquired behavior without an unconditioned stimulus. Learning & Motivation, 14, 1–29. Spence, J. T., & Spence, K. (1966). The motivational components of manifest anxiety: Drive and drive stimuli. In C. Spielberger (Ed.), Anxiety and behavior (pp. 291–326.) New York: Academic Press. Stegen, K., De Bruyne, K., Rasschaert, W., Van de Woestijne, K., & Van den Bergh, O. (1999). Fear-relevant images as conditioned stimuli for somatic complaints, respiratory behavior, and reduced end-tidal pCO2. Journal of Abnormal Psychology, 108, 143–152. Swartzentruber, D. (1995). Modulatory mechanisms in Pavlovian conditioning. Animal Learning & Behavior, 23, 123–143.

322 Mark E. Bouton, et al. Taylor, S., & Rachman, S. (1994). Klein’s suffocation theory of panic. Archives of General Psychiatry, 51, 505–506. Teasdale, J. (1988). Cognitive models and treatments for panic. A critical evaluation. In S. Rachman & J. D. Maser (Eds.), Panic: Psychological perspectives (pp. 189–203). Hillsdale, NJ: Erlbaum. Telch, M. J. (1995, July). Singular and combined efficacy of in vivo exposure and CBT in the treatment of panic disorder with agoraphobia. Paper presented at the World Congress of Behavioural and Cognitive Therapies, Copenhagen, Denmark. Telch, M. J., Brouillard, M., Telch, C. F., Agras, W. S., & Taylor, C. B. (1989). Role of cognitive appraisal in panic related avoidance. Behaviour Research and Therapy, 27, 373–383. Tellegen, A. (1985). Structure of mood and personality and their relevance to assessing anxiety, with an emphasis on self-report. In A. H. Tuma & J. D. Maser (Eds.), Anxiety and the anxiety disorders (pp. 681–706). Hillsdale, NJ: Erlbaum. Testa, T. J. (1975). Effects of similarity of location and temporal intensity pattern of conditioned and unconditioned stimuli on the acquisition of conditioned suppression in rats. Journal of Experimental Psychology: Animal Behavior Processes, 1, 114–121. Thorn, G. R., Chosak, A., Baker, S. L., & Barlow, D. H. (1999). Psychological theories of panic disorder. In D. J. Nutt, J. C. Ballenger, & J.-P. Lepine (Eds.), Panic disorder: Clinical diagnosis, management, and mechanisms (pp. 93–108). London: Martin Dunitz. Thyer, B., & Himle, J. (1985). Temporal relationships between panic attack onset and phobic avoidance in agoraphobia. Behaviour Research and Therapy, 23, 607–608. Timberlake, W. (1994). Behavior systems, associationism, and Pavlovian conditioning. Psychonomic Bulletin & Review, 1, 405–420. Timberlake, W., & Grant, D. L. (1975). Auto-shaping in rats to the presentation of another rat predicting food. Science, 190, 690–692. Timberlake, W., & Silva, K. M. (1995). Appetitive behavior in ethology, psychology, and behavioral systems. In N. S. Thompson (Ed.), Perspectives in ethology (Vol. 11, pp. 211–253). New York: Plenum. Timberlake, W., Wahl, G., & King, D. (1982). Stimulus and response contingencies in the misbehavior of rats. Journal of Experimental Psychology: Animal Behavior Processes, 8, 62–85. Tolman, E. C. (1932). Purposive behavior in animals and men. New York: Century. Trapold, M. A., & Overmier, J. B. (1972). The second learning process in instrumental learning. In A. H. Black & W. F. Prokasy (Eds.), Classical conditioning. II: Current research and theory (pp. 427–452). New York: Appleton-Century-Crofts. Tuma, A. H., & Maser, J. D. (Eds.). (1985). Anxiety and the anxiety disorders. Hillsdale, NJ: Erlbaum. Turkat, I. (1982). An investigation of parental modeling in the etiology of diabetic illness behavior. Behaviour Research and Therapy, 20, 547–552. Tweed, D. L., Schoenbach, V. J., George, L. K., & Blazer, D. G. (1989). The effects of childhood parental death and divorce on six-month history of anxiety disorders. British Journal of Psychiatry, 154, 823–828. Uhde, T., Boulenger, J., Roy-Byrne, P., Geraci, M., Vittone, B., & Post, R. (1985). Longitudinal course of panic disorder: Clinical and biological considerations. Progressive-Neuropsychopharmacology & Biological Psychiatry, 9, 39–51. Van den Hout, M. A. (1988). The explanation of experimental panic. In S. Rachman & J. D. Maser (Eds.), Panic: Psychological perspectives. Hillsdale, NJ: Erlbaum. VanDercar, D. H., & Schneiderman, N. (1967). Interstimulus interval functions in different response systems during classical discrimination conditioning of rabbits. Psychonomic Science, 9, 9–10.

A Modern Learning Theory Perspective on the Etiology 323 Van Houten, R., O’Leary, K. D., & Weiss, S. J. (1970). Summation of conditioned suppression. Journal of the Experimental Analysis of Behavior, 13, 75–81. Wagner, A. R. (1969). Stimulus selection and a modified continuity theory. In G. H. Bower & J. T. Spence (Eds.). The psychology of learning and motivation (Vol. 36, pp. 1–41). New York: Academic Press. Wagner, A. R. (1976). Priming in STM: An information-processing mechanism for self-generated or retrieval-generated depression in performance. In T. R. Tighe & R. N. Leaton (Eds.), Habituation: Perspectives from child development, animal behavior, and neurophysiology (pp. 95–128). Hillsdale, NJ: Erlbaum. Wagner, A. R. (1981). SOP: A model of automatic memory processing in animal behavior. In N. E. Spear & R. R. Miller (Eds.), Information processing in animals: Memory mechanisms (pp. 5–47). Hillsdale, NJ: Erlbaum. Wagner, A. R., & Brandon, S. E. (1989). Evolution of a structured connectionist model of Pavlovian conditioning (AESOP). In S. B. Klein & R. R. Mowrer (Eds.), Contemporary learning theories: Pavlovian conditioning and the status of traditional learning theory (pp. 149–189). Hillsdale, NJ: Erlbaum. Watson, D., & Clark, L. A. (1984). Negative affectivity: The disposition to experience negative emotional states. Psychological Bulletin, 96, 465–490. Watson, J. B., & Rayner, R. (1920). Conditioned emotional reactions. Journal of Experimental Child Psychology, 3, 1–14. Weiss, J. (1971). Effects of coping behavior in different warning-signal conditions on stress pathology in rats. Journal of Comparative and Physiological Psychology, 77, 1–13. Weiss, S. J., & Emurian, H. H. (1970). Stimulus control during the summation of conditioned suppression. Journal of Experimental Psychology, 85, 204–209. White, T. L., & Depue, R. A. (1999). Differential association of traits of fear and anxiety with norepinephrine- and dark-induced pupil reactivity. Journal of Personality and Social Psychology, 77(4), 863–877. Whitehead, W. W., Bush, C., Heller, B., & Costa, P. (1986). Social learning influences of menstrual symptoms and illness behavior. Health Psychology, 5, 13–23. Whitehead, W. W., Winget, C., Fedoravicius, A., Wooley, S., & Blackwell, B. (1982). Learned illness behavior in patients with irritable bowel syndrome and peptic ulcer. Digestive Diseases and Science, 27, 202–208. Williams, J., & Maier, S. (1977). Transituational immunization and therapy of learned helplessness in the rat. Journal of Experimental Psychology: Animal Behavior Processes, 3, 240–252. Williams, S. L., Kinney, P. H., Harap, S. T., & Liebmann, M. (1997). Thoughts of agoraphobic people during scary tasks. Journal of Abnormal Psychology, 106, 511–520. Wittchen, H. U., & Essau, C. A. (1991). The epidemiology of panic attacks, panic disorder and agoraphobia. In J. R. Walker, G. R. Norton, & C. A. Ross (Eds.), Panic disorder and agoraphogia (pp. 103–149). Monterey, CA: Brooks/Cole. Wolpe, J., & Rowan, V. C. (1988). Panic disorder: A product of classical conditioning. Behaviour Research and Therapy, 26, 441–450. Woods, S. C., & Strubbe, J. H. (1994). The psychobiology of meals. Psychonomic Bulletin & Review, 1, 141–155. Zinbarg, R. E., Barlow, D. H., Liebowitz, M., Street, L., Broadhead, E., Katon, W., Roy-Byrne, P., Lepine, J. P., Teherani, M., Richard, J., Brantley, P. J., & Kraemer, H. (1994). The DSM-1V field trial for mixed-anxiety depression. The American Journal of Psychiatry, 151, 1153–1162.

324 Mark E. Bouton, et al. Zinbarg, R., Brown, T., Barlow, D., & Rapee, R. (in press). Fear responses to hyperventilation and 5.5% carbon dioxide enriched air and anxiety sensitivity: A reanalysis of Rapee, Brown, Antony, & Barlow (1992). Journal of Abnormal Psychology. Zinbarg, R., & Mohlman, J. (1998). Individual differences in the acquisition of affectively valenced associations. Journal of Personality and Social Psychology, 74, 1024–1040. Zucker, D., Taylor, C. B., Brouillard, M., Ehlers, A., Margraf, J., Telch, M., Roth, W. T., & Agras, W. S. (1989). Cognitive aspects of panic attacks: Content, course, and relationship to laboratory stressors. British Journal of Psychiatry, 155, 86–91.

Article 16

A Proposal for a Dimensional Classification System Based on the Shared Features of the DSM-IV Anxiety and Mood Disorders: Implications for Assessment and Treatment Timothy A. Brown and David H. Barlow Center for Anxiety and Related Disorders, Boston University

A wealth of evidence attests to the extensive current and lifetime diagnostic comorbidity of the DSM-IV anxiety and mood disorders. Research has shown that the considerable cross-sectional covariation of DSM-IV emotional disorders is accounted for by common higher-order dimensions such as neuroticism/behavioral inhibition (N/BI) and low positive affect/behavioral activation. Longitudinal studies have indicated that the temporal covariation of these disorders can be explained by changes in N/BI and in some cases, initial levels of N/BI are predictive of the temporal course of emotional disorders. Moreover, the marked phenotypal overlap of the DSM-IV anxiety and mood disorder constructs is a frequent source of diagnostic unreliability (e.g., temporal overlap in the shared features of generalized anxiety disorder and mood disorders, situation specificity of panic attacks in panic disorder and specific phobia). Although dimensional approaches have been considered as a method to address the drawbacks associated with the extant prototypical nosology (e.g., inadequate assessment of individual differences in disorder severity), these proposals do not reconcile key problems in current classification such as modest reliability and high comorbidity. The current paper considers an alternative approach that emphasizes empirically supported common dimensions of emotional disorders over disorder-specific criteria sets. The selection and assessment of these dimensions are discussed along with how these methods could be implemented to promote more reliable and valid diagnosis, prognosis, and treatment planning. For instance, the advantages of this classification system are discussed in context of current transdiagnostic treatment protocols that are efficaciously applied to a variety of disorders by targeting their shared features.

326 Timothy A. Brown and David H. Barlow

Researchers have long recognized the limitations associated with the predominately categorical approach to diagnostic classification that has been adopted in the various editions of the Diagnostic and Statistical Manual of Mental Disorders (DSM). Although the development of the fifth edition of the DSM is well underway (DSM-V), no well-defined alternative nosological systems have been articulated to address these limitations. After providing an overview of the drawbacks inherent to the current DSM system, this paper will outline alternative approaches to diagnostic classification ranging from systems that incorporate dimensional elements into the extant diagnostic categories to systems that place greater emphasis on empirically supported common dimensions over disorder-specific criteria sets. The selection and operationalization of these dimensions will be discussed with emphasis on how these methods might be implemented to promote more reliable and valid clinical assessment, prognosis, treatment planning, and outcome evaluation. Limitations to the Current DSM System Diagnostic Reliability

Our nosological systems for mental disorders have moved from vague, ill-defined concepts reflecting hypothetical etiological constructs (DSM-II) to an atheoretical precise set of criteria based on presenting clinical features rather than assumptions about etiology (Brown & Barlow, 2002). This extreme “splitting” approach, present from DSM-III until now, was a necessary intermediate step in our clinical science; that is, becoming familiar with and defining more carefully the specific psychopathological phenomena comprising anxiety and mood disorders was essential to building an adequate nosological system. Indeed, diagnostic reliability studies of the DSM-III-R and DSM-IV disorders have indicated that this endeavor has been largely successful, reflected by the fact that most anxiety and mood disorder categories are associated with favorable reliability (e.g., Brown, Di Nardo, Lehman, & Campbell, 2001; Di Nardo et al., 1993; Mannuzza et al., 1989; Williams et al., 1992). For instance, in our investigation of the diagnostic reliability of current and lifetime DSM-IV anxiety and mood disorders (Brown et al., 2001), 362 outpatients underwent two, independent administrations of the Anxiety Disorders Interview Schedule for DSM-IV: Lifetime version (ADIS-IV-L; Di Nardo, Brown, & Barlow, 1994). All DSM-IV principal and lifetime diagnostic categories evidenced good to excellent reliability with the exception of dysthymia (DYS) and posttraumatic stress disorder (PTSD). Moreover, excellent reliability was obtained for each of the specific phobia (SPEC) types introduced in DSM-IV. In comparison to our reliability study of DSM-III-R disorders (Di Nardo et al., 1993), improved reliability was noted for the majority of DSM-IV categories, and no DSM-IV category was associated with a markedly lower reliability estimate. Diagnoses showing the most improvement were panic disorder (PD) and generalized anxiety disorder (GAD). The improvement in GAD was encouraging because this

A Proposal for a Dimensional Classification System Based on the Shared Features 327

category was in jeopardy of being removed from DSM-IV, in part due to evidence of poor to fair reliability in DSM-III-R (cf. Brown, Barlow, & Leibowitz, 1994). Although reliability estimates are generally favorable, limitations of the current classification system are evident when examining the sources of diagnostic disagreements. Brown et al. (2001) recorded the primary source of unreliability for each diagnostic disagreement using the following rating system: (a) difference in report—patient gives different information to the two interviewers (e.g., variability in responses to inquiry about the presence, severity, or duration of key symptoms); (b) threshold—consistent symptom report is provided across interviews, but interviewers disagree on whether these symptoms cause sufficient interference and distress to satisfy the DSM-IV threshold for a clinical disorder; (c) change in clinical status—clear change in the severity or presence of symptoms between interviews; (d) interviewer error—interviewer improperly applies DSM-IV diagnostic or exclusion rules or fails to obtain necessary diagnostic information during ADIS-IV-L administration (e.g., skips out of an ADIS-IV-L diagnostic section prematurely); (e) diagnosis subsumed under another condition—disagreement on whether symptoms are attributable to, or better accounted for by, a co-occurring disorder; and (f) DSM-IV inclarity—disagreement stems from limitations of the DSM-IV criteria in providing clear direction for differential diagnosis. As seen in Table 16.1, the prevailing sources of unreliability differed substantially across the DSM-IV anxiety and mood disorders. For instance, the majority of disagreements involving social phobia (SOC), SPEC, and obsessive-compulsive disorder (OCD; 62% to 67%) entailed cases where one interviewer assigned the diagnosis at a clinical level, and the other rated the diagnosis as subclinical; for other categories (e.g., panic disorder with agoraphobia [PDA], GAD, major depressive disorder [MDD], DYS), “threshold” disagreements were a relatively rare source of unreliability. “Difference in patient report” was otherwise the most prevalent source of unreliability, although with wide-ranging frequency across the anxiety and mood disorders (i.e., from 22% in SPEC to 100% in PTSD). Considerable variability was also evident across categories for the frequency with which other disorders were involved in diagnostic disagreements. Whereas disagreements with other disorders were relatively uncommon for SOC, OCD, and PTSD (8% to 13%), another clinical diagnosis was involved in over half of the disagreements with DYS, PDA, MDD, and GAD (54% to 74%). As shown in Table 16.1, disagreements with another clinical diagnosis often involved disorders with overlapping definitional features, and that differed mainly in the duration or severity of symptoms (e.g., PD vs. PDA; SPEC vs. agoraphobia without a history of panic disorder; MDD vs. DYS). Consistent with prior evidence that mood disorders may pose the greatest boundary problem for GAD (cf. Brown et al., 1994), 63% of the GAD disagreements with another diagnosis involved the mood disorders (DYS = 10, MDD = 9, depressive disorder NOS = 2, bipolar = 1). In addition, this overlap was evident in disagreements with anxiety disorder NOS and depressive disorder NOS

PD

PDA

Total Disagreements (n) 13 26 Proportion (and n) of: Disagreements Involving Another Clinical Diagnosis .38(5) .54(14) Disorders involved in the disagreement: PD .15(4) PDA .31(4) SPEC — .19(5) SOC — — GAD — .04(1) OCD — — PTSD — .04(1) MDD — — DYS — — Depression NOS — — Anxiety NOS .08(1) .08 (2) AGw/oPD — .04(1) Adjustment Disorder — — Other — — Clinical vs. Subclinical Disagreementsa .46 (6) .12(3) Sources of Unreliability Thresholdb .46 (6) .12(3) Difference in patient report .31(4) .50(13) Interviewer error .23 (3) .19(5) Diagnosis subsumed under — .15(4) comorbid disorder

Diagnosis:

.67 (26) .41 (16) .36 (14) .10(4) .10(4)

.62 (23) .46(17) .22 (8) .14(5) .08 (3)

— — —

— .08 (3) — — — — — — .05 (2) .03(1) — —

.08 (3)

.27(10)

— .03 (1) — — — — — — .14(5) — HYPOc

39

SOC

37

SPEC

.09 (4) .55 (26) .19(9) .15(7)

.13(6)

— .02(1) .19(9) .21(10) .04(2) .21(10) — .02(1) BIP

.02(1) — — —

.74(35)

47

GAD

Table 16.1 Sources of diagnostic disagreements for current DSM-IV Anxiety and mood disorders

.29 (6) .48 (10) .14(3) .10(2)

.62(13)

— — — — .10(2) — — —

— — — — —

.10(2)

21

OCD

— 1.00(8) — —

.25 (2)

— — — .13(1) — — —

— — — — — —

.13(1)

8

PTSD

.15(8) .55 (29) .11(6) —

.17(9)

.23 (12) .28(15) .02(1) — .06 (3) PAIN

— — — — .04 (2) — —

.64 (34)

53

MDD

.05 (2)d .66 (27) .15(6) .15(6)



.12(5) — — .02(1) SOM

— — — — .15(6) — — .39(16)

.71 (29)

41

DYS

— — —

.04(1) — —

.03 (1) .05 (2) .03 (1)

— .03(1) —

.02(1) — —

— — —

— — —

.17(9) — .02(1)

— — —

Cases where both raters recorded the diagnosis, but disagreed on whether it should be assigned at the clinical level.

This disagreement pertained to a boundary issue between specific phobia, other type (contracting an illness) and hypochondriasis.

Table reproduced from Brown,T.A., Di Nardo, P.A., Lehman, C. L., & Campbell, L.A. (2001). Reliability of DSM-IV anxiety and mood disorders: Implications for the classification of emotional disorders. Journal of Abnormal Psychology, 110, 49–58.

d

In these 2 cases, disagreements stemmed from a threshold issue of MDD (chronic) vs. DYS; that is, whether the features were sufficiently severe to be classified as a chronic major depressive episode.

c

b

Note that, for SPEC and other diagnoses, the frequency of cases where “Threshold” was the primary source of unreliability does not necessarily equal the number of disagreements entailing assignment of the disorder at clinical and subclinical levels. This is because each of the other unreliability sources (e.g., change in clinical status, variability in patient report regarding number or severity of symptoms) could also result in clinical vs. subclinical disagreements.

a

Note. PD = panic disorder; PDA = panic disorder with agoraphobia; SPEC = specific phobia; SOC = social phobia; GAD = generalized anxiety disorder; OCD = obsessive-compulsive disorder; PTSD = posttraumatic stress disorder; MDD = major depressive disorder; DYS = dysthymia; Depression NOS = depressive disorder not otherwise specified; Anxiety NOS = anxiety disorder not otherwise specified; AGw/oPD = agoraphobia without a history of panic disorder; HYPO = hypochondriasis; BIP = bipolar disorder; PAIN = pain disorder; SOM = somatization disorder.

Change in clinical status DSM-IV inclarity Missing

330 Timothy A. Brown and David H. Barlow

diagnoses. For example, a category frequently involved in disagreements with GAD was anxiety disorder NOS, where one interviewer noted clinically significant features of GAD, but judged that all criteria for a formal DSM-IV GAD diagnosis had not been met (e.g., number or duration of worries or associated symptoms). This was also the case for the NOS diagnoses associated with disagreements in other disorders [e.g., in the two OCD disagreements involving another disorder, both were with anxiety NOS (OCD)]. Comorbidity

Consistent findings of high comorbidity among anxiety and mood disorders underscore the poor distinguishability of the emotional disorders (e.g., Andrews, 1990). Comorbidity studies based on DSM-III-R criteria routinely indicated that at least 50% of patients with a principal anxiety disorder have one or more additional diagnoses at the time of assessment (e.g., BrawmanMintzer et al., 1993; Brown & Barlow, 1992; Sanderson et al., 1990). Similar findings were obtained in a study of outpatients (N = 1,127) diagnosed with DSM-IV anxiety and mood disorders (Brown et al., 2001). It is important to note that this study, like others, probably yielded conservative estimates of diagnostic co-occurrence due to limits in generalizability stemming from various exclusion criteria (e.g., cases with active suicidality were excluded), the outpatient setting, and so forth. Nevertheless, comorbidity rates for many categories were quite high. Results indicated that 55% of patients with a principal anxiety or mood disorder had at least one additional anxiety or depressive disorder at the time of the assessment (Table 16.2); this rate increased to 76% when lifetime diagnoses were considered. The principal diagnostic categories of PTSD, MDD, DYS, and GAD had the highest comorbidity rates, and SPEC, the lowest. Analyses examining patterns of covariation among current and lifetime disorders yielded many interesting findings. For example, significant relative risks were obtained for associations between SOC and the mood disorders (MDD, DYS). The differential aggregation of SOC and mood disorders was noteworthy in view of structural findings (Brown, 2007; Brown et al., 1998) indicating that, unlike other DSM-IV disorders, the latent DSMIV factors of SOC and Depression (DEP) are influenced by the higher-order trait of low positive affect. Thus, the high diagnostic comorbidity between SOC and mood disorders may due to a shared vulnerability dimension such as low positive affect/behavioral activation (cf. Clark et al., 1994; Eley & Brown, 2009). Moreover, PTSD was associated with significantly increased risk of PDA. Temporal sequences analyses indicated that PDA rarely (28%) preceded PTSD. This association is noteworthy because PTSD and PDA are the only two emotional disorders characterized by high levels of autonomic arousal (Brown & McNiff, in press). However, the findings of Brown et al. (2001) also revealed how the examination of comorbidity at the level of DSM-IV diagnosis can produce misleading findings about the overlap among disorders. For instance, the presence of PDA was associated with decreased relative risk of conditions such as SOC and SPEC. Rather than reflecting a true lack of association between these conditions (indeed, one would expect considerable phenotypic overlap of these

08 (.41) 24(1.41*) 23(1.28) 08 (.99) 07 (.73) 07 (.74) 03 (.74) 03 (.87) 03 (.84)

— — — 15 (.53*) 16(1.64*) 22 (.78*) 07(1.08) 15(1.28) 04(1.95) 04(1.81)

03 (3.25) 15 (2.09*) 18 (2.24*) 36(2.33*) — — 04 (.59) 12 (.91) 01 (.30) 01 (.25)

14 (.66*) 26(1.39) 13(1.75*) 06 (.66) 05 (1.40) 06(1.71)

01 (.28) 03 (.24*) 03 (.24*) — 13(1.10) 21 (.78) 08(1.29) 08 (.58*) 03 (1.09) 03(1.10)

00 (—) 05 (.58) 05 (.49) 09 (.38*) 05 (.40*) 07 (.26*) 03 (.38) 15b (1.17) 00 (—) 03 (.90)

34 (.56*) 33 (.56*) 27 (.61*) 10 (.34*)

22(1.14) 03 (.12*) 10 (1.26) 04 (.40) 04(1.05) 04 (.98)

01 (.86) 08 (.88) 09 (.88) 26(1.21) 12 (.93) 16 (.59) — 12 (.91) 00 (—) 03 (.86)

57(1.01) 53 (.96) 39 (.91) 32(1.17) 04 (3.25) 15(1.90*) 19(2.08*) 41 (2.07*) 05 (.37*) 67(3.16*) 09(1.35) 15(1.18) 06 (2.84*) 02 (.81)

69(1.25*) 68(1.26*) 64(1.58*) 11 (.37*)

69(3.67*) — 23 (2.80*) 11(1.36) 00 (—) 00 (—)

00 (—) 23 (2.75) 23 (2.33) 15 (.70) 23(1.87) 38(1.52) 23 (3.62*) 15(1.21) — 00 (—)

92(1.64*) 92(1.69*) 62(1.45) 77 (2.79*)

03 (2.48) 15 (1.96*) 18 (2.03*) 42 (2.24*) 05 (.36*) 72 (3.73*) 08(1.21) 14(1.09) 05 (2.20) 02 (.63)

71 (1.28*) 70(1.30*) 63 (1.56*) 17 (.56*)

(Continued )

20 08 04

01 09 10 22 13 25 07 13 03 03

57 55 43 28

MDD/DYS Overalla (102)

33 (1.73) — — — 05 (1.29) 01 (.24)

00 (—) 14(1.68) 14 (1.42) 48 (2.25*) 05 (.38) 90(3.80*) 05 (.71) 10 (.75) 00 (—) 00 (—)

76(1.36*) 76(1.40*) 57(1.35) 38(1.36)

— — — 15 (.58*) 16(1.50*) 22 (.79) 07(1.12) 15(1.36) 04 (2.30*) 04(1.86)

68(1.24*) 65(1.21*) 52(1.25*) 36(1.32*)

— — — 08 (.37) 19(1.59) 22 (.87) 06 (.83) 08 (.65) 00 (—) 03 (.92)

46 (.78*) 45 (.79*) 28 (.61*) 29(1.04)

62(1.14*) 60(1.16*) 47(1.17*) 33 (1.28*)

42 (.73) 42 (.75) 36 (.84) 17 (.58)

Any Axis I Any anxiety/mood Any anxiety disorder Any mood disorder Anxiety Disorders PD PDA PD or PDA SOC GAD GAD: no hierarchy OCD SPEC PTSD Other anxiety Mood Disorders MDD DYS Other mood Other Disorders

60(1.09) 59(1.12) 46(1.14) 31(1.19)

PD (36) PDA (324) PD/A (360) SOC (186) GAD (120) OCD (77) SPEC (110) PTSD (13) MDD (81) DYS (21)

Current Additional Diagnosis

DSM-IV Principal Diagnosis (n)

Table 16.2 Percentage (and relative risk) of current additional diagnoses in patients with current principal anxiety and mood disorders (N = 968)

04(3.53*) 01 (.83) 00 (—) 06 (2.25) 06 (2.41) 02 (.58)

00 (—) 00 (—)

02(1.68) 00 (—) 06 (2.28) 00 (—)

02(1.31) 05(1.77)

02 03

MDD/DYS Overalla (102)

Overall frequency category was assigned as an additional diagnosis.

* Significantly increased or decreased risk of co-occurring disorder (95% confidence interval). Table reproduced from Brown, T. A., Campbell, L. A., Lehman, C. L., Grisham, J. R., & Mancill, R. B. (2001). Current and lifetime comorbidity of the DSM-IV anxiety and mood disorders in a large clinical sample. Journal of Abnormal Psychology, 110, 585–599.

a

Note. DSM-IV = Diagnostic and Statistical Manual of Mental Disorders, 4th Edition (American Psychiatric Association, 1994); PD = panic disorder; PDA = panic disorder with agoraphobia; PD/A = panic disorder with or without agoraphobia; SOC = social phobia; GAD = generalized anxiety disorder; GAD: no hierarchy = generalized anxiety disorder ignoring DSM-IV hierarchy rule with mood disorders; OCD = obsessive-compulsive disorder; SPEC = specific phobia; PTSD = posttraumatic stress disorder; MDD = major depressive disorder; DYS = dysthymia; MDD/DYS = major depressive disorder or dysthymia.

01 (.65) 01 (.31)

00 (—) 02 (.99) 03 (.92) 02 (.63)

Somatoform Other Axis I

01 (.84) 02 (.64)

PD (36) PDA (324) PD/A (360) SOC (186) GAD (120) OCD (77) SPEC (110) PTSD (13) MDD (81) DYS (21)

Current Additional Diagnosis

DSM-IV Principal Diagnosis (n)

Table 16.2 (Continued)

A Proposal for a Dimensional Classification System Based on the Shared Features 333

disorders; e.g., situational avoidance), such findings can be a byproduct of DSM-IV differential diagnostic guidelines (i.e., features of social or specific fear/avoidance were often judged to be better accounted for and thereby subsumed under the PDA diagnosis). A clear instance of this phenomenon was findings on the comorbidity of GAD and mood disorders. When adhering strictly to DSM-IV diagnostic rules, the comorbidity between GAD and DYS was 5%. However, when ignoring the hierarchy rule that GAD should not be assigned when it occurs exclusively during a course of a mood disorder, the comorbidity estimate rose to 90%. While curtailing (or distorting) comorbidity, DSM’s differential diagnostic guidelines also forfeit salient clinical information. Strict adherence to DSM-IV criteria does not acknowledge the common situation where clinically significant GAD co-resides with a mood disorder or PTSD. Nonetheless, such symptoms are relevant to the overall severity of the clinical presentation, and may have strong implications for treatment planning, untreated course, and so forth (Kessler et al., 2008; Lawrence, Liverant, Rosellini, & Brown, 2009; Zimmerman & Chelminski, 2003). Categorical Classification

Although there are obvious practical advantages to a predominately categorical diagnostic classification system (e.g., use in clinical practice; cf. First, 2005), there are many serious drawbacks to this approach. For instance, as noted earlier, our diagnostic reliability study of the DSM-IV anxiety and mood disorders (Brown, Di Nardo, et al., 2001) found that for many categories (e.g., SOC, OCD), diagnostic disagreements were primarily due to problems in defining and applying a categorical threshold on the number, severity, or duration of symptoms. This threshold problem is also manifested in diagnostic disagreements where both raters concur that the key features of a disorder are present, but disagree as to whether these features cause sufficient interference or distress to satisfy the DSM-IV threshold for a clinical disorder (common with SOC and SPEC). Moreover, the problem is evident in the high rates of disagreements involving NOS diagnoses (both raters agree on the presence of clinically significant features of the disorder, but one rater does not assign a formal anxiety or mood disorder diagnosis due to subthreshold patient report of the number or duration of symptoms; common with generalized anxiety disorder and major depression). A similar problem is at the root of diagnostic disagreements involving MDD vs. DYS (core features of clinically significant depression are observed by both raters, but disagreement occurs with regard to the severity or duration of these symptoms). In addition to introducing measurement error (cf. MacCallum, Zhang, & Preacher, 2003), imposing categories on dimensional phenomena leads to a substantial loss of potentially valuable clinical information. As noted by Widiger and Samuel (2005), DSM does not provide adequate coverage for clinically significant symptom presentations that fail to meet criteria for formal diagnostic categories.

334 Timothy A. Brown and David H. Barlow

This is reflected in part by the high rate in which NOS diagnoses are assigned as current and lifetime conditions (e.g., Brown, Campbell, et al., 2001; Lawrence & Brown, in press), but it is likely that clinically significant subthreshold cases are altogether undetected by the categorical system. Although rarely used in clinical practice and applied research, DSM does provide a coarse mechanism for recording the severity of disorders above the diagnostic threshold (i.e., as detailed in the “Use of the Manual” section of the DSM-IV, the ordinal specifiers of mild, moderate, and severe may be used for any current disorder meeting full diagnostic criteria). The only DSM-IV anxiety and mood disorder category where individual differences in symptom severity are embedded in a disorder’s diagnostic criteria is MDD (i.e., severity specifiers of mild, moderate, and severe with/without psychotic features are recorded when assigning the MDD diagnosis). Yet, whereas the dimensional ratings of the severity of MDD symptoms are reliable (r = .74), the DSM-IV categorical severity specifiers of this disorder are not (e.g., κ = .30; Brown et al., 2001). This further attests to the problem of measurement error associated with imposing nominal or ordinal cutoffs on continuous symptom features. Summary and Illustration of DSM-IV Limitations

The preceding sections summarized some of the research that highlighted various limitations associated with DSM’s categorical approach to diagnostic classification (e.g., measurement error due to operationalizing and applying a categorical cutoff on dimensional features such as symptom frequency, severity, and duration; distorted rates of diagnostic comorbidity due to overlapping criteria sets and diagnostic decision rules; inadequate coverage of subthreshold presentations and individual differences in severity of threshold cases). As noted earlier, the issues of modest diagnostic reliability and high comorbidity have been routinely associated with GAD, thus making it an appropriate diagnostic category for further illustrating the shortcomings of the DSM-IV classification system. For clinical presentations entailing the features of GAD (excessive worry and associated features such as muscle tension and irritability), the potential sources of diagnostic unreliability are multifold. As is the case with most disorder categories, DSM-IV’s operationalization of GAD categorical diagnostic threshold is not based on compelling empirical evidence and is difficult to reliably implement in clinical practice. For instance, some of the requirements for a DSM-IV diagnosis of GAD are “excessive anxiety and worry, occurring more days than not for at least 6 months, about a number of events or activities” (Criterion A), “the anxiety and worry are associated with three (or more) of six symptoms (with at least some symptoms present for more days than not for the past 6 months)” (Criterion C), and “does not occur exclusively during a mood disorder” (and other conditions such as PTSD and psychotic disorders) (Criteria D and F). In addition to their dubious scientific basis (e.g., limited evidence to support the symptom number and duration cutoffs specified in Criterion C), the criteria are vaguely worded and thus require considerable clinical judgment in their implementation for establishing the presence/absence of a GAD diagnosis.

A Proposal for a Dimensional Classification System Based on the Shared Features 335

While this was intentional (owing in part to the absence of data in support of more specific symptom cutoffs), the DSM-IV approach to operationalizing disorder thresholds introduces considerable measurement error. For instance, what “number” of worry areas are necessary to meet Criterion A (two? several?)?; how many associated symptoms must be present more days than not over the prior 6 months (one? more than one?)? Should Criterion F be interpreted as indicating that GAD can be assigned in context of a co-occurring mood disorder so long as there has been a period of at least a few days or weeks when GAD was present and the mood disorder was not (or is a full 6-month duration of GAD without a co-occurring mood disorder necessary to meet this criterion?). Research laboratories may be inclined to develop more refined operationalizations of these vaguely worded diagnostic criteria in order to foster the reliability of diagnostic judgments within their setting. However, such operationalizations may vary across research laboratories (i.e., variability in the manner in which GAD diagnoses are assigned) which would thus complicate the comparability and interpretability of findings across studies. In addition to obscuring true patterns of symptom and syndrome comorbidity, the diagnostic hierarchy rule stating GAD should not be assigned if it occurs during the course of a mood disorder can lead to instances where GAD cannot be formally assigned despite the fact that it is associated with more distress and impairment than the mood disorder it is subsumed under (e.g., severe GAD occurring during a mild MDD). Finally, as is the case with other disorders, the failure of DSM-IV to recognize individual differences in the frequency, severity, and duration of symptoms can result in clinically salient GAD presentations that are not recognized or a generically coded as anxiety disorder NOS (cf. Lawrence & Brown, in press). Alternative Classification Systems For decades, researchers have acknowledged the potential utility of incorporating dimensional elements into our diagnostic classification systems (e.g., Barlow, 1988; Kendell, 1975; Maser & Cloninger, 1990; Widiger, 1992). Over this considerable time span, however, no strong proposals have emerged with regard to exactly how dimensional classification could be introduced in the DSM. In an earlier paper (Brown & Barlow, 2005), we forwarded an initial proposal for this endeavor. Given that clinical practicality is a compelling reason for retaining categorical distinctions in the nosology, we suggested the introduction of dimensional severity ratings to the extant diagnostic categories and/or the constituent symptom criteria (along the lines of the procedures used in the Anxiety Disorders Interview Schedule for DSM-IV; cf. Brown et al., 1998). Compared to more drastic proposals (e.g., multidimensional assessment, in which categorical diagnostic labels are subsequently imposed on the basis of quantitative algorithms), it was argued that this alternative would be relatively practical because the categorical system would remain intact and the dimensional rating system could be optional in settings where its implementation is less feasible (e.g., primary care). Several potential advantages of this dimensional system were noted including the ability to address key

336 Timothy A. Brown and David H. Barlow

shortcomings and sources of unreliability in the DSM, such as its failure to convey disorder severity as well as other clinically significant features that are either subsumed by other disorders (e.g., GAD in mood disorders and PTSD) or fall just below conventional thresholds due to a DSM technicality (e.g., subclinical or NOS diagnoses where the clinical presentation is a symptom or two short of a formal disorder). Moreover, because the dimensional ratings would be added to the current diagnostic categories, this approach would have other advantages including: (a) its basis on a pre-existing and widely studied set of constructs; and (b) the ability to retain functional analytic and temporal (duration) aspects of diagnosis that are difficult to capture in a purely psychometric approach. It would also provide a standardized assessment system that would foster across-site comparability in the study of dimensional models of psychopathology. Finally, we argued that this approach could be regarded as a prudent “first step” that would assist in determining the feasibility of more ambitious dimensional systems (e.g., quantifying higher-order dimensions). Nonetheless, this initial proposal is not without immediately apparent limitations. For instance, as noted earlier, “difference in patient report” (i.e., patient gives different information to independent interviewers in response to inquiries about the presence, severity, or duration of symptoms) is a very common source of diagnostic unreliability that would be relevant to dimensional clinical assessment. In fact, because the dimensional ratings would simply be added onto the existing criteria sets, most sources of unreliability present in the current diagnostic system would continue to be germane (e.g., measurement error associated with vaguely operationalized symptom criteria and differential diagnosis decision rules; see GAD example in preceding paragraph). Perhaps more importantly, because the various disorder categories would remain unchanged, a dimensional system of this nature would not address the problem of high diagnostic comorbidity. Higher-Order Models of Classification

Thus, more ambitious proposals have been suggested which place emphasis on dimensions corresponding to broader biologically and environmentally based constructs of temperament and personality (e.g., neuroticism/negative affectivity; Clark, 2005). These strategies follow from the theories and evidence that the observed overlap in families of disorders (e.g., comorbidity and symptom overlap in anxiety and mood disorders) is due to the fact that these conditions emerge from shared biologic/genetic and psychosocial diatheses (e.g., Barlow, 2002; Kendler et al., 1992; Mineka et al., 1998). Under this framework, the DSM disorders represent different manifestations of these core vulnerabilities, and such variability stems from the influence of other, more specific etiologic agents (e.g., environmentally based psychological vulnerabilities, other genetic or biological influences; cf. Barlow, 2002; Suárez, Bennett, Goldstein, & Barlow, 2009). Two genetically based core dimensions of temperament have been posited as instrumental in the etiology and course of the emotional disorders: neuroticism/negative affectivity and extraversion/positive affectivity. Extensive evidence indicates these constructs are strongly heritable (e.g., Fanous, Gardner,

A Proposal for a Dimensional Classification System Based on the Shared Features 337

Prescott, Cancro, & Kendler, 2002; Fulker, 1981; Hettema, Prescott, & Kendler, 2004; Tellegen et al., 1988; Viken, Rose, Kaprio, & Koskenvuo, 1994) and stable over time (Costa & McCrae, 1988; Kasch, Rottenberg, Arnow, & Gotlib, 2002; Watson & Clark, 1984). Whereas neuroticism/negative affect is considered to be etiologically relevant to the full range of emotional disorders, the influence of extraversion/positive affect appears to be more specific to depression, SOC, and mania, with depression and SOC associated with low positive affect and mania with high positive affect (e.g., Brown, 2007; Brown, Chorpita, & Barlow, 1998; Gruber, Johnson, Oveis, & Keltner, 2008; Johnson, Gruber, & Eisner, 2007; Mineka et al., 1998; Watson, Clark, & Carey, 1988). Although the theoretical frameworks were developed independently (cf. Eysenck, 1981; Tellegen, 1985), neuroticism/negative affect and extraversion/positive affect are closely related to Gray’s (1987) constructs of behavioral inhibition and behavioral activation, respectively, at both the conceptual and empirical levels (e.g., Campbell-Sills, Liverant, & Brown, 2004; Carver & White, 1994; Clark et al., 1994; Fowles, 1994; Kasch et al., 2002). A substantial literature underscores the roles of these constructs in accounting for the onset, overlap, and maintenance of anxiety and depression (e.g., Brown, 2007; Brown et al., 1998; Gershuny & Sher, 1998; Kasch et al., 2002; Watson et al., 1988). For instance, in a sample of outpatients, Brown et al. (1998) found that virtually all the considerable covariance among latent variables corresponding to the DSM-IV constructs of unipolar depression (DEP), social phobia (SOC), generalized anxiety disorder (GAD), obsessive-compulsive disorder (OCD), and panic disorder/agoraphobia (PD/A) was explained by the higher-order dimensions of negative affect and positive affect (bipolar disorder was not included). Although the results were consistent with the notion of neuroticism/negative affect as a broadly relevant dimension of vulnerability, results indicated the DSM-IV disorders were differentially related to negative affect, with DEP and GAD evidencing the strongest associations (see also Brown & McNiff, 2009). In accord with a reformulated hierarchical model of anxiety and depression (Mineka et al., 1998), positive affect was predictive of DEP and SOC only (see Figure 16.1). Brown (2007) extended these findings by examining the temporal course and temporal structural relationships of dimensions of temperament (neuroticism/behavioral inhibition, N/ BI; behavioral activation/positive affect, BA/P) and selected DSM-IV disorder constructs (DEP, SOC, GAD). Outpatients with anxiety and unipolar mood disorders (N = 606) were reassessed at one-year (T2) and two-year (T3) follow-ups. As the majority (76%) of patients received treatment after intake, the overall rate of anxiety and mood disorders declined significantly over follow-up (e.g., 100% to 58% for T1 and T3, respectively). Nonetheless, each of the five constructs examined (N/BI, BA/P, DEP, GAD, SOC) evidenced a favorable level of longitudinal measurement invariance. Despite the marked decline in DSM-IV diagnoses by T3, test-retest correlations of the factors and unconditional latent growth models (LGMs) indicated that BA/P evidenced a very high degree of temporal stability, consistent with its conceptualization as a trait vulnerability construct that is relatively unaffected by treatment. However, of the five constructs examined, N/BI evidenced the greatest amount of temporal change and was the dimension associated

338 Timothy A. Brown and David H. Barlow

Figure 16.1 Structural Model of the Interrelationships of DSM-IV Disorder Constructs and Negative Affect, Positive Affect, and Autonomic Arousal. Completely standardized estimates are shown (path coefficients with asterisks are statistically significant, p < .01). Figure reproduced from Brown, T. A., Chorpita, B. F., & Barlow, D. H. (1998). Structural relationships among dimensions of the DSM-IV anxiety and mood disorders and dimensions of negative affect, positive affect, and autonomic arousal. Journal of Abnormal Psychology, 107, 179–192.

with the largest treatment effect. In addition to its inconsistency with prior research (e.g., Kasch et al., 2002), this finding seems at odds with conceptual speculations that core dimensions of temperament may be more resilient to psychological treatment. As discussed in Brown (2007), this result may partly reflect the fact that the assessment of temperament is prone to mood-state distortion (cf. Clark et al., 2003; Widiger et al., 1999). That is, it is likely the measurement of N/BI consists of some combination of stable temperament variance (i.e., vulnerability) and variability attributable to generalized distress (i.e., more prone to mood-state distortion, subject to greater temporal fluctuation). Thus, the considerable covariance between N/BI and the emotional disorders is due partly to a temperamental component (N/BI acts as vulnerability, is relatively stable), but also because of the distress associated with having a disorder (mood-state distortion). Presumably, the latter aspect is less temporally unstable and more apt to covary with temporal fluctuations in the severity of disorders. Indeed, findings indicated that N/BI operated differently than the DSM-IV disorder constructs in several ways. For instance, unconditional LGMs of each DSM-IV disorder construct revealed inverse relations

A Proposal for a Dimensional Classification System Based on the Shared Features 339

between the Intercept and Slope; i.e., higher initial disorder severity was associated with greater change over time. However, the Intercept and Slope of N/BI were positively correlated (r = .47), indicating that patients with higher initial levels of N/BI tended to show less change in this dimension over time; and conversely, patients with lower initial levels of N/BI tended to evidence greater change. Thus, unlike the DSM-IV disorders, the stability of N/BI increased as a function of initial severity. This may indicate the influence of mood-state distortion/general distress on the measurement of N/BI is most pronounced at the lower end of its continuum (i.e., it is the lower range of N/BI that is less temporally stable and presumably more apt to covary with temporal change in disorder severity). In addition, parallel-process LGMs indicated that higher initial levels of N/BI were associated with less change in the DSM-IV constructs of GAD and SOC. Moreover, lower BA/P predicted poorer outcome of SOC, although this effect only approached statistical significance (p = .07) with N/BI in the analysis. Although no temporal relations were obtained for DEP, these results are in line with earlier work and theory that N/BI and BA/P have directional temporal effects on Axis I psychopathology (cf. Gershuny & Sher, 1998; Kasch et al., 2002; Meyer, Johnson, & Winters, 2001). Consistent with prediction and theory, initial levels of the DSM-IV disorders did not predict increases in temperament over time. Finally, Brown (2007) found that temporal change in DEP, SOC, and GAD was significantly related to change in N/BI. Of particular interest is the finding that all the temporal covariance of the DSM-IV disorder constructs was accounted for by change in N/BI; i.e., when N/BI was specified as a predictor, the temporal overlap among disorder constructs was reduced to zero. The correlational nature of these findings precludes firm conclusions about the direction of these effects. Although counter to earlier initial evidence and conceptualization (Brown et al., 1995; Kasch et al., 2002), N/BI may be therapeutically malleable and this in fact mediates the extent of change in the emotional disorders. Alternatively or additively, a reduction in disorder severity is associated with a decrease in general distress, a feature shared by the emotional disorders and is partially reflected in the measurement of N/BI. Finally, in the case of bipolar disorder, Johnson and colleagues (Gruber et al., 2008; Johnson et al., 2007; Meyer et al., 2001) have demonstrated that risk for mania is associated with stable, elevated BA/P. A Nosological Emphasis on Shared Dimensions

Compared to the more modest proposal of adding dimensional elements to the extant diagnostic categories (cf. Brown & Barlow, 2005), the inclusion of higher-order dimensions in the nosology would address the considerable overlap that exists among the current DSM-IV anxiety and mood disorder constructs. As we noted previously (Brown & Barlow, 2005), because dimensions such as N/BI and BA/P reflect general vulnerability constructs, their inclusion would make the diagnostic system germane to the primary and secondary prevention of mental disorders (e.g., identification of individuals with elevated N/BI who thus may be at risk for psychopathology if exposed to other vulnerability factors such as life stressors). Of course, this is a far more challenging endeavor because these broader phenotypes are not currently

340 Timothy A. Brown and David H. Barlow

recognized by the DSM and thus must be further validated and measured in a manner that is feasible in clinical practice and research. Moreover, an exclusive focus on higher-order dimensions could be viewed as overly reductionistic because there is considerable variability in which these core vulnerabilities are manifested as clinical psychopathology. Although the covariance of SOC and MDD can be fully explained by the shared vulnerability dimensions of N/BI and BA/P (Brown, 2007; Brown et al., 1998), further differentiation is important because constructs more specific to these disorders provide information about treatment planning, risk of complications (e.g., suicidality), and so forth, than does only knowing a patient’s standing on the continuum of these nonspecific higher-order dimensions. However, while more germane to some DSM-IV disorder constructs than to others, these lower-order dimensions are also present in varying degrees across the emotional disorders (e.g., while fear of negative social evaluation is the defining feature of SOC, it is germane to other DSM-IV disorder constructs such as PD/A, mood disorders, and body dysmorphic disorder). While having the advantage of providing important additional information about the nature of an individual’s psychopathology, the inclusion of these lower-order dimensions would perpetuate the rather weak boundaries among the DSM disorder constructs. But by also including broader dimensions such as N/BI and B/ PA, the nosological system would possess a hierarchical framework that accounts for the overlap at the lower-order level. Indeed, current conceptual models assert that dimensions of temperament do not act alone in determining the etiology, course, and complications of emotional disorders (e.g., Barlow, 2002; Carver, Johnson, & Joormann, 2008; Clark et al., 1994; Mineka et al., 1998). For instance, Barlow (2000, 2002; Suárez et al., 2009) has formulated a triple vulnerability model of emotional disorders, which draws from and integrates the rich literatures of genetics, personality, cognitive and neuroscience, and emotion and learning theories. This model specifies the existence of a generalized biological vulnerability to experience anxiety consisting of a well-established genetic contribution to this diathesis accounting for approximately 30% to 50% of the variance. In addition, a generalized psychological vulnerability emerges from early childhood experience characterized by a stressful, unpredictable environment and/or the influence of specific parenting styles described in detail in the attachment theory literature that inhibit the development of effective coping procedures and the emergence of self efficacy. These early experiences lead to a general sense of unpredictability and uncontrollability over life events that, along with elevated sympathetic nervous system arousal, forms the core of the process of anxiety (Barlow, 2002). If these two generalized vulnerabilities or diatheses line up, the individual is at increased risk for experiencing generalized anxiety and/or depression, in the context of triggering stressful events (Chorpita & Barlow, 1998; Suárez et al., 2009). But a third diathesis, referred to as a specific psychological vulnerability, comes into play in the form of learning a particular focus for anxiety, or learning that some situations, objects, or internal somatic states are potentially dangerous even if objectively they are not. These early learning experiences can be as straightforward as watching parents’ model severe fears of specific objects or situations such as small animals (e.g., as in SPEC), or

A Proposal for a Dimensional Classification System Based on the Shared Features 341

more subtle such as experiencing heightened attention from caregivers to the potential danger of experiencing unexplained somatic sensations (e.g., as in PD/A or hypochondriasis). One may notice that the two generalized vulnerabilities, one biological and the other psychological, describe a more stable disposition to experience anxiety, and thus, could be considered more accurately as a temperament. Indeed we consider anxiety in this regard as simply the expression of the temperament of N/BI with the addition, in many cases, of a specific focus (or several specific foci) dictated by the learning experiences that comprise the third vulnerability. Evidence for this relationship is presented in detail elsewhere (Barlow, 2002; Suárez et al., 2009). A New Proposal for a Dimensional Classification System These considerations, then, lead to a dimensional classification system for emotional disorders emphasizing common features but encompassing the variety of phenotypic expressions emerging from different foci of anxiety. At the highest level, an individual would experience various levels of severity in terms of intensity, frequency, and distress associated with anxiety or N/BI (which we now refer to as anxiety/neuroticism/behavioral inhibition, or AN). This temperament is associated with chronic generalized distress, particularly involving the hypothalamic-pituitary-adrenocortical axis, perceptions of uncontrollability regarding future threatening and challenging events, excessive vigilance, and low self confidence/self-efficacy over one’s ability to cope with future threatening events. In addition to assessing the presence and severity of AN, the second higher-order dimension of behavioral activation/positive affectivity (BA/P) must also be considered (see Figure 16.1). Low positive affect reflects principally low enthusiasm, interest, and an overriding pessimistic sense, and is associated with the DSM-IV disorder constructs of MDD and SOC, as noted above. Once again, structural models demonstrate that AN provides substantial contributions to these two DSM-IV disorder constructs, but that low BA/P is also part of the picture. In the case of SOC, this may well be associated with the chronically impoverished network of social relationships and support which contributes so substantially to positive affect based on the work of scientists studying the determinants of positive affect in the normal population (e.g., Diener & Seligman, 2002; Gilbert, 2006). In clinical populations, low-level manifestations of BA/P merge into a presentation that has been referred to as “mixed anxiety depression” (Moras et al., 1996; Zinbarg et al., 1994). High level manifestations of BA/P are associated with euthymic stages of bipolar and cyclothymic disorders. Each individual experiencing varying degrees of AN would also almost certainly display substantial avoidance behaviors in all of its behavioral and cognitive manifestations. These manifestations would include situational avoidance, most obviously, in the case of phobic avoidance. Because SPEC in isolation is not characterized by marked AN, specification of situational avoidance could simply be represented on this dimension. A variant of situational

342 Timothy A. Brown and David H. Barlow

avoidance that is common across the anxiety disorders is the avoidance of internal somatic cues such as avoiding a hot and stuffy room, physical exertion, and situations provoking perceptual distortions associated with dissociation. Although much of this behavioral avoidance may appear to be situational, the cues for avoidance are actually interoceptive (e.g., a hot and stuffy room). But more subtle cognitive and or emotional avoidance exemplified most commonly by attempts at distraction or the presence of safety signals or talismans meant to (magically) protect against the occurrence of AN and/or the eruptions of discrete emotions such as panic attacks comprise another facet of avoidance. The extent of this avoidance could also be dimensionalized, leading to further specification with implications for treatment planning. Table 16.3 provides examples of various avoidant strategies. Table 16.3 Examples of Avoidance Strategies

Avoidance Strategy I. Behavioral and Interoceptive Avoidance Situational avoidance/escape Avoidance/escape from phobic situations (e.g., crowds, parties, elevators, public speaking, theaters, animals) Subtle behavioral avoidance Avoidance of eye contact (e.g., wearing sunglasses) Avoidance of sensation-producing activities (e.g., physical exertion, caffeine, hot rooms) Avoidance of “contaminated” objects (e.g., sinks, toilets, doorknobs, money) Perfectionistic behavior at work or home Procrastination (avoidance of emotionally salient tasks) Repetitive or ritualistic behaviors Compulsive acts (e.g., excessive checking, cleaning) II. Cognitive and Emotional Avoidance Cognitive avoidance/escape Distraction (e.g., reading a book, watching television) Avoidance of thoughts or memories about trauma Effort to prevent thoughts from coming into mind Worry Rumination Thought suppression Safety signals Carrying a cell phone Holding onto “good luck” charms Carrying water or empty medication bottles Having reading material always on hand Carrying self-protective materials (e.g., mace, siren)

Disorder Often Associated

PD/A, SOC, SPEC SOC PD/A OCD GAD, SOC, OCD GAD, DEP OCD

GAD, DEP, PD/A PTSD OCD, PTSD GAD DEP All disorders PD/A, GAD OCD PD/A SOC, GAD PTSD

Note. PD/A = panic disorder with/without agoraphobia; SOC = social anxiety disorder; SPEC = specific phobia; OCD = obsessive-compulsive disorder; GAD = generalized anxiety disorder; DEP = depressive disorders (e.g., major depression, dysthymic disorder); PTSD = posttraumatic stress disorder.

A Proposal for a Dimensional Classification System Based on the Shared Features 343

Having established the presence and dimensional severity of AN, BA/P, and avoidance behavior, it then becomes important to examine the functional relationships of AN with specific facets of emotional expression or other experiences that have traditionally been considered as “key features” of existing criteria sets. As noted above, for GAD, stressful life events trigger the diathesis of AN to produce a clinical syndrome. While GAD is simply a pathologically strong expression of AN, depression (e.g., MDD, DYS), which is characterized by cognitive and motor slowing, hopelessness, and marked anhedonia, emerges from AN in combination with (low) BA/P when activated by life stressors (cf. Mineka et al., 1998). Mania (and hypomania), on the other hand, emerges from high BA/P when triggered by life stressors. For the anxiety disorders other than GAD, the third (specific) vulnerability is activated in which AN, through a process of learning, is specifically associated with an object or event. This process results in one or more key diagnostic features that become the principal focus or foci of AN. These major features are described next. The occurrence of discrete eruptions of emotion leads to several common manifestations of key features (Barlow, 2002). One manifestation is evident in the form of an ethologically primitive sympathetic nervous system surge known as the flight/fight response, which is the action tendency of the basic emotion of fear. When fear is triggered by relatively non-dangerous internal or external cues and therefore occurs at an inappropriate time, the experience is called a panic attack. The fact that panic attacks subsequently may become a major focus of AN has been explicated in considerable detail in the case of PD/A (Bouton, Mineka, & Barlow, 2001; White & Barlow, 2002). But the same phenomenon has been demonstrated to occur in PTSD where it is referred to as “flashbacks” (Brown & McNiff, in press; Jones & Barlow, 1990). We refer to the object of this focus as panic and related autonomic surges. Somatic symptoms or experiences can become functionally related to AN and may become a second focus. For example, in the case of hypochondriasis, certain unexplained physical symptoms are associated in a chronic manner with potential threatening illness or disease. Similarly, sensations of unreality, if they become the focus of AN, can evolve into a prominent feature of PTSD and depersonalization disorder. Apprehension about the physical symptoms of anxiety and panic attacks is a key feature of PD/A (Clark, 1986; Taylor, 1999; White & Barlow, 2002). We refer to this focus as somatic anxiety. A third common focus of AN is ego-dystonic intrusive cognitions. Although most characteristic of OCD and body dysmorphic disorder (BDD), studies have shown that these types of egodystonic intrusive thoughts are fairly common in the normal population as a function of stressful life events (e.g., Fullana et al., 2009; Parkinson & Rachman, 1981; Rachman & de Silva, 1978), but can reach substantial levels of severity in the context of a wide range of emotional disorders even in the absence of a diagnosis of OCD or BDD. Thus, it is common to see intrusive ego-dystonic thoughts of, for example, harming one’s children in cases where the principal diagnosis is PD/A or GAD, and these are associated with varying amounts of AN (Barlow, 2002). A fourth dimension could be described as social evaluation. Once again, although typically associated with SOC where it becomes the key diagnostic

344 Timothy A. Brown and David H. Barlow

feature, it is common for this focus to be represented on a dimension of severity across most emotional disorders (e.g., Rapee, Sanderson, & Barlow, 1988). A fifth focus of AN is past trauma with all of its ramifications. When the focus is on past trauma, it also quite common, as noted above, to experience sympathetic surges in the form of flashbacks triggered by internal or external cues originally associated with the trauma. But the focus of AN on past traumatic experience in those individuals already vulnerable to the expression of AN is the key feature of the disorder (Keane & Barlow, 2002; Steenkamp, McLean, Arditte, & Litz, in press). Thus, it becomes essential through a functional analysis to determine the focus of AN. One would accomplish this by first ascertaining the severity of AN along with the particular key features that might be the focus of AN to varying degrees. Following past precedent, the focus would correspond to current DSM-IV categories, but one could also determine the extent to which other key features are present on dimensions of severity that would include intensity, frequency, and distress, thereby providing potentially important information for treatment planning and prediction of course. In this way, information that is currently discarded in the system of prototypes that comprises DSM-IV (unless patients happen to meet all of the criteria for a comorbid diagnosis) would be saved and incorporated into the clinical picture. In sum, AN, and BA/P, as well as types of avoidant strategies, would be rated on severity dimensions for each individual presenting with emotional disorders. If severe enough or exacerbated by stress, these diatheses present as GAD and, in the presence of low BA/P, unipolar depression (e.g., MDD, DYS). High BA/P is, of course, uniquely associated with bipolar disorder. It is also important to determine any foci of anxiety and avoidance. An example of a recent patient presenting to our Center, whose presenting clinical symptoms are organized in this way, is presented below. Case Example

The patient was a male high school teacher in his mid-fifties, who was referred to the Center after worsening symptoms following a severe car accident some four months previously. Symptoms included intrusive memories of the crash, images of his wife’s injured face, very clear flashbacks to certain aspects of the accident itself, accompanied by some memory difficulties for other aspects of the accident, and hypervigilance to his surroundings as well as a strong startle reaction to cues relating to the accident. These symptoms intermingled with a similar set of symptoms emerging from multiple traumatic events experienced during the Vietnam War. In addition, he reported spending a great deal of the day worrying about a variety of life events, including his own health and that of his family, as well as his performance at work. He noted that he worried approximately one-third of the day and had been like this since returning from the war. He had previously sought treatment after suffering an episode of severe grief following the death of his parents some 15 years earlier. He continued to exhibit relatively mild symptoms of depression. In addition, he was concerned about interacting inappropriately with his students and that he would be evaluated badly by other staff members leading to dismissal from his teaching job.

A Proposal for a Dimensional Classification System Based on the Shared Features 345

On the basis of this clinical presentation, the only DSM-IV diagnosis formally assigned was PTSD. Although the patient met all the key and associated features of GAD, this diagnosis could not be assigned if adhering to a DSM-IV hierarchy rule (i.e., the disturbance occurred during the course of PTSD). In addition, whereas mild depression including some lethargy was part of the clinical picture, the patient’s symptoms were not sufficiently severe to meet the diagnostic threshold for a unipolar mood disorder such as MDD or DYS. In Figure 16.2, we present the patient’s diagnostic profile using the scheme described above. As seen in this figure, the patient presented with high levels of AN and moderately low BA/P. Behavioral avoidance was also elevated as reflected in avoidance of driving in certain locations, driving long distances, or engaging in activities or conversations associated with war. Although avoidance of intense emotion (blunting) and interoceptive cues were present for a time after the war, this type of cognitive/emotional avoidance had lessened considerably over the years. The foci of anxiety were primarily on past trauma, along with elevations for panic-related autonomic surges (which in this case represented flashbacks) and somatic anxiety. Because all intrusive thoughts were related to trauma, ego-dystonic intrusive thoughts were largely absent. Anxiety focused on social evaluation was present at moderate intensity in the context of evaluation at work. Accordingly, this dimensional profile provides a more complete portrayal of the patient’s clinical presentation relative to a single categorical diagnosis of PTSD. The profile captures the dimensional severity of the key features of PTSD (i.e., focus on past trauma, panic-related autonomic surges, avoidance), in addition to clinical features that are often associated and functionally related to PTSD but not recognized by the DSM-IV definition of this disorder (e.g., temperament, somatic anxiety). Moreover, although GAD could not be assigned under the DSM-IV framework, the profile recognizes the presence of high AN that may be relevant to prognosis and treatment planning. A sub-diagnostic threshold level of depression is also reflected that would be missed by DSM-IV diagnosis but which may be nonetheless clinically salient. Notably, while providing a richer depiction of the patient’s presenting symptoms and their severity, the dimensional profile avoids the various diagnosis decision rules (e.g., diagnostic hierarchy, temporal overlap, symptom counts, determination of whether symptoms are better accounted for by another condition) that are a common sources of DSM-IV diagnostic unreliability and that often result in non-veridical patterns of diagnostic comorbidity. Transdiagnostic Treatment Protocols

This diagnostic conception also has implications for treatment. Elsewhere, we have described the conceptual underpinnings of a unified, transdiagnostic treatment for emotional disorders (Barlow et. al., 2004; Fairholme, Boisseau, Ellard, Ehrenreich, & Barlow, in press; Moses & Barlow, 2006). These developments arise from work over the last few decades that has empirically supported a number of psychological treatments targeting individual anxiety and mood disorders, usually organized around specific DSM-IV criteria sets (Barlow, 2008; Nathan & Gorman, 2007). However, the aforementioned

346 Timothy A. Brown and David H. Barlow

conceptualizations of diagnosis and psychopathology emphasize commonalities rather than their differences among these disorders. In addition, treatment outcome research points to generalization of treatment response from a treated disorder (such as PDA) across other comorbid anxiety and mood disorders (e.g., Brown et al., 1995). Careful analysis also highlights a common set of therapeutic operations across these disparate protocols (Barlow et. al., 2004). Thus, research lends support to a unified, transdiagnostic approach to emotional disorders that considers these commonalities and distills common therapeutic procedures across evidenced-based protocols. At the same time, research on emotion regulation has provided a novel perspective on the treatment of anxiety and mood disorders, which are essentially disorders of emotion (Barlow, 1991, 2002). It is now clear that emotion regulation, or how individuals influence the occurrence, intensity, expression, and experience of emotions, plays an important role in phenomenology of anxiety and mood disorders (Campbell-Sills & Barlow, 2007; Gross, 2007). Deficits in emotion regulation are found in each of these disorders, as individuals with anxiety and mood disorders often use maladaptive regulation strategies that contribute to the persistence of symptoms. Thus, increased attention to emotional dysregulation in treatment is grounded in current research findings and is consistent with emerging conceptualizations emphasizing underlying commonalities among emotional disorders. Based on these advances, we have developed a treatment designed to be applicable to all anxiety and unipolar mood disorders, and possibly other disorders with strong emotional components (e.g., many somatoform and

Figure 16.2 Example Profile of Patient Evaluated with a Dimensional Classification System. Note: AN = anxiety-neuroticism; BA/P = behavioral activation-positive affect; DEP = unipolar depression; MAN = mania; SOM = somatic anxiety; PAS = panic and related autonomic surges; IC = intrusive cognitions; SOC = social evaluation; TRM = past trauma; AV-BI = behavioral and interoceptive avoidance, AV-CE = cognitive and emotional avoidance; higher scores on the Y-axis (0–100) indicate higher levels of the X-axis dimension, but otherwise the Y-axis metric is arbitrary and used for illustrative purposes.

A Proposal for a Dimensional Classification System Based on the Shared Features 347

dissociative disorders) that addresses core affective properties rather than narrowly construed DSM-IV key features. The unified protocol capitalizes on the contributions made by cognitive-behavioral theorists by distilling and incorporating the common principles among existing evidenced based psychological treatments such as restructuring maladaptive cognitive appraisals, changing action tendencies associated with the disordered emotion, preventing emotional avoidance, and utilizing emotional exposure procedures. It also draws on the innovations from the field of emotion science, incorporating and addressing the deficits in emotion regulation common in emotional disorders. Because this transdiagnostic intervention addresses core psychopathological features that cut across the DSM-IV disorders, the proposed nosological scheme described above facilitates a more comprehensive conceptualization and assessment of temperaments and associated foci of AN. Some evidence suggests that changes in AN may be substantial in some patients during successful psychological treatment (e.g., Brown, 2007), and this change may be associated with the widely noted outcome of enduring treatment effects from successful psychological treatment relative to drug treatment (e.g., Barlow et. al., 2000). Furthermore, we have previously reported that generalized improvement in comorbid emotional disorders after treatment of the principal disorder is not necessarily sustained over time, even if improvement in the “target” disorder is sustained (Brown et al., 1995). Transdiagnostic treatment and comprehensive conceptualization and assessment may obviate the outcome of incomplete treatment and the necessity of repeated, sequential, and narrowly targeted courses of intervention for each DSM-IV disorder. Practical Considerations and Future Directions The goal of this paper was to outline our vision of how a dimensional classification system based on common features might look for the anxiety and mood disorders. Clearly, this clinical assessment system is more radical than a previous proposal to simply incorporate dimensional severity ratings into the existing DSM disorder criteria sets (cf. Brown & Barlow, 2005). While this assessment system possesses several advantages over comparatively modest dimensional alternatives (e.g., better capturing salient clinical features that would not be recognized when adhering to DSM differential diagnosis decision rules; evaluation of core dimensions of temperament), researchers would be faced with many challenges in the development and implementation of this approach. A key consideration is clinical utility and all its components (e.g., acceptability and proper use by clinicians; First, 2005; First et al., 2004). Based on practical considerations, it may be difficult to envision a diagnostic classification system that is fully devoid of nominal elements. For instance, it has been argued that categorical diagnoses have greater clinical utility for endeavors such as clinical communication among practitioners and medical record keeping (First, 2005). Thus, the question arises as to what level could a dimensional system best exist in the formal nosology. A less radical approach would entail using the dimensional system to complement categorical diagnosis. In other words, although the primary emphasis would remain on the extant DSM classification system, a dimensional profile such as the one illustrated in

348 Timothy A. Brown and David H. Barlow

Figure 16.2 could be used in tandem with categorical diagnoses to recognize important clinical features that would otherwise be ignored by DSM (e.g., core dimensions of temperament, symptoms subsumed by another condition). However, this intermediate position could be viewed as unsatisfactory because it would perpetuate many of the most serious practical and empirical drawbacks associated with the categorical system (e.g., high comorbidity and modest reliability of disorders). Akin to proposals for the personality disorders (e.g., Widiger, Costa, & McCrae, 2002), a bolder approach would entail a nosological system that is driven by dimensional assessment and classification. In this scenario, a profile such as the one illustrated in Figure 16.2 would be generated for a patient via the administration of dimensional measures. Subsequently, categorical diagnostic labels, guided by empirically derived cut-points, would be assigned that characterize the patient’s dimensional profile (along the lines of the methods used in the MMPI). If a viable assessment system was developed, this approach would reconcile many common sources of diagnostic unreliability as well as the problem of high or distorted disorder comorbidity. But considerable work would be required to develop and validate an assessment system that is accepted and used in clinical and research settings alike. While the constructs in Figure 16.2 are firmly established in the empirical literature and are well-known by most practitioners, they are currently assessed by a disparate array of clinician-administered and self-report measures. To be implemented feasibly in clinical practice, a single multidimensional measure of temperament, mood, avoidance, and foci of AN would have to be developed and validated. Another practical issue is that the dimensional system would require evaluation of the full spectrum of emotional disorder psychopathology. This would be a distinct departure from the manner by which diagnoses are currently made in clinical practice (i.e., determination of whether a patient’s clinical presentation corresponds to the criteria set for a specific DSM disorder). To avoid unreasonable time demands on the clinician, a greater emphasis on self-report may be required to assess each component of the dimensional profile (cf. Widiger et al., 2002). In addition to its own practical issues (e.g., cost, availability, time demand on patient), self-report assessment is often less capable than clinical interviewing in capturing some aspects of a clinical presentation such as the duration, temporal sequence, and functional relationships of symptoms. However, whereas these aspects are germane to the DSM (e.g., determination of whether the symptoms of a disorder are better accounted for or occur during another disorder), they would be less relevant to a nosology based on dimensional profiling where the emphasis is on the patterning and severity of common features rather than differential or hierarchical diagnosis and symptom counts. Nevertheless, as has been repeatedly evidenced in the personality disorders literature (cf. Samuel & Widiger, 2006; Verheul, 2005), various concerns about clinical utility would likely be cited as primary reasons for why a dimensionally based classification system should not replace categorical DSM anxiety and mood diagnoses. These objections often pertain to concerns that clinicians would find a dimensional system unacceptable because they are unfamiliar

A Proposal for a Dimensional Classification System Based on the Shared Features 349

with the dimensions, they perceive the dimensional classification system as overly cumbersome, they believe it will hinder communicating clinical information to patients and mental health professionals, or that the dimensions could not be used to guide treatment decisions or predict clinical course and prognosis (First et al., 2004; Samuel & Widiger, 2006; Verheul, 2005). However, there is some evidence that these concerns may be unfounded. For instance, in a study by Samuel and Widiger (2006), 245 practicing psychologists evaluated personality disorder cases using both the dimensional five-factor model (FFM) and the DSM-IV and then rated both assessment systems on several aspects of clinical utility. Despite the fact that the clinicians indicated that they were much less familiar with the FFM than DSM-IV, ratings on the ease of application of these assessment systems did not differ. Interestingly, the FFM was consistently rated as significantly more useful than the DSM-IV assessment model on several dimensions of clinical utility (e.g., better at providing a global description of the patient’s personality, communicating information to patients, capturing all of the patient’s clinically salient personality difficulties, and assisting the clinician in developing effective treatments). It would be important to ascertain whether a dimensional classification system for anxiety and mood disorders is associated with comparable levels of clinical utility and acceptability found in the personality disorders literature (e.g., Blais, 1997; Samuel & Widiger, 2006; Sprock, 2003). Key to this endeavor would be the explication and validation of nominal cut-points or dimensional profiles necessary to inform clinical decisions (e.g., treatment planning and prognosis) and to facilitate communication among clinicians, patients, third-party payors, and so forth. Indeed, it might be argued that this empirical necessity is daunting to an extent that it undermines the potential realization of a dimensional system. However, as has been detailed in this article, one must be mindful of the fact that the DSM’s categorical diagnostic thresholds for mental disorders (i.e., the cut-point indicating the presence/absence of psychopathology) were set primarily by panel consensus (i.e., the various DSM workgroups) but are not grounded by strong empirical evidence. Indeed, the existence of a dimensional assessment system would launch the quest to establish empirically supported cut-points, an endeavor that has not been undertaken in DSM and perhaps which could not be compellingly addressed under this system (e.g., given the DSM’s reliance on nominal diagnoses and diagnostic criteria, cut-points in DSM are predetermined). Moreover, the possibility exists that this scientific venture would identify different sets of cut-points for a given dimension or dimensional profile to guide different clinical judgments, decisions, and predictions (e.g., inform the appropriate level or type of clinical intervention). This can only be accomplished through the analysis of the rich array of clinical information garnered by dimensional assessment, but which is lost or distorted by binary judgments of the presence or absence of psychopathology. Note The authors thank Ben Emmert-Aronson for his assistance with this paper.

350 Timothy A. Brown and David H. Barlow

References Andrews G. Classification of neurotic disorders. Journal of Royal Society of Medicine 1990; 83:606–607. Andrews G. Comorbidity in neurotic disorders: The similarities are more important than the differences. In: Rapee RM., editor. Current controversies in the anxiety disorders. New York: Guilford Press; 1996. p. 3–20. Barlow DH. Anxiety and its disorders: The nature and treatment of anxiety and panic. New York: Guilford Press; 1988. Barlow DH. Anxiety and its disorders: The nature and treatment of anxiety and panic. 2. New York: Guilford Press; 2002. Barlow DH. Disorders of emotion. Psychological Inquiry 1991; 2:58–71. Barlow DH. Unraveling the mysteries of anxiety and its disorders from the perspective of emotion theory. American Psychologist 2000; 55:1247–1263. [PubMed: 11280938] Barlow DH., editor. Clinical handbook of psychological disorders: A step-by-step treatment manual. 4. New York: Guilford Press; 2008. Barlow DH, Allen LB, Choate ML. Toward a unified treatment for emotional disorders. Behavior Therapy 2004; 35:205–230. Barlow DH, Gorman JM, Shear MK, Woods SW. Cognitive-behavioral therapy, imipramine, or their combination for panic disorder: A randomized controlled trial. Journal of American Medical Association 2000; 283:2529–2536. Blais MA. Clinician ratings of the five-factor model of personality and the DSM-IV personality disorders. Journal of Nervous and Mental Disease 1997; 185:388–393. [PubMed: 9205425] Bouton ME, Mineka S, Barlow DH. A modern learning theory perspective on the etiology of panic disorder. Psychological Review 2001; 108:4–32. [PubMed: 11212632] Brawman-Mintzer O, Lydiard RB, Emmanuel N, Payeur R, Johnson M, Roberts J, Jarrell MP, Ballenger JC. Psychiatric comorbidity in patients with generalized anxiety disorder. American Journal of Psychiatry 1993; 150:1216–1218. [PubMed: 8328567] Brown TA. Validity of the DSM-III-R and DSM-IV classification systems for anxiety disorders. In: Rapee RM, editor. Current controversies in the anxiety disorders. New York: Guilford Press; 1996. p. 21–45. Brown TA. Temporal course and structural relationships among dimensions of temperament and DSMIV anxiety and mood disorder constructs. Journal of Abnormal Psychology 2007; 116:313–328. [PubMed: 17516764] Brown TA, Antony MM, Barlow DH. Diagnostic comorbidity in panic disorder: Effect on treatment outcome and course of comorbid diagnoses following treatment. Journal of Consulting and Clinical Psychology 1995; 63:408–418. [PubMed: 7608353] Brown TA, Barlow DH. Comorbidity among anxiety disorders: Implications for treatment and DSMIV. Journal of Consulting and Clinical Psychology 1992; 60:835–844. [PubMed: 1460147] Brown TA, Barlow DH. Classification of anxiety and mood disorders. In: Barlow DH, editor. Anxiety and its disorders: The nature and treatment of anxiety and panic. 2. New York: Guilford Press; 2002. p. 292–327. Brown TA, Barlow DH. Categorical vs dimensional classification of mental disorders in DSM-V and beyond. Journal of Abnormal Psychology 2005; 114:551–556. [PubMed: 16351377] Brown TA, Barlow DH, Liebowitz MR. The empirical basis of generalized anxiety disorder. American Journal of Psychiatry 1994; 151:1272–1280. [PubMed: 8067480] Brown TA, Campbell LA, Lehman CL, Grisham JR, Mancill RB. Current and lifetime comorbidity of the DSM-IV anxiety and mood disorders in a large clinical sample. Journal of Abnormal Psychology 2001; 110:585–599. [PubMed: 11727948]

A Proposal for a Dimensional Classification System Based on the Shared Features 351 Brown TA, Chorpita BF, Barlow DH. Structural relationships among dimensions of the DSM-IV anxiety and mood disorders and dimensions of negative affect, positive affect, and autonomic arousal. Journal of Abnormal Psychology 1998; 107:179–192. [PubMed: 9604548] Brown TA, Di Nardo PA, Lehman CL, Campbell LA. Reliability of DSM-IV anxiety and mood disorders: Implications for the classification of emotional disorders. Journal of Abnormal Psychology 2001; 110:49–58. [PubMed: 11261399] Brown TA, McNiff J. Specificity of autonomic arousal to DSM-IV panic disorder and posttraumatic stress disorder. Behaviour Research and Therapy; in press. Campbell-Sills L, Barlow DH. Incorporating emotion regulation into conceptualizations and treatments of anxiety and mood disorders. In: Gross JJ, editor. Handbook of emotion regulation. New York: Guilford Press; 2007. p. 542–560. Campbell-Sills LA, Liverant G, Brown TA. Psychometric evaluation of the Behavioral Inhibition/ Behavioral Activation Scales (BIS/BAS) in large clinical samples. Psychological Assessment 2004; 16:244–254. [PubMed: 15456380] Carver CS, Johnson SL, Joormann J. Serotonergic function, two-mode models of self-regulation, and vulnerability to depression: What depression has in common with impulsive aggression. Psychological Bulletin 2008; 134:912–943. [PubMed: 18954161] Carver CS, White TL. Behavioral inhibition, behavioral activation, and affective responses to impending reward and punishment: The BIS/BAS scales. Journal of Personality and Social Psychology 1994; 67:319–333. Chorpita BF, Barlow DH. The development of anxiety: The role of control in the early environment. Psychological Bulletin 1998; 124:3–21. [PubMed: 9670819] Clark DM. A cognitive approach to panic. Behaviour Research and Therapy 1986; 24:461–470. [PubMed: 3741311] Clark LA. Temperament as a unifying basis for personality and psychopathology. Journal of Abnormal Psychology 2005; 114:505–521. [PubMed: 16351374] Clark LA, Vittengl J, Kraft D, Jarrett RB. Separate personality traits from states to predict depression. Journal of Personality Disorders 2003; 17:152–172. [PubMed: 12755328] Clark LA, Watson D, Mineka S. Temperament, personality, and the mood and anxiety disorders. Journal of Abnormal Psychology 1994; 103:103–116. [PubMed: 8040472] Costa PT, McCrae RR. Personality in adulthood: A six-year longitudinal study of self-reports and spouse ratings on the NEO Personality Inventory. Journal of Personality and Social Psychology 1988; 54:853–863. [PubMed: 3379583] Diener E, Seligman MEP. Very happy people. Psychological Science 2002; 13:81–84. [PubMed: 11894851] Di Nardo PA, Brown TA, Barlow DH. Anxiety Disorders Interview Schedule for DSM-IV: Lifetime Version (ADIS-IV-L). New York: Oxford University Press; 1994. Di Nardo PA, Moras K, Barlow DH, Rapee RM, Brown TA. Reliability of DSM-III-R anxiety disorder categories using the Anxiety Disorders Interview Schedule–Revised (ADIS-R). Archives of General Psychiatry 1993; 50:251–256. [PubMed: 8466385] Eley TC, Brown TA. Phenotypic and genetic/environmental structure of anxiety and depressive disorder symptoms in adolescence. 2009. Manuscript submitted for publication. Eysenck HJ. A model for personality. New York: Springer; 1981. Fairholme CP, Boisseau CL, Ellard KK, Ehrenreich JT, Barlow DH. Emotions, emotion regulation, and psychological treatment: A unified perspective. In: Kring AM, Sloan DM, editors. Emotion regulation and psychopathology. New York: Guilford Press; in press. Fanous A, Gardner CO, Prescott CA, Cancro R, Kendler KS. Neuroticism, major depression, and gender: A population-based twin study. Psychological Medicine 2002; 32:719–728. [PubMed: 12102386]

352 Timothy A. Brown and David H. Barlow First MB. Clinical utility: A prerequisite for the adoption of a dimensional approach in DSM. Journal of Abnormal Psychology 2005; 114:560–564. [PubMed: 16351379] First MB, Pincus HA, Levine JB, Williams JBW, Ustun B, Peele R. Clinical utility as a criterion for revising psychiatric diagnoses. American Journal of Psychiatry 2004; 161:946–954. [PubMed: 15169680] Fowles, DC. A motivational theory of psychopathology. In: Spaulding W, editor. Nebraska Symposium on Motivation: Integrated views of motivation, cognition, and emotion. Vol. 41. Lincoln, NE: University of Nebraska Press; 1994. p. 181–238. Fulker DW. The genetic and environmental architecture of psychoticism, extroversion, and neuroticism. In: Eysenck HJ., editor. A model for personality. New York: Springer-Verlag; 1981. p. 88–122. Fullana MA, Mataix-Cols D, Caspi A, Harrington H, Grisham JR, Moffitt TE, Poulton R. Obsessions and compulsions in the community: Prevalence, interference, help-seeking, developmental stability, and co-occurring psychiatric conditions. American Journal of Psychiatry 2009; 166:329–336. [PubMed: 19188283] Gershuny BS, Sher KJ. The relation between personality and anxiety: Findings from a 3-year prospective study. Journal of Abnormal Psychology 1998; 107:252–262. [PubMed: 9604554] Gilbert DT. Stumbling on happiness. New York: Knopf; 2006. Gray, J. A. (1987). The psychology of fear and stress (2nd ed.). Cambridge, England: Cambridge University Press. Gross JJ., editor. Handbook of emotion regulation. New York: Guilford Press; 2007. Gruber J, Johnson SL, Oveis C, Keltner D. Risk for mania and positive emotional responding: Too much of a good thing? Emotion 2008; 8:23–33. [PubMed: 18266513] Hettema JM, Prescott CA, Kendler KS. Genetic and environmental sources of covariation between generalized anxiety disorder and neuroticism. American Journal of Psychiatry 2004; 161:1581–1587. [PubMed: 15337647] Johnson SL, Gruber J, Eisner LR. Emotion and bipolar disorder. In: Rottenberg J, Johnson SL, editors. Emotion and psychopathology: Bridging affective and clinical science. Washington, DC: American Psychological Association; 2007. p. 123–150. Jones JC, Barlow DH. The etiology of posttraumatic stress disorder. Clinical Psychology Review 1990; 10:299–328. Kasch KL, Rottenberg J, Arnow BA, Gotlib IH. Behavioral activation and inhibition systems and the severity and course of depression. Journal of Abnormal Psychology 2002; 111:589–597. [PubMed: 12428772] Keane TM, Barlow DH. Posttraumatic stress disorder. In: Barlow DH., editor. Anxiety and its disorders: The nature and treatment of anxiety and panic. 2. New York: Guilford Press; 2002. p. 418–453. Kendell RE. The role of diagnosis in psychiatry. Oxford: Blackwell; 1975. Kendler KS, Neale MC, Kessler RC, Heath AC, Eaves LJ. Major depression and generalized anxiety disorder: Same genes, (partly) different environments? Archives of General Psychiatry 1992; 49:716–722. [PubMed: 1514877] Kessler RC, Gruber M, Hettema JM, Hwang I, Sampson N, Yonkers KA. Comorbid major depression and generalized anxiety disorders in the National Comorbidity Survey follow-up. Psychological Medicine 2008; 38:365–374. [PubMed: 18047766] Lawrence AE, Brown TA. Differentiating generalized anxiety disorder from anxiety disorder not otherwise specified. Journal of Nervous and Mental Disease; in press. Lawrence AE, Liverant GI, Rosellini AJ, Brown TA. Generalized anxiety disorder within the course of major depressive disorder: Examining the utility of the DSM-IV hierarchy rule. 2009. Manuscript submitted for publication.

A Proposal for a Dimensional Classification System Based on the Shared Features 353 MacCallum RC, Zhang S, Preacher KJ. On the practice of dichotomization of quantitative variables. Psychological Methods 2003; 7:19–40. [PubMed: 11928888] Mannuzza S, Fyer AJ, Martin LY, Gallops MS, Endicott J, Gorman JM, Liebowitz MR, Klein DF. Reliability of anxiety assessment: I. Diagnostic agreement. Archives of General Psychiatry 1989; 46:1093–1101.9. [PubMed: 2589923] Maser JD, Cloninger CR, editors. Comorbidity of mood and anxiety disorders. Washington, DC: American Psychiatric Press; 1990. Meyer B, Johnson SL, Winters R. Responsiveness to threat and incentive in bipolar disorder: Relations of the BIS/BAS scales with symptoms. Journal of Psychopathology and Behavioral Assessment 2001; 23:133–143. Mineka S, Watson D, Clark LA. Comorbidity of anxiety and unipolar mood disorders. Annual Review of Psychology 1998; 49:377–412. Moras K, Clark LA, Katon W, Roy-Byrne P, Watson D, Barlow DH. Mixed anxiety-depression. In: Widiger TA, Frances AJ, Pincus HA, Ross R, First MB, Davis WW, editors. DSM-IV sourcebook. Washington DC: American Psychiatric Association; 1996. p. 623–643. Moses EB, Barlow DH. A new unified treatment approach for emotional disorders based on emotion science. Current Directions in Psychological Science 2006; 15:146–150. Nathan PE, Gorman JM., editors. A guide to treatments that work. New York: Oxford University Press; 2007. Parkinson L, Rachman S. Intrusive thoughts: The effects of an uncontrived stress. Advances in Behaviour Research and Therapy 1981; 3:111–118. Rachman S, de Silva P. Abnormal and normal obsessions. Behaviour Research and Therapy 1978; 16:233–248. [PubMed: 718588] Rapee RM, Sanderson WC, Barlow DH. Social phobia features across the DSM-III-R anxiety disorders. Journal of Psychopathology and Behavioral Assessment 1988; 10: 287–299. Samuel DB, Widiger TA. Clinician’s judgments of clinical utility: A comparison of the DSM-IV and five-factor models. Journal of Abnormal Psychology 2006; 115:298–308. [PubMed: 16737394] Sanderson WC, Di Nardo PA, Rapee RM, Barlow DH. Syndrome comorbidity in patients diagnosed with a DSM-III-R anxiety disorder. Journal of Abnormal Psychology 1990; 99:308–312. [PubMed: 2212281] Sprock J. Dimensional versus categorical classification of prototypic and nonprototypic cases of personality disorder. Journal of Clinical Psychology 2003; 59:991–1014. [PubMed: 12945064] Steenkamp M, McLean CP, Arditte KA, Litz BT. Assessment of adults exposed to trauma. In: Antony MM, Barlow DH., editors. Handbook of assessment, treatment planning, and outcome evaluation: Empirically supported strategies for psychological disorders. New York: Guilford Press; in press. Suárez L, Bennett S, Goldstein C, Barlow DH. Understanding anxiety disorders from a “triple vulnerabilities” framework. In: Antony, MM, Stein, MB, editors. Oxford handbook of anxiety and related disorders. New York: Oxford; 2009. p. 153–172. Taylor S. Anxiety sensitivity: Theory, research, and treatment of the fear of anxiety. Mahwah, NJ: Erlbaum; 1999. Tellegen A. Structures of mood and personality and their relevance to assessing anxiety with an emphasis on self-report. In: Tuma AH, Maser JD., editors. Anxiety and the anxiety disorders. Hillsdale, NJ: Erlbaum; 1985. p. 681–706. Tellegen A, Lykken DT, Bouchard TJ, Wilcox KJ, Segal NL, Rich S. Personality similarity in twins reared apart and together. Journal of Personality and Social Psychology 1988; 54:1031–1039. [PubMed: 3397862]

354 Timothy A. Brown and David H. Barlow Verheul R. Clinical utility of dimensional models for personality pathology. Journal of Personality Disorders 2005; 19:283–302. [PubMed: 16175737] Viken RJ, Rose RJ, Kaprio J, Koskenvuo M. A developmental genetic analysis of adult personality: Extraversion and neuroticism from 18 to 59 years of age. Journal of Personality and Social Psychology 1994; 66:722–730. [PubMed: 8189349] Watson D, Clark LA. Negative affectivity: The disposition to experience aversive emotional states. Psychological Bulletin 1984; 96:465–490. [PubMed: 6393179] Watson D, Clark LA, Carey G. Positive and negative affectivity and their relation to the anxiety and depressive disorders. Journal of Abnormal Psychology 1988; 97:346–353. [PubMed: 3192830] White KS, Barlow DH. Panic disorder and agoraphobia. In: Barlow DH, editor. Anxiety and its disorders: The nature and treatment of anxiety and panic. New York: Guilford Press; 2002. p. 328–379. Widiger TA. Categorical versus dimensional classification: Implications from and for research. Journal of Personality Disorders 1992; 6:287–300. Widiger TA, Costa PT, McCrae RR. A proposal for Axis II: Diagnosing personality disorders using the five-factor model. In: Costa PT, Widiger TA., editors. Personality disorders and the five-factor model of personality. Washington, DC: American Psychological Association; 2002. p. 431–456. Widiger TA, Samuel DB. Diagnostic categories or dimensions: A question for DSM-V. Journal of Abnormal Psychology 2005; 114:494–504. [PubMed: 16351373] Widiger TA, Verheul R, van den Brink W. Personality and psychopathology. In: Pervin LA, John OP., editors. Handbook of personality: Theory and research. New York: Guilford Press; 1999. p. 347–366. Williams JBW, Gibbon M, First MB, Spitzer RL, Davies M, Borus J, Howes MJ, Kane J, Pope HG, Rounsaville B, Wittchen H. The Structured Clinical Interview for DSM-IIIR (SCID) II. Multisite test-retest reliability. Archives of General Psychiatry 1992; 49:630–636.2. [PubMed: 1637253] Zimmerman M, Chelminski I. Generalized anxiety disorder in patients with major depression: Is DSMIV’s hierarchy correct? American Journal of Psychiatry 2003; 160:504–521. [PubMed: 12611832] Zinbarg RE, Barlow DH, Liebowitz M, Street L, Broadhead E, Katon W, et al. The DSMIV field trial for mixed anxiety-depression. American Journal of Psychiatry 1994; 151:1153–1162. [PubMed: 8037250]

Article 17

The Nature, Diagnosis, and Treatment of Neuroticism: Back to the Future David H. Barlow 1, Shannon Sauer-Zavala 1, Jenna R. Carl 1, Jacqueline R. Bullis 1, and Kristen K. Ellard 2,3 Department of Psychology, Boston University; 2Massachusetts General Hospital, Boston, Massachusetts; and 3Harvard Medical School

1

We highlight the role of neuroticism in the development and course of emotional disorders and make a case for shifting the focus of intervention to this higher-order dimension of personality. Recent decades have seen great emphasis placed on differentiating disorders into Diagnostic and Statistical Manual of Mental Disorders diagnoses; however, evidence has suggested that splitting disorders into such fine categories may be highlighting relatively trivial differences. Emerging research on the latent structure of anxiety and mood disorders has indicated that trait neuroticism, cultivated through genetic, neurobiological, and psychological factors, underscores the development of these disorders. We raise the possibility of a new approach for conceptualizing these disorders—as emotional disorders. From a service-delivery point of view, we explore the possibility that neuroticism may be more malleable than previously thought and may possibly be amenable to direct intervention. The public-health implications of directly treating and even preventing the development of neuroticism would be substantial. Throughout history, philosophers and researchers have attempted to understand the nature of anxiety and its disorders. As early as 450 BC, the relationship between an individual’s discrete emotions in response to stressors and his or her enduring proclivity for such experiences was of interest. For example, the Roman philosopher Cicero proposed a theory of emotion that distinguished between the temporary emotional state of anxiety (angor) and the enduring tendency to experience anxiety (anxi-etas; Lewis, 1970). Furthermore, in perhaps one of the best-known historical conceptualizations of anxiety, Freud (1924) distinguished between objective anxiety signaling the presence of an immediate threat (e.g., a loaded gun pressed to the temple) and neurotic anxiety generated when an individual’s defense mechanisms are no longer able to successfully repress an early traumatic experience, which results in a persistent state of distress (Reiss, 1997). In this article, we review a variety of research areas with the purpose of shifting focus from

356 David H. Barlow, et al.

the study of discrete anxiety, mood, and related disorders as defined in the Diagnostic and Statistical Manual of Mental Disorders (5th ed.; DSM-5; American Psychiatric Association, APA, 2013) to a broader understanding of emotional or internalizing disorders. In particular, we seek to call attention to higher-order temperamental factors that may be a more appropriate target for assessment and intervention than may symptom-level manifestations of these traits. Although relevant higher-order temperamental dimensions associated with the experience of frequent and intense negative emotions have acquired a number of labels, we focus on the construct of “neuroticism” as the first (Eysenck, 1947) and still most popular (Lahey, 2009) conception of this trait. Neuroticism is typically defined as the tendency to experience frequent and intense negative emotions in response to various sources of stress. These negative emotions are usually broadly construed to include anxiety, fear, irritability, anger, sadness, and so forth, but the greatest focus has been on the experience of anxious or depressive mood. Accompanying this exaggerated negative emotionality is the pervasive perception that the world is a dangerous and threatening place, along with beliefs about one’s inability to manage or cope with challenging events. These beliefs often are manifested in terms of heightened focus on criticism, either self-generated or from others, as confirming a general sense of inadequacy and perceptions of lack of control over salient events (Barlow, 2002; L. A. Clark & Watson, 2008; Eysenck, 1947; Goldberg, 1993). The label neuroticism was first used by Eysenck (1947) to describe this statistically derived personality trait; Eysenck coined the term after the commonly used clinical label at that time: neurosis—the early DSM category comprising people with anxiety, depression, and related disorders. Building on earlier work by Slater (1943), who used the term neurotic constitution, Eysenck, no fan of things Freudian even in the 1940s, regretted employing this term and noted that “it would no doubt be preferable in some ways to use a more neutral kind of label” (p. 49). Nevertheless, he made it clear that individuals with the diagnosis of neurosis occupied the pathological extreme of the personality trait of neuroticism. Eysenck also noted, in a prescient bit of writing, that these two constructs could be relatively independent and that “some so called ‘neurotic’ inmates (of a hospital) show very little evidence of the ‘neurotic constitution’ and would likely be situated rather towards the normal end of the distribution” (p. 48). He went on to theorize that these individuals would have been subjected to extreme life stress, whereas individuals high on the trait of neuroticism would require relatively little life stress to trigger neurosis, thereby clearly presaging the stress-diathesis model of psychopathology. Although the study of trait or temperament models of anxiety and related negative emotions has continued for decades, this literature has had decreasing influence on nosological schemes. DSM-III (3rd ed.; APA, 1980) represented the advent of an objective, empirically based classification system for mental disorders; patients previously receiving a diagnosis of neurosis were now classified more narrowly into specific anxiety, depressive, and related disorders. These new discrete diagnostic categories indeed had a meaningful

The Nature, Diagnosis, and Treatment of Neuroticism 357

impact on the development of pharmacological and individual psychological treatments, particularly for anxiety and mood disorders (e.g., Barlow et al., 1984). This approach allowed for the development of interventions targeting specific forms of psychopathology and the ability to evaluate treatment response on the basis of discrete outcomes. Although this “splitting” approach to nosology ensured high rates of diagnostic reliability, there is evidence that it may have come at the expense of validity; in other words, the current diagnostic system may be overemphasizing categories that are minor variations of a broader underlying syndrome (T. A. Brown & Barlow, 2005, 2009). Neuroticism, combined with other temperamental variables, likely represents this syndrome. Despite the marked emphasis during the past 30 years on discrete DSM-based categories of emotional disorders and their treatment, a few investigators have continued to focus attention on the existence and salience of broader underlying syndromes. For example, Andrews (1990) and Tyrer (1989) each argued for the existence of a “general neurotic syndrome” as a more parsimonious and heuristic account of emotional disorders. In addition, Lahey (2009), summarizing the growing evidence for the public-health significance of neuroticism, documented that neuroticism is strongly associated with and predicts many different mental and physical disorders, as well as treatment seeking for not only mental disorders but also general health services. Indeed, he underscored evidence that neuroticism may act as a salient predictor of the quality and longevity of our lives. Lahey called for a more substantial focus on the nature and origins of neuroticism and the mechanisms through which this trait is linked to both mental and physical disorders. Moreover, Cuijpers et al. (2010) examined the economic cost of the trait of neuroticism and found, in a large representative sample of the Dutch population, that these economic costs were enormous and exceeded costs of common mental disorders. Cuijpers et al. noted that “we should start thinking about interventions that focus not on each of the specific negative outcomes of neuroticism but rather on the starting point itself ” (p. 1086). The current review begins with a section highlighting commonalities among the emotional disorders, including high rates of comorbidity, broad rather than narrow treatment responsiveness among comorbid disorders, and shared neurobiological mechanisms. Next, we present the emerging research on the latent structure of emotional disorders that may underlie these observed commonalities followed by a brief description of theoretical accounts of the origins of neuroticism or trait anxiety. The following section describes common functional relationships in emotional disorders, particularly among emotional expression, negative appraisals under stress, and avoidance, as well as new approaches to conceptualizing these disorders and suggests the possibility of more satisfying dimensional nosological and assessment schemes. Finally, a review of diverse research on the malleability of neuroticism in both normal and pathological expressions sets the stage for a discussion of possible new strategies for treatment and prevention that focus not on individual negative outcomes of neuroticism in the form of the DSM-defined emotional disorders but, rather, on neuroticism itself as well as related temperaments.

358 David H. Barlow, et al.

Commonalities and Dimensions Among Anxiety and Related Disorders Empirical conceptions of the anxiety and major emotional disorders are emerging that underscore their commonalities (Barlow, 2002; T. A. Brown, 2007; T. A. Brown & Barlow, 2009). Major developments in at least three areas—high rates of comorbidity, broad treatment response across comorbid disorders, and shared neuro-biological mechanisms—support this conception. First, studies of phenomenology and nosology with a particular focus on comorbidity have suggested considerable overlap among disorders. At the diagnostic level, this overlap is most evident in high rates of current and lifetime comorbidity (e.g., Allen et al., 2010; T. A. Brown, Campbell, Lehman, Grisham, & Mancill, 2001; Kessler, Berglund, & Demler, 2003; Kessler et al., 1996; Kessler et al., 1998; Kessler et al., 2008). In a study of 1,127 patients presenting at the Center for Anxiety and Related Disorders (CARD) at Boston University, 55% of patients with a principal anxiety disorder had at least one additional anxiety or depressive disorder at the time of assessment, and this rate increased to 76% when lifetime diagnoses were considered (T. A. Brown et al., 2001). To take one example, of 324 patients diagnosed with DSM-IV-TR (4th ed., text rev.; APA, 2000) panic disorder with or without agoraphobia, 60% met criteria for an additional anxiety or mood disorder, breaking down to 47% with an additional anxiety disorder and 33% with an additional mood disorder. When lifetime diagnoses were considered, the percentages rose to 77% of the patients experiencing any comorbid anxiety or mood disorder, breaking down to 56% for an additional anxiety disorder and 60% for a mood disorder. The principal diagnostic categories of posttraumatic stress disorder and generalized anxiety disorder were associated with the highest comorbidity rates. Similarly, Merikangas, Zhang, and Aveneoli (2003) followed approximately 500 individuals for 15 years and found that relatively few people suffered from anxiety or depression alone; when patients did meet criteria for a single disorder at one point in time, an additional anxiety or depressive episode disorder almost certainly emerged at a later time. These summaries are most likely conservative as a result of artifactual constraints in DSM-IV-TR (which were continued in DSM-5), such as the nature of inclusion-exclusion criteria used. For instance, when adhering strictly to DSM-IV-TR diagnostic rules, the comorbidity between dysthymia and generalized anxiety disorder in the T. A. Brown et al. (2001) study was 5%. However, when the hierarchical rule that generalized anxiety disorder should not be assigned when occurring exclusively during a course of a mood disorder was suspended, the comorbidity estimate increased to 90%. These data also ignore the presence of subthreshold symptoms that did not meet diagnostic thresholds for one disorder or another. Second, psychological treatments for a given anxiety disorder often produce improvement in additional comorbid anxiety or mood disorders that are not specifically addressed in treatment (Allen et al., 2010; Borkovec, Abel, & Newman, 1995; T. A. Brown, Antony, & Barlow, 1995; Tsao, Lewin, & Craske, 1998; Tsao, Mystkowski, & Zucker, 2002). For example, we examined the course of additional diagnoses in a sample of 126 patients who were being treated at CARD for panic disorder with or without agoraphobia (T. A. Brown et al., 1995).

The Nature, Diagnosis, and Treatment of Neuroticism 359

A significant pre- to posttreatment decline in overall comorbidity was noted (40% to 17%, respectively). This effect could represent the generalization of elements of treatment to independent facets of both disorders or the effective targeting of “core” features of emotional disorders. The fact that a wide range of emotional disorders (e.g., major depressive disorder, obsessive-compulsive disorder, panic disorder) respond approximately equivalently to antidepressant medications has also been interpreted by some researchers as indicating shared features among these disorders (e.g., Hudson & Pope, 1990). There are several possible explanations for high rates of comorbidity and overlapping treatment response that we have reviewed extensively elsewhere (T. A. Brown & Barlow, 2002, 2009), including overlapping definitional criteria, artifactual reasons (e.g., differential base rates of occurrence in our setting), and the possibility that disorders are sequentially related such that the features of one disorder act as risk factors for another disorder. Another more intriguing explanation, noted earlier, is that this pattern of comorbidity argues for the existence of what has been called a general neurotic syndrome (Andrews, 1990, 1996; Tyrer, 1989). Under this conceptualization, heterogeneity in the expression of emotional disorder symptoms (e.g., individual differences in the prominence of social anxiety, panic attacks, anhedonia) is regarded as trivial variation in the manifestation of a broader syndrome. We return to this argument later. Third, recent research from affective neuroscience has suggested the existence of a biological syndrome that is common across emotional disorders. For example, research among individuals with anxiety and related disorders has suggested that hyperexcitability of limbic structures, along with limited inhibitory control by cortical structures, may be one explanation for the increased negative emotionality among individuals with such diagnoses (Etkin & Wager, 2007; Mayberg et al., 1999; Porto et al., 2009; Shin & Liberzon, 2010). Specifically, increased “bottom-up” processing through amygdala overactivation, coupled with inefficient or deregulated cortical inhibition of amygdala responses, has been shown in studies of social anxiety disorder (Lorberbaum et al., 2004; Phan, Fitzgerald, Nathan, & Tancer, 2006; Tillfors, Furmark, Marteinsdottir, & Fredrikson, 2002), posttraumatic stress disorder (Shin et al., 2005), generalized anxiety disorder (Etkin, Prater, Hoeft, Menon, & Schatzberg, 2010; Hoehn-Saric, Schlund, & Wong, 2004; Paulesu et al., 2010), specific phobia (Paquette et al., 2003; Straube, Mentzel, & Miltner, 2006), and depression (Holmes et al., 2012). This same amygdala overactivation has also been found in individuals high in the personality dimension of neuroticism (Keightley et al., 2003). Of course, a few unique and idiosyncratic neurobiological factors have also been associated with discrete DSM-IV-TR diagnoses (Blair et al., 2008; Chorpita, Albano, & Barlow, 1998), but it seems likely that the broader based genetic and neurobiological commonalities reviewed earlier may better account for the nature of emotional disorders. Although these three commonalities among emotional disorders have garnered attention, a recent focus on the hierarchical structure of emotional disorders may be more heuristic. Latent Temperamental Structure of Emotional Disorders Emerging research on the latent dimensional features of emotional disorders has revealed a hierarchical structure that places emphasis on two genetically

360 David H. Barlow, et al.

based core dimensions of temperament: neuroticism and, to a lesser degree, extraversion (Barlow, 2002). Extraversion, commonly associated with social qualities, broadly refers to an energetic approach to the world, including activity and positive emotionality in addition to sociability. Although these traits have received various labels, including negative affect, behavioral inhibition, trait anxiety, and harm avoidance as alternate terms for neuroticism and positive affect or behavioral activation as alternate terms for extraversion, substantial existing literature has underscored the roles of these constructs in accounting for the onset, overlap, and maintenance of anxiety, depressive, and related disorders (T. A. Brown, 2007; T. A. Brown & Barlow, 2009; T. A. Brown, Chorpita, & Barlow, 1998; Gershuny & Sher, 1998; Griffith et al., 2010; Kasch, Rottenberg, Arnow, & Gotlib, 2002; Kessler et al., 2011; Krueger, 1999; Watson, Clark, & Carey, 1988). The study of trait or temperament models of anxiety and related negative emotions has been ongoing for decades in spite of the decreasing influence of this work on nosological schemes. It is interesting that almost every theory of personality structure references neuroticism––and extraversionlike traits, which suggests the fundamental importance of these dimensions for functioning (John & Srivastava, 1999). Current well-accepted personality conceptualizations, such as the Big Three (Eysenck & Eysenck, 1975; Tellegen, 1985; Watson & Clark, 1993) and the Big Five (Digman, 1990; John, 1990; McCrae & Costa, 1987), prominently feature these dimensions of personality despite disagreement on additional traits (e.g., constraint in the Big Three and agreeableness, openness, and conscientiousness in the Big Five) and different methods of formulation. In addition to understanding the structure of personality, researchers have been interested in the neurobiological basis for such traits. Eysenck’s (1961, 1981) influential theory led to the development of the Big Three, and he was the first to implicate neuroticism and extraversion. He based his trait theory on variations in levels of cortical activation and autonomic nervous system reactivity and suggested that extraversion/positive emotion is associated with moderate levels of arousal, whereas neuroticism/negative emotion is associated with under- or overarousal. A number of researchers have examined the relationship of neuroticism (and extraversion) in the development of anxiety and other negative emotions. For example, Gershuny and Sher (1998) found, in a sample composed of 466 young adults, that the combination of high neuroticism and low extraversion at Time 1 seemed to play an important and predisposing role in the emergence of anxiety assessed 4 years later. Further bolstering the importance of neuroticism and extraversion in the experience of negative emotion, Gray (1982; Gray & McNaughton, 1996) described a similar trait theory and its neurobiological correlates that map onto Eysenck’s (1961, 1981) traits, namely, the behavioral inhibition system, behavioral approach system, and fight-flight system. In Gray’s theory, the biological basis for anxiety is the behavioral inhibition system’s (over)reaction to novel signals or punishment with exaggerated inhibition. High levels on Gray’s behavioral inhibition system roughly relate to high levels of neuroticism and low levels of extraversion in Eysenck’s model, and the behavioral approach system roughly corresponds to high extraversion and low neuroticism (Barlow,

The Nature, Diagnosis, and Treatment of Neuroticism 361

2002). The fight-flight system involves unconditioned escape behavior (i.e., flight) and defensive aggression (i.e., fight) in response to unconditioned punishment, such as pain, and unconditioned frustrative nonrewards (Gray, 1991; Gray & McNaughton, 1996). As such, the fight-flight system may represent a biological vulnerability specifically to the distinct emotion of fear/panic, as opposed to anxiety more generally. In another trait theory, Kagan (1989, 1994) examined children’s approach-and-withdrawal behavior and created profiles characterizing their levels of behavioral inhibition. Kagan’s (1989) definition of behavioral inhibition is similar to Gray’s (1982) in that it involves a low threshold for limbic arousal and uncertainty regarding unfamiliar events. Kagan considered these stable profiles as representing temperaments; physiological (increased salivary cortisol levels and muscle tension, greater pupil dilation, and elevated urinary catecholamine levels) and external (subsequent development of anxiety disorders) correlates of behavioral inhibition have also been found (Biederman et al., 1993; Hirshfeld-Becker et al., 1992). Robinson, Kagan, Reznick, and Corley (1992) suggested that temperament is clearly heritable; however, only 30% of individuals who clearly met criteria for behavioral inhibition as young children went on to develop anxiety disorders (Biederman, Rosenbaum, Hirshfeld, & Faraone, 1990). Moreover, temperament (as described in Kagan & Snidman, 1991) appears to be somewhat malleable, which suggests that environmental factors are also important determinants in temperament and anxiety vulnerability. These findings support the notion of a “constraining” biological vulnerability (in contrast to a “determining” role of temperament) in the development of anxiety in adolescence and adulthood, a theme to which we return in our later discussion of treatment. Finally, in one of the best known modern conceptualizations of temperaments related to anxiety and depression, L. A. Clark and Watson (1991) proposed two genetically based core dimensions, neuroticism/negative emotionality and extraversion/positive emotionality, as part of their tripartite theory (L. A. Clark, 2005; L. A. Clark, Watson, & Mineka, 1994; Watson, 2005). These concepts, originally based on Eysenck’s (1961, 1981) model, are also closely related to Gray’s (1982) constructs of behavioral inhibition and behavioral activation both conceptually and empirically. Although the traits reviewed earlier may turn out to be distinct in some way, current evidence in this area from CARD and elsewhere (L. A. Clark, 2005; Watson, 2005) lump these concepts together in a partially heritable temperament that we have labeled neuroticism/behavioral inhibition (or just neuroticism1) and behavioral activation/positive affect (T. A. Brown, 2007; T. A. Brown & Barlow, 2009; T. A. Brown et al., 1998; Campbell-Sills, Liverant, & Brown, 2004). We have been studying the latent structure of emotional disorders for the past 20 years (e.g., T. A. Brown et al., 1998; Zinbarg & Barlow, 1996) and have confirmed, with some modifications, the tripartite model of emotional disorders first proposed by L. A. Clark and Watson (1991). For example, the findings from T. A. Brown et al. (1998), which used a sample composed of 350 patients with DSM-IV-TR anxiety and mood disorders, confirmed a hierarchical structure. In this structure, neuroticism and extraversion emerged as higher-order factors to the DSM-IV-TR disorder factors, with significant paths

362 David H. Barlow, et al.

from neuroticism to each of the five DSM-IV-TR factors. In accord with a reformulated hierarchical model of anxiety and depression (Mineka, Watson, & Clark, 1998), results showed that extraversion was predictive of unipolar depression and social anxiety disorder only (see also T. A. Brown & McNiff, 2009). In addition, Rosellini, Lawrence, Meyer, and Brown (2010) found recently that agoraphobia (but not panic disorder) was associated with low extraversion, which provided support for the change in DSM-5 that separates agoraphobia from panic disorder as a distinct, new, diagnosis. In this model, autonomic arousal, which we consider to reflect largely the phenomenon of panic, emerges as a lower-order factor with significant paths from panic disorder and generalized anxiety disorder, which demonstrated a negative relationship with autonomic surges. These findings on latent structure have recently been extended both by our research team (T. A. Brown, 2007; T. A. Brown & Barlow, 2009) and by other researchers (e.g., Griffith et al., 2010; Kessler et al., 2011). For example, Griffith et al. (2010), studying a large sample of ethnically diverse adolescents and including both self-report and peer-report measures of neuroticism, found that a single internalizing factor was common to lifetime diagnoses of mood and anxiety disorders and that this internalizing factor was all but isomorphic with neuroticism. Noting the replication of earlier findings (e.g., T. A. Brown et al., 1998), Griffith et al. suggested that these results provided further evidence that neuroticism may be the core of “internalizing” disorders. Using factor analysis, Krueger (1999) similarly found that the variance in seven anxiety and mood disorders can be accounted for by the higher-order dimension of internalizing/neuroticism. Although the key features of the DSM anxiety and mood disorders (i.e., the specific symptoms used to discriminate between diagnoses) cannot be collapsed indiscriminately into higher-order temperamental dimensions, it seems safe to conclude, on the basis of these studies, that what is common outweighs what is not. Thus, virtually all of the considerable covariance among latent variables corresponding to the DSM-IV-TR constructs of depression, social anxiety disorder, generalized anxiety disorder, obsessive-compulsive disorder, and panic disorder was explained by the higher-order dimension of neuroticism (and extraversion); bipolar disorder was not included in these studies. Our own framework for understanding the origins of neuroticism as well as emotional disorders describes three separate but interacting diatheses or vulnerabilities (i.e., triple vulnerability theory; Barlow, 1988, 2000; Barlow, Ellard, Sauer-Zavala, Bullis, & Carl, 2013). We have explicated these vulnerabilities in some detail elsewhere (Barlow et al., 2013) but included a general biological (heritable) vulnerability common across disorders, a general psychological vulnerability consisting of a heightened sense of unpredictability and uncontrollability and associated changes in brain function resulting from adverse early experiences, and a more specific psychological vulnerability, also largely learned, accounting for why one particular emotional disorder may emerge instead of another. It seems increasingly evident that the two generalized vulnerabilities identified in the triple vulnerability theory are implicated in the development and expression of neuroticism itself (T. A. Brown, 2007; T. A. Brown & Barlow, 2009). Indeed, we hypothesize that these two generalized

The Nature, Diagnosis, and Treatment of Neuroticism 363

vulnerabilities function as direct risk factors for the development of neuroticism, which in turn mediates risk for the development of anxiety and mood disorders. What is notable for our purposes is that adverse experiences contribute strongly to changes in brain function and that later experiences may alter resulting temperamental expression and associated brain circuits (Kagan & Snidman, 1991; Nemeroff, 2013). The Nature of Emotional Disorders Negative reactivity to emotional experience appears fundamentally connected to neuroticism and resulting emotional disorder pathology. Emotional disorder is a term that has been used to group anxiety, unipolar mood, and related disorders, such as somatoform and dissociative disorders (Barlow, 1991; Barlow et al., 2011; T. A. Brown & Barlow, 2009). These disorders are characterized by a number of shared emotional disturbances, which appear to be closely linked to neuroticism. As described earlier, individuals with emotional disorders, compared with healthy individuals, have higher levels of negative affect (e.g., T. A. Brown & Barlow, 2009) and report experiencing more frequent and intense negative emotions (Campbell-Sills, Barlow, Brown, & Hofman, 2006; Mennin, Heimberg, Turk, & Fresco, 2005). In addition, individuals with emotional disorders, compared with healthy individuals, report less emotional clarity (Baker, Holloway, Thomas, Thomas, & Owens, 2004; McLaughlin, Mennin, & Farach, 2007; Tull & Roemer, 2007; Weiss et al., 2012) and acceptance of emotional experiences (McLaughlin et al., 2007; Tull & Roemer, 2007; Weiss et al., 2012) and find the experience of negative emotions more unpleasant (Roemer, Salters, Raffa, & Orsillo, 2005). In view of these reactions to negative emotions, it is not surprising that individuals with emotional disorders also display a range of cognitive and behavioral strategies aimed at reducing encounters with or the impact of negative emotions. Individuals with emotional disorders exhibit early vigilant information-processing biases toward negative information, but then they tend to quickly turn their attention away from such negative information (MacLeod & Mathews, 2012; Mathews & MacLeod, 2005). These individuals also react strongly to negative emotions when they occur with attempts to suppress or avoid the emotional experience (Aldao, Nolen-Hoeksema, & Schweizer, 2010; Baker et al., 2004; Moore, Zoellner, & Mollenholt, 2008; Tull & Roemer, 2007; Turk, Heimberg, Luterek, Mennin, & Fresco, 2005). In addition, individuals with emotional disorders, compared with healthy individuals, display greater intolerance for uncertainty, ambiguity, or situations that are perceived as uncontrollable, which leads to heightened negative affect. For example, intolerance of uncertainty and distress have been demonstrated across a range of disorders, including depression, generalized anxiety disorder, and social anxiety disorder (Boelen, Vrinssen, & van Tulder, 2010; Boswell, Thompson-Hollands, Farchione, & Barlow, 2013; Lee, Orsillo, Roemer, & Allen, 2010; Tolin, Abramowitz, Brigidi, & Foa, 2003). Increased perceptions of emotions as uncontrollable and intolerable as well as increased attempts at emotional control are also evident across disorders. A wide range of research has suggested that these ways of interpreting and responding to negative

364 David H. Barlow, et al.

emotions paradoxically serve to increase and maintain negative emotions and emotional disorder symptomatology. Thus, we consider this pathological reaction to emotional experience as the phenotypic core of emotional disorders. The following section discusses in detail the nature of these response tendencies and their sequelae. Evidence has suggested that how one interprets or reacts to negative emotions when they occur can affect the intensity and duration of the emotional experience (Sauer & Baer, 2009; Sauer-Zavala et al., 2012). A clear example of how this process unfolds can be seen in early models of panic disorder (Barlow, 1988; D. M. Clark, 1986). In these models, physical symptoms associated with initial panic attacks (e.g., increased heart rate) evoke anxiety about impending consequences (e.g., heart attack), which intensifies anxiety and its related physical symptoms and possibly triggers additional panic attacks. It is important to note that in individuals without panic disorder, occasional panic attacks do not evoke strong emotional reactions (nonclinical panic; Bouton, Mineka, & Barlow, 2001). Thus, the central issue in panic disorder is not the experience of panic attacks but the negative emotional reaction to the intense fear (panic) itself. Negative reactions or interpretations of emotions that intensify the experience are also prominent in other anxiety and depressive disorders. For example, in research on obsessive-compulsive disorder, Rachman and de Silva (1978) found that the content of ego-dystonic intrusive thoughts under stress are similar in patients diagnosed with obsessive-compulsive disorder and nonclinical control participants, which supports the notion that the way these thoughts are interpreted and managed has implications for the development of this emotional disorder. In addition, individuals with generalized anxiety disorder may find unexpected, uncontrolled emotional reactions that result from unplanned or mildly threatening situations particularly aversive and may engage in worry or checking behavior to regulate this emotional experience (Newman & Llera, 2011). Once again, the focus of an emotional disorder (panic attacks, intrusive thoughts, and social evaluation) may be determined by early learning experiences; however, the negative emotional reaction itself and one’s attempts to cope with or downregulate this emotional reaction are at the core of the disorder. Several constructs capturing the problematic reactions to emotions that may be implicated in the transdiagnostic development and maintenance of emotional disorders have been identified; specifically, these constructs measure the tendency of individuals to find emotional experiences aversive and, as such, engage in attempts to avoid them. Anxiety sensitivity, defined as a propensity for developing beliefs that anxiety-related symptoms will have negative somatic, cognitive, and social consequences (Reiss, 1991), is one such construct. Anxiety sensitivity represents an individual’s characteristic way of evaluating and responding to an emotional experience (specifically anxiety) when it occurs, distinct from the frequency or intensity of anxiety itself (Cox, Taylor, & Enns, 1999; Lilienfeld, 1999). Although anxiety sensitivity was originally introduced as a risk factor for panic disorder (Reiss, Peterson, Gursky, & McNally, 1986) and has predominantly been studied in the

The Nature, Diagnosis, and Treatment of Neuroticism 365

context of this disorder (e.g., Maller & Reiss, 1992; Plehn & Peterson, 2002; Rassovsky, Kushner, Schwarze, & Wangensteen, 2000), a large literature also has implicated anxiety sensitivity in the development of other anxiety disorders and depression (see Boswell, Farchione, et al., 2013; Naragon-Gainey, 2010; Taylor, 1999). For example, prospective studies have demonstrated that anxiety sensitivity predicts the onset of anxiety and depressive disorders (Maller & Reiss, 1992; Schmidt, Keough, Timpano, & Richey, 2008) beyond the contributions of the tendency to experience anxiety (see McNally, 1996, for a review) and that reductions in anxiety sensitivity during treatment predict symptom improvement (Gallagher et al., 2013). In addition, anxiety sensitivity has demonstrated incremental validity above trait neuroticism in the prediction of most mood and anxiety disorders (Collimore, McCabe, Carelton, & Asmundson, 2008; Cox, Enns, Walker, Kjernisted, & Pidlubny, 2001; Kotov, Watson, Robles, & Schmidt, 2007; Norton et al., 1997; Reardon & Williams, 2007). These results support the notion that how one relates to negative emotions is just as important in the development of emotional disorders as is the frequency and intensity of emotional experience. Another transdiagnostic construct that has been implicated in the development and maintenance of emotional disorders is experiential avoidance, defined as the unwillingness to remain in contact with uncomfortable internal experience (e.g., thoughts, emotions, sensations, memories, urges) through escape or avoidance (Hayes, Wilson, Gifford, Follette, & Strosahl, 1996). Self-report studies have demonstrated that individuals with anxiety disorders (Begotka, Woods, & Wetterneck, 2004; Hayes, Luoma, Bond, Masuda, & Lillis, 2006; Kashdan, Breen, Afram, & Terhar, 2010) and depressive disorders (Berking, Neacsiu, Comtois, & Linehan, 2009; Hayes et al., 2006; Shahar & Herr, 2011; Tull, Gratz, Salters, & Roemer, 2004) display high levels of self-reported experiential avoidance. The existing literature also has suggested that experiential avoidance both predicts generalized anxiety disorder symptoms even when the variance associated with frequency of negative affect is parceled out (Lee et al., 2010) and mediates the relationship between neuroticism and posttraumatic stress disorder symptoms (Maack, Tull, & Gratz, 2012; Pickett, Lodis, Parkhill, & Orcutt, 2012). Furthermore, Cheavens and Heiy (2011) recently found that avoidant coping partially mediates the relationship between the experience of negative emotions and major depressive symptoms among individuals high in experiential avoidance. Taken together, these findings suggest that emotional disorder symptoms are not simply a product of high levels of negative affect; instead, the combination of strong negative emotions and how one relates to them when they occur appears to be important for the development of these disorders. Individuals with emotional disorders also show deficits in mindfulness (Cheavens et al., 2005; Hayes et al., 1996), a related construct that refers to attention and awareness toward the present moment in an accepting manner regardless of how unpleasant the experience (Kabat-Zinn, 1982). Studies have shown that deficits in mindfulness are common across the emotional disorders (Baer, Smith, Hopkins, Kritemeyer, & Toney, 2006; K. Brown & Ryan, 2003;

366 David H. Barlow, et al.

Cash & Whittingham, 2010; Rasmussen & Pidgeon, 2011), which supports the case for similar underlying constructs in these disorders. Results of a recent study on the effects of dispositional mindfulness on response to a laboratory stressor suggested that individuals reporting higher levels of mindfulness display lower levels of self-reported anxiety and an attenuated cortisol response than do individuals endorsing lower levels of this construct (K. Brown, Weinstein, & Creswell, 2011). The impact of mindfulness on stress-related cortisol secretion has been associated with attenuated amygdala activation in response to threat (Creswell, Way, Eisenberger, & Leiberman, 2007). Consistent with data on experiential avoidance, results from studies have suggested that the degree to which one responds to negative emotions in a mindful manner predicts psychological symptoms over and above the contributions of a traitlike tendency to experience negative emotions (Sauer & Baer, 2009; Segal, Williams, & Teasdale, 2002). Evidence has suggested that many of the maladaptive behaviors associated with emotional disorders serve a function of facilitating escape or avoidance of intense emotions. Such behaviors include overt situational avoidance as well as more subtle forms of avoidance and safety behaviors. For example, situational avoidance is a hallmark feature of social anxiety disorder, specific phobias, posttraumatic stress disorder, depression, agoraphobia, and panic disorder (APA, 2013). Subtle forms of avoidance are also typical across most emotional disorders. In generalized anxiety disorder and obsessive-compulsive disorder, engaging in worry or compulsions (Newman & Llera, 2011), respectively, are subtle ways of avoiding the distress associated with experiencing anxiety. In social anxiety, subtle avoidance can include behaviors such as decreased eye contact or standing farther away from people during conversations or safety behaviors, such as engaging in social interactions only with a close friend present. In panic disorder, subtle avoidance includes avoiding activities that produce anxietylike sensations, such as exercise or drinking coffee (i.e., interoceptive avoidance). Safety behaviors across disorders include carrying around medications or even empty medication bottles, making sure to always have a cell phone or water on hand, or engaging in certain activities only with a “safe” person. Addressing such behavioral avoidance is an important element of most cognitive-behavioral treatments for emotional disorders, such as through use of fear and avoidance hierarchies in anxiety disorder protocols or activity scheduling in depression treatments. Some treatment protocols posit behavioral avoidance as comprising the core of the dysfunction, as in the example of behavioral activation, a well-supported treatment for depression, which is based on the notion that depressive symptoms are maintained by chronic avoidance of engagement or activity (Manos, Kanter, & Busch, 2010). In addition to engaging in problematic avoidant behaviors, individuals with emotional disorders engage in cognitive coping motivated by avoidance. Such processes include emotion suppression and rumination. Emotion suppression is a strategy in which individuals deliberately attempt to push unpleasant, emotion-inducing cognitions out of awareness; paradoxically, this strategy has been shown to produce rebound effects in which the

The Nature, Diagnosis, and Treatment of Neuroticism 367

suppressed thoughts return with greater frequency and intensity (Rassin, Muris, Schmidt, & Merkelbach, 2000; Wegner, Schneider, Carter, & White, 1987). High levels of emotion suppression have been demonstrated across emotional disorders, including depression, generalized anxiety disorder, obsessive-compulsive disorder, and posttraumatic stress disorder (Purdon, 1999), and have also been shown to exacerbate symptoms (Abramowitz, Tolin, & Street, 2001). In particular, emotional suppression has been associated with increased physiological arousal (Hofmann, Heering, Sawyer, & Asnaani, 2009). It is hypothesized that emotion suppression, like other forms of avoidance, is a negatively reinforced behavior that produces short-term reductions in negative affect, despite then spawning increased negative emotions in the longer term. Rumination refers to repetitively and passively focusing on negative mood and its possible causes, meanings, and consequences (Nolen-Hoeksema, 1991). Like suppression, rumination can be conceptualized as an avoidant strategy because passive focus on surface matters may serve to protect individuals from more distressing concerns (Lyubomirsky & Nolen-Hoeksema, 1995; Lyubomirsky, Tucker, Caldwell, & Berg, 1999). Rumination has been shown to intensify negative affect (Nolen-Hoeksema, Wisco, & Lyubomirsky, 2008), which leads to more rumination, in what Selby, Anestis, and Joiner (2008) described as an emotional cascade; this process continues until a maladaptive behavior (reassurance seeking, substance use, binge eating, etc.) interrupts the cycle. Rumination appears to be prominent across emotional disorders (see Aldao et al., 2010) and prospectively predicts increases in anxiety and depressive symptoms (Butler & Nolen-Hoeksema, 1994; Calmes & Roberts, 2007; Hong, 2007; Nolen-Hoeksema, 2000; Nolen-Hoeksema, Larson, & Grayson, 1999; O’Connor, O’Connor, & Marshall, 2007; Sarin, Abela, & Auerbach, 2005; Segerstrom, Tsao, Alden, & Craske, 2000). It seems clear that negative reactions to strong emotions lead to similar forms of avoidant cognitive coping (e.g., emotion suppression and rumination) common to all emotional disorders. These strategies that intensify already strong negative emotions appear to lead to greater use of disorder-specific behavioral coping. Overall, the literature has suggested that individuals with emotional disorders experience strong negative emotions with frequency and evaluate these experiences as aversive. As a result of these negative reactions to their emotions, such individuals are more likely to engage in avoidant coping strategies to manage emotional experiences, and these strategies, in turn, paradoxically increase the frequency/intensity of negative emotions. Once again, we suggest that this functional relationship, driven by neuroticism, is at the core of disorders of emotion. Neuroticism/Trait Anxiety, Nosology, and Dimensional Diagnosis of Emotional Disorders It was long thought that DSM-IV-TR represented the zenith of a splitting approach to nosology we described earlier and that began in the 1980s with the publication of DSM-III, an approach basically unchanged in DSM-5. The

368 David H. Barlow, et al.

advent of an objective empirically based system of classification of mental disorders was an enormous advance over previous systems based on unsupported etiological theories best exemplified by the term neurosis. Beginning in the 1980s, with the splitting of neuroses into anxiety, depressive, somatoform, and related disorders, meaningful research on outcomes of pharmacological and individual psychological treatments, particularly cognitive-behavioral therapy, targeting these disorders appeared (e.g., Barlow et al., 1984). It also became clear at that time, on the basis of the pioneering work of Strupp (1973), that clinical trials require the generation of detailed individual therapeutic protocols to specify an independent variable. As a result, these psychotherapeutic treatments were increasingly characterized by well-articulated individual protocols targeted to specific forms of psychopathology as articulated in DSM-III and its successors, particularly anxiety and depressive disorders. These treatments were then evaluated empirically and found efficacious in a variety of formats, uses, and settings (Barlow, 1996, 2004, 2008; Barlow, Gorman, Shear, & Woods, 2000; Heimberg, Liebowitz, & Hope, 1998; Nathan & Gorman, 2007). It is fair to say that these findings have had a substantial impact in that public-health authorities have allocated billions of dollars for training and dissemination of these treatments (McHugh & Barlow, 2010). This approach to nosology also has ensured high rates of diagnostic reliability; however, as mentioned earlier, there was growing suspicion both that advances in classification and treatment development represented by this approach came at the expense of diagnostic validity and that the current system may be overemphasizing categories that are minor variations of broader underlying syndromes. The careful consideration of these broader underlying syndromes as a conceptual approach to nosology would not imply a return to a nonempirical system of classification based on theories of etiology. Rather, this thinking points to a quantitative approach using structural equation modeling to examine the full range of anxiety and mood disorders without the constraints of artificial categories, given their strong relationship and potential overlap. Thus, our evolving view is that DSM-5 emotional disorder categories do not qualify in any sense as real entities (Kendell, 1975) but do seem to be useful concepts or constructs that emerge as “blips” on a general background of temperament. The conceptualization of emotional disorders in a more dimensional fashion should result in a more satisfactory representation of salient aspects of these disorders that would eliminate vexing issues of comorbidity. But moving from this conceptualization to a dimensional system of diagnostic assessment with implications for treatment has proven an exceedingly difficult task even in areas such as personality disorders in which there is widespread agreement that this would be the preferred approach (Widiger & Crego, 2013). The recent failure to accomplish this goal in DSM-5 underscores these difficulties. In a preliminary attempt to conceptualize how this effort might work for emotional disorders, we proposed a dimensional classification scheme to reflect the research described earlier (T. A. Brown & Barlow, 2009). The purpose would be to create a profile that may provide a more complete

The Nature, Diagnosis, and Treatment of Neuroticism 369

portrayal of a patient’s clinical presentation than would a more categorical approach that often consists of several individual comorbid diagnoses. The profile would highlight levels of constructs thought to be important in forming a useful case conceptualization, including neuroticism, extraversion (referred to as behavioral activation/positive affectivity in this model), avoidance, mood, and specific foci of anxiety (e.g., panic and other autonomic surges, somatic symptoms, intrusive cognitions, social evaluation, and trauma). Scores on the dimension of trait neuroticism, arguably the most important construct in this model, reflect the frequency, intensity, and distress associated with negative emotions, as well as perceptions of uncontrollability regarding future challenges and low self-efficacy regarding one’s ability to cope. The higher-order dimension of extraversion/positive affect is also represented because low levels of this trait are specifically associated with major depressive disorder, social anxiety disorder, and agoraphobia, whereas high levels are associated with euthymic states of bipolar and cyclothymic disorders. As highlighted earlier, individuals with high levels of neuroticism are likely to display avoidance behaviors. In this dimensional system, avoidance is broken into two types: behavioral/interoceptive and cognitive/emotional. Specific examples of disorder profiles highlight the heuristic clinical value associated with dimensional classification of emotional disorders. For example, individuals with a principal diagnosis of panic disorder would likely display profiles with high levels of neuroticism, avoidance, and preoccupation with panic/autonomic arousal and other somatic symptoms. In contrast, patients presenting with posttraumatic stress disorder might display high neuroticism and preoccupation with panic/autonomic arousal (flashbacks) and past trauma. Although each diagnostic category is linked to a prototypic dimensional profile, this classification system would allow clinicians to determine the extent to which other key features are present that would potentially affect treatment planning. A new measure, the multidimensional emotional disorder inventory (MEDI), recently was developed to assess these important vulnerabilities and characteristics of emotional disorders with a single assessment tool; the ongoing MEDI validation study has suggested that this measure may be a reliable and valid method for assessing important emotional disorder dimensions (Rosellini, 2013). A representation of what a MEDI profile might look like is shown in Figure 17.1. These data present clinical estimates of constructs composing the dimensional scheme from a patient seen at CARD with a principal diagnosis of posttraumatic stress disorder (motor vehicle accident) and additional diagnoses of generalized anxiety disorder and subclinical depression. Under the current diagnostic system, unless a patient meets full diagnostic criteria for a comorbid disorder, information on the dimensions not associated with the primary diagnosis are discarded. Given that the rates of comorbidity among emotional disorders are high, with even greater overlap at the subclinical level (T. A. Brown & Barlow 2009), a dimensional classification system would allow for the integration of several important areas of functioning.

0 10 20 30 40 50 60 70 80 90 100

Score

370 David H. Barlow, et al.

A/N

BA/P

Temperament

DEP

MAN

SOM

Mood

PAS

IC

SOC

Focus of Anxiety

TRM

AV-BI AV-CE Avoidance

Figure 17.1 Example profile of patient evaluated with a dimensional classification system. Higher scores on the y-axis indicate higher levels of the x-dimension; otherwise, the y-axis metric is arbitrary and used for illustrative purposes. Note: A/N = anxiety/neuroticism; BA/P = behavioral activation/positive affect; DEP = unipolar depression; MAN = mania; SOM = somatic anxiety; PAS = panic and related autonomic surges; IC = intrusive cognitions; SOC = social evaluation;TRM = past trauma; AV-BI = behavioral and interoceptive avoidance; AV-CE = cognitive and emotional avoidance.

Malleability of Neuroticism The evidence regarding the malleability of personality traits over time or in response to therapeutic intervention is mixed. As noted earlier, Kagan (1989, 1994) described his conceptualization of behavioral inhibition as strongly heritable and stable; however, empirical evidence, including Kagan’s own work, has suggested that only 30% of children who clearly met criteria for this trait as young children went on to develop anxiety disorders, although some of the participants may have remained shy. Several additional studies have yielded similar results (e.g., Hayward, Killen, Kraemer, & Taylor, 1998; Hirshfeld-Becker et al., 2007; Schwartz, Snidman, & Kagan, 1999). These results led Kagan to view behavioral inhibition as a constraining factor subject to environmental influences. Such influences may include stress and having parents diagnosed with anxiety disorders; behaviorally inhibited third to sixth graders were more likely to experience increased anxiety if they reported more daily hassles (Brozina & Abela, 2006) and were more likely to be diagnosed with an anxiety disorder if their parents also experienced such disorders (Biederman et al., 2001). L. A. Clark (2009) has reviewed research examining stability and change in personality disorders in adults. Contrary to the thinking of most personality theorists, as well as a statement in the DSM-5, which holds that personality traits are stable, inflexible, and pervasive, is Clark’s observation that the collection of traits and behaviors that make up the definition of most personality disorders do change, albeit slowly over time, with the greatest change occurring

The Nature, Diagnosis, and Treatment of Neuroticism 371

in the behavioral manifestations of these traits. Furthermore, researchers examining longitudinal changes in negative traits, such as neuroticism, in the normal population have observed gradual age-related decreases that continue long into old age (Roberts & Mroczeck, 2008; Roberts, Walton, & Viechtbauer, 2006). Similar age-related decreases in the related higher order construct of internalizing have also been found; specifically, internalizing means and indicator intercepts were lower for older-aged cohorts (Eaton, Krueger, & Oltmanns, 2011; Oltmanns & Balsis, 2011). This pattern of results is found at the mean level, which suggests that, summed across individuals, neuroticism decreases across time. In addition, change in neuroticism has been examined at the individual level using growth modeling (Mroczek & Spiro, 2003); results suggested that there is great variability in extent of change on temperamental variables, with some people remaining at stable levels and other people changing a great deal (Helson, Jones, & Kwan, 2002; Small, Hertzog, Hultsch, & Dixon, 2003). At CARD, we have also investigated longitudinal change in neuroticism, with particular focus on responsiveness to psychological intervention, and have encountered mixed findings. For example, in a study on the treatment of panic disorder (T. A. Brown & Barlow, 1995), comorbid anxiety and depressive diagnoses improved immediately after successful treatment even though they were not specifically targeted, which suggests, as described earlier, the existence of an underlying temperamental vulnerability. But at a 2-year follow-up, comorbid diagnoses had returned to a level (30%) that was no longer significantly different from pretreatment. This result occurred despite the fact that, in the aggregate, patients maintained or improved on gains for panic disorder across the follow-up interval, which indicates considerable independence between panic disorder symptoms and related comorbidity. Although speculative, and contrary to the more usual interpretations offered earlier, these findings may suggest that current cognitive-behavioral treatments are generally effective in addressing the specific symptoms and maintaining processes of the targeted disorder (in this case, panic disorder) both immediately and at follow-up but do not result in substantial reductions in general predispositional features (e.g., neuroticism), which leaves patients vulnerable to the emergence or persistence of other disorders. These findings also raise the possibility that current psychological treatments have become overly specialized because they focus on disorder-specific features, such as panic attacks in panic disorder or rituals connected to obsessional thought in obsessive-compulsive disorder, neglecting broader dimensions that might produce more favorable long-term outcomes across all disorders. In a more direct evaluation of this issue, the temporal stability (8 months) and predictive utility of self-reported levels of neuroticism and extraversion in 41 individuals diagnosed with major depressive disorder (most of whom received some kind of treatment during this time) has also been examined (Eaton et al., 2011; Oltmanns & Balsis, 2011). Low levels of extraversion at Time 1 predicted poorer clinical outcome of major depressive disorder at the 8-month reassessment. Moreover, neuroticism and extraversion were remarkably stable over time, despite changes in clinical state. In fact, although more than one third of depressed participants were classified as no longer depressed

372 David H. Barlow, et al.

at the 8-month follow-up, neuroticism and extraversion displayed the same high level of temporal stability in this group as in a subgroup of participants who were depressed at both assessment points. Despite these overall findings, because there was no information available regarding what types of treatment patients received during this study, it is difficult to determine how specific treatments may have affected the temperamental constructs. In contrast, other researchers have indeed found changes in neuroticism as a function of time and treatment. For example, we examined the temporal course and temporal structural relationships of dimensions of temperament (neuroticism, extraversion) within DSM-IV-TR disorder constructs of depression, social anxiety disorder, and generalized anxiety disorder (T. A. Brown, 2007; T. A. Brown & Barlow, 2009). Outpatients with these disorders (N = 606) were first examined at intake and then reassessed at 1- and 2-year follow-ups. The majority (76%) of patients received some kind of treatment after intake, although not all at CARD, of varying duration and quality. The overall rate of diagnosed anxiety and mood disorders declined significantly during follow-up from 100% at intake to 58% at the 2-year follow-up. Despite the marked decline in DSM-IV-TR diagnoses by the 2-year follow-up, test-retest correlations of the factors and unconditional latent growth models indicated that extraversion evidenced a very high degree of temporal stability, consistent with its conceptualization as a trait vulnerability construct that is relatively unaffected by treatment. However, of the five constructs examined, neuroticism evidenced the greatest amount of temporal change and was the dimension associated with the largest treatment effect. In addition to its inconsistency with some prior research, such as that reviewed earlier (e.g., Kasch et al., 2002), this finding is clearly at odds with conceptual assumptions that core dimensions of temperament are stable, inflexible, and more resistant to psychological treatment. Levels of neuroticism and extraversion have also been explored in a more specified context across a large randomized controlled trial of cognitive therapy compared with placebo for adults with major depressive disorder (Tang et al., 2009). Results indicated that cognitive therapy produced greater changes in both neuroticism and extraversion than did placebo, but contrary to T. A. Brown’s (2007) results, cognitive therapy maintained a significant effect only for extraversion after controlling for changes in depression symptoms during treatment. Nevertheless, the Tang et al. (2009) study differs from Brown’s study in the clinical features of the sample (pure major depressive disorder vs. heterogeneous anxiety and unipolar depressive disorders) and the degree of control provided regarding the treatment condition (placebo controlled and with stringent adherence/fidelity procedures vs. naturalistic, heterogeneous treatment). There is also evidence that neuroticism may operate differently than the DSM-IV-TR disorder constructs in several ways (T. A. Brown, 2007). For instance, unconditional latent growth models of each DSM-IV-TR disorder construct revealed inverse relations between the intercept and the slope; that is, higher initial disorder severity was associated with greater change over time. However, the intercept and the slope of neuroticism were positively correlated (r = .47), which indicated that patients with higher initial levels of neuroticism

The Nature, Diagnosis, and Treatment of Neuroticism 373

tended to show less change in this dimension over time, and, conversely, patients with lower initial levels of neuroticism tended to evidence greater change. Thus, unlike the DSM-IV-TR disorders, the stability of neuroticism increased as a function of initial severity. In addition, parallel-process latent growth models indicated that higher initial levels of neuroticism were associated with less change in the DSM-IV-TR constructs of generalized anxiety disorder and social anxiety disorder. Although neuroticism alone demonstrated no temporal relation with depression, the presence of chronic stress moderated the relationship such that high neuroticism resulted in less improvement in depression as the level of chronic stress increased (T. A. Brown & Rosellini, 2011). These results are consistent with results from earlier work, as well as theory, that showed that neuroticism has directional temporal effects on Axis I psychopathology (cf. Gershuny & Sher, 1998; Kasch et al., 2002; Meyer, Johnson, & Winters, 2001) but that the converse does not seem to occur; that is, initial levels of the DSM-IV-TR disorders did not predict increases in temperament over time. Finally, evidence has suggested that change in DSM disorder constructs (e.g., depression, social anxiety disorder, and generalized anxiety disorder) is significantly related to change in neuroticism (T. A. Brown, 2007). Of particular interest is the finding that all the temporal covariance of the DSM-IV-TR disorder constructs was accounted for by change in neuroticism; that is, when neuroticism was specified as a predictor, there was no temporal overlap among disorder constructs. The correlational nature of these findings precludes firm conclusions about the direction of these effects. Nevertheless, and counter to some earlier evidence and conceptualizations (T. A. Brown et al., 1995; Kasch et al., 2002), all of these findings suggest that neuroticism may be therapeutically malleable, and it is this malleability that mediates the extent of change in the emotional disorders. Some authors have suggested that decreases in neuroticism over time may partly reflect the fact that measures of temperament overlap to some degree with symptomatic measures of anxiety and depression, which results in distortions on estimates of temperament (mood-state distortion; cf. L. A. Clark, Vittengl, Kraft, & Jarrett, 2003; Jylhä & Isometsä, 2006; Widiger, Verheul, & van den Brink, 1999). That is, the measurement of neuroticism consists of some combination of stable temperamental variance (i.e., vulnerability) and variability attributable to generalized distress that would be subject to greater temporal fluctuation and would imply that neuroticism is apt to covary with temporal fluctuations in the severity of disorders. However, results from a recent study have suggested that measures of neuroticism (and extraversion) primarily capture true temperamental variance even in individuals with emotional disorders (Naragon-Gainey, Gallagher, & Brown, 2013). In addition, a number of longitudinal studies have controlled for the periodic occurrence of anxious or depressive symptoms and still found that neuroticism acted independently in predicting anxiety and mood (Lahey, 2009; Spijker, de Graf, Oldehinkel, Nolen, & Ormel, 2007). In sum, the malleability of neuroticism and other temperamental variables, particularly in response to treatment, remains an unsettled question. The studies described earlier, in which researchers have examined changes

374 David H. Barlow, et al.

in temperamental variables in the context of naturalistic treatments or treatments targeting disorder-specific symptoms, have yielded mixed findings. In some studies, temperament dimensions appeared to change during the course of treatment in the expected directions (e.g., T. A. Brown, 2007; Kennedy, Rapee, & Edwards, 2009), whereas in other studies, no changes in temperament occurred (e.g., Kasch et al., 2002). Inconsistencies also exist in the degree to which the different dimensions of temperament respond to treatment across studies. Indeed, the research reviewed raises questions about the nature and mechanisms of change of temperament during treatment of emotional disorders, how best to measure temperament, and whether directly targeting temperament therapeutically would lead to more definitive results. Prevention and Treatment of Neuroticism In most of the studies in which researchers have examined changes in temperament in response to psychological interventions, including the studies described in the previous section, researchers have not provided a priori hypotheses regarding how and why the study treatment might affect temperament other than as a by-product of symptom reduction. The interventions used were not designed to target features of temperament but, rather, to address symptoms. Within such studies, it is difficult to interpret changes in temperament. One explanation for the mixed findings across studies is that there may be specific interactions between the intervention and the dimensions of temperament that influence which temperament variables respond and to what extent. Such interactions have been largely unexplored in studies of psychological-treatment outcomes. However, recent research from pharmacological-treatment studies has supported this treatment-temperament interaction hypothesis, and emerging research from our own laboratory has suggested that dimensions of temperament can be more directly targeted through specialized treatments. In this section, we review recent evidence for treatments specifically designed to affect temperament and discuss briefly our own efforts to develop a psychological treatment focused on directly addressing neurotic temperament. Most of the studies in which researchers examine interventions specifically designed to target temperament have come from the literature on psychopharmacology (for a review, see Soskin, Carl, Alpert, & Fava, 2012). Results from these studies have provided some evidence for specific interactions between treatment agents and temperament variables. To summarize, this research has indicated that serotonergic drug agents (i.e., selective serotonin reuptake inhibitors) produce dampening effects on neuroticism (Fu et al., 2004; Harmer et al., 2009; Harmer, Mackay, Reid, Cowen, & Goodwin, 2006; Murphy, Yiend, Lester, Cowen, & Harmer, 2009; Quilty, Meusel, & Bagby, 2008) and possibly to a lesser extent on extraversion (McCabe, Mishor, Cowen, & Harmer, 2010), whereas catecholaminergic (i.e., noradrenergic/dopaminergic) agents produce specific enhancement of extraversion (McCabe et al., 2010; Tomarken, Dichter, Freid, Addington, & Shelton, 2004). The specific neurobiological properties of these agents have been hypothesized to mediate such observed effects on temperament variables. For example, serotonergic agents have been shown to decrease hyperreactivity of the amygdala in response to fear-inducing stimuli and to inhibit dopaminergic neurotransmission in areas of the prefrontal

The Nature, Diagnosis, and Treatment of Neuroticism 375

cortex. In contrast, catecholaminergic agents upregulate noradrenergic and dopaminergic neurotransmission, particularly within the mesolimbic reward circuitry (Soskin et al., 2012). Despite the obvious differences between pharmacological and psychological treatments, the studies on preferential effects of pharmacological agents on dimensions of temperament have provided some support for the notion that treatments can be designed to selectively target temperamental variables. As a corollary, the pharmacologic results also suggested that treatments, whether pharmacological or psychological, should not be expected to produce equivalent changes in temperament. Some treatments may produce no effects on temperament, whereas other treatments may produce generalized or specific effects on dimensions of temperament. Behavioral interventions designed to specifically address temperamental vulnerabilities are limited in number. Rapee, Kennedy, Ingram, Edwards, and Sweeney’s (2005) intervention for children identified as behaviorally inhibited was designed with the purpose of preventing the later onset of anxiety and related disorders and serves as one example. The program consists of a parent-focused intervention that includes psychoeducation about the nature of anxiety, traditional cognitive-behavioral strategies (i.e., exposure and cognitive restructuring) directed toward personal concerns, and training in behavior management techniques that prevent an overprotective parenting style. Results from randomized controlled trials (Rapee et al., 2005; Rapee, Kennedy, Ingram, Edwards, & Sweeney, 2010) have indicated that this program is clearly successful at preventing anxiety disorders, but for our purposes, the most interesting findings are those on the effects of the program on temperament. Specifically, using a brief version of this program, Rapee et al. (2010) found that levels of behavioral inhibition did not differ significantly on the basis of either parent report or laboratory observation, despite the success in preventing the later onset of anxiety disorders. However, when the program was administered in a more intensive format with higher risk children, compared with a group that did not receive the treatment, reductions in measures of temperament did occur (Kennedy et al., 2009). Rapee et al. also noted that differences among groups seemed to increase with time, which suggested to them that interventions directed at temperament (and related risk factors) might produce an increasing trajectory of change in temperament over the years, at least in children. In addition, some work has been conducted in an effort to identify intervention strategies specifically for targeting positive affect (extraversion). For example, in an experience-sampling study, Mata et al. (2012) found that both participants diagnosed with major depressive disorder and control participants reported increases in positive affect directly following physical activity and that depressed participants in particular demonstrated a dose-response effect such that longer and more intense instances of physical activity led to greater increases in positive affect. Speisman, Kumar, Rani, Foster, and Ormerod (2012) recently demonstrated in animal laboratories that exercise increases neurogenesis, which could possibly be one mechanism of action in successful psychological treatments using exercise. Despite strong preliminary evidence, this theory must now undergo the slow process of scientific confirmation. Given the clinical promise of therapeutically addressing temperamental vulnerabilities, we have devoted more than a decade to developing a psychological

376 David H. Barlow, et al.

treatment that targets the putative, fundamental, underlying processes of anxiety and mood disorders that may be more closely related to temperament. This treatment, the unified protocol for transdiagnostic treatment of emotional disorders (UP), which has been described in detail elsewhere (Barlow et al., 2011), is a cognitive-behavioral intervention designed to address core temperamental processes in emotional disorders. The UP targets identification and modification of the strong negative reactions to emotions that lead to problematic, avoidant coping across emotional disorders (Ellard, Fairholme, Boisseau, Farchione, & Barlow, 2010). Amelioration of negative reactions to emotions in turn changes the frequency and intensity of future emotional experiences and thereby affects temperamental constructs. The UP has now been evaluated for its efficacy in treating anxiety disorders in a series of preliminary trials culminating with a small randomized controlled trial (N = 37) comparing a treatment group with a wait list control group (Ellard et al., 2010; Farchione et al., 2012). Results from these studies have indicated that the UP is an efficacious treatment for a range of anxiety disorders with stable improvements out to an 18-month follow-up (Bullis, Fortune, Farchione, & Barlow, 2013). We have also recently begun a large randomized controlled equivalence trial (N = 250) comparing the UP with four well-established single anxiety disorder treatment protocols on the basis of patients’ principal diagnoses (generalized anxiety disorder, obsessive-compulsive disorder, social anxiety disorder, or panic disorder with or without agoraphobia). In following with the goal of assessing the extent to which the UP addresses temperamental vulnerabilities in addition to current symptoms, we have also conducted an investigation of the effects of the UP on dimensions of temperament (see Carl, Gallagher, Sauer-Zavala, Bentley, & Barlow, 2014) in the context of the randomized controlled trial mentioned in the previous paragraph (i.e., Farchione et al., 2012). In brief, our results indicated that in the treatment group, compared with the wait list group, the UP produced small to moderate effects on both neuroticism and extraversion from pre- to posttreatment, and these changes in temperament are associated with improvements in core symptomatology, functional impairment, and quality of life (Carl et al., 2013). The results of this investigation suggest the importance of changes in temperament as they affect treatment outcomes. Neuroticism and extraversion contributed to both shared and distinct treatment outcomes. Decreased neuroticism at posttreatment and at 6-month follow-up predicted decreased anxiety and depressive symptoms. Increased extraversion was associated with decreased depressive symptoms at posttreatment and decreased anxiety symptoms at 6-month follow-up, and changes in both temperament variables predicted reductions in functional impairment at 6-month follow-up. Finally, extraversion alone was associated with higher quality of life at posttreatment and at 6-month follow-up. Although preliminary, these results suggest that there are both common and differentiated outcomes associated with changes in specific dimensions of temperament. Although both neuroticism and extraversion can affect depression and anxiety symptoms and functional impairment, only extraversion appears to directly influence quality of life. In future research, it will be important to investigate and gain a better understanding of what accounts for the variability in the effects of these temperamental predictors on treatment outcomes.

The Nature, Diagnosis, and Treatment of Neuroticism 377

In summary, contrary to traditional conceptions, a variety of research has suggested that dimensions of temperament may be malleable over time or in response to treatment, but such findings have been mixed, which indicates that more research is required to identify specific conditions that affect temperament. Specifically, recently developed interventions that target dimensions of temperament more directly have provided preliminary support for the notion that psychological interventions can address temperamental vulnerabilities and that such improvements are associated with a range of beneficial treatment outcomes (Carl et al., 2013; Farchione et al., 2012; Kennedy et al., 2009). If confirmed, these findings may shift the focus of investigation into the nature, diagnosis, and treatment of emotional disorders. Future Research Directions The construct of neuroticism is almost as old as the study of psychopathology itself, but recent developments described herein suggest fresh, new directions for research. Considered broadly, can we develop targeted psychological interventions for neuroticism? If so, will these interventions provide a more efficient and effective way to affect the broad sweep of phenomena across the spectrum of emotional disorders, including common patterns of comorbidity and subthreshold symptomatic presentations? Will these conceptions move us further along toward the goal of a more satisfactory dimensional system for classifying the emotional disorders and facilitate the development of diagnostic instruments that will greatly simplify the process of assessment? And can we usefully extend these research objectives to other relevant temperaments, such as positive affect and perhaps constraint? The accumulation of important basic research covered briefly in this review suggests that it may now be possible to translate these concepts into a fundamentally new approach to the diagnosis, assessment, and treatment of emotional disorders. Notes 1

We henceforth refer to this construct as neuroticism; however, it should be noted that the individual studies described may have used alternate terms. D. H. Barlow developed the main thesis of the manuscript. D. H. Barlow, S. Sauer-Zavala, J. R. Carl, J. R. Bullis, and K. K. Ellard all contributed substantially to the literature review and subsequent drafting of the manuscript in support of this thesis. Portions of this article were presented by D. H. Barlow as the James McKeen Cattell Address at the Annual Meeting of the Association for Psychological Science, May 2012. The authors declared that they had no conflicts of interest with respect to their authorship or the publication of this article.

References Abramowitz, J., Tolin, D., & Street, G. (2001). Paradoxical effects of thought suppression: A meta-analysis of controlled studies. Clinical Psychology Review, 21, 683–703. Aldao, A., Nolen-Hoeksema, S., & Schweizer, S. (2010). Emotion-regulation strategies across psychopathology: A meta-analytic review. Clinical Psychology Review, 30, 217–237. doi:10.1016/j.cpr.2009.11.004

378 David H. Barlow, et al. Allen, L. B., White, K. S., Barlow, D. H., Shear, K. M., Gorman, J. M., & Woods, S. W. (2010). Cognitive-behavior therapy (CBT) for panic disorder: Relationship of anxiety and depression comorbidity with treatment outcome. Journal of Psychopathology and Behavioral Assessment, 32, 185–192. American Psychiatric Association. (1980). Diagnostic and statistical manual of mental disorders (3rd ed.). Washington, DC: Author. American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th ed., text rev.). Washington, DC: Author. American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Arlington, VA: American Psychiatric Publishing. Andrews, G. (1990). Classification of neurotic disorders. Journal of the Royal Society of Medicine, 83, 606–607. Andrews, G. (1996). Comorbidity in neurotic disorders: The similarities are more important than the differences. In R. M. Rapee (Ed.), Current controversies in the anxiety disorders (pp. 3–20). New York, NY: Guilford Press. Baer, R. A., Smith, G. T., Hopkins, J., Kritemeyer, J., & Toney, L. (2006). Using self-report assessment methods to explore facets of mindfulness. Assessment, 13, 27–45. Baker, R., Holloway, J., Thomas, P., Thomas, S., & Owens, M. (2004). Emotional processing and panic. Behaviour Research and Therapy, 42, 1271–1287. Barlow, D. H. (1988). Anxiety and its disorders: The nature and treatment of anxiety and panic. New York, NY: Guilford Press. Barlow, D. H. (1991). Disorders of emotion. Psychological Inquiry, 2, 58–71. Barlow, D. H. (1996). The effectiveness of psychotherapy: Science and policy. Clinical Psychology: Science and Practice, 3, 236–240. Barlow, D. H. (2000). Unraveling the mysteries of anxiety and its disorders from the perspective of emotion theory. American Psychologist, 55, 1247–1263. Barlow, D. H. (2002). Anxiety and its disorders: The nature and treatment of anxiety and panic (2nd ed.). New York, NY: Guilford Press. Barlow, D. H. (2004). Psychological treatments. American Psychologist, 59, 869–878. Barlow, D. H. (2008). Clinical handbook of psychological disorders: A step-by-step treatment manual (4th ed.). London, England: Oxford University Press. Barlow, D. H., Cohen, A. S., Waddell, M. T., Vermilyea, B. B., Klosko, J. S., Blanchard, E. B., & DiNardo, P. A. (1984). Panic and generalized anxiety disorders: Nature and treatment. Behavior Therapy, 15, 431–449. Barlow, D. H., Ellard, K. K., Fairholme, C., Farchione, T. J., Boisseau, C., Allen, L., & Ehrenreich-May, J. (2011). Unified protocol for the transdiagnostic treatment of emotional disorders. New York, NY: Oxford University Press. Barlow, D. H., Ellard, K. K., Sauer-Zavala, S., Bullis, J. R., & Carl, J. R. (2013). The origins of trait neuroticism. Manuscript in preparation. Barlow, D. H., Gorman, J. M., Shear, M. K., & Woods, S. W. (2000). Cognitive-behavioral therapy, imipramine, or their combination for panic disorder: A randomized controlled trial. Journal of the American Medical Association, 283, 2529–2536. Begotka, A., Woods, D., & Wetterneck, C. (2004). The relationship between experiential avoidance and the severity of trichotillomania in a nonreferred sample. Journal of Behavior Therapy and Experimental Psychiatry, 35, 17–24. Berking, M., Neacsiu, A., Comtois, K., & Linehan, M. (2009). The impact of experiential avoidance on the reduction of depression in treatment for borderline personality disorder. Behaviour Research and Therapy, 47, 663–670. Biederman, J., Hirshfeld-Becker, D., Rosenbaum, J., Herot, C., Friedman, D., Snidman, N., . . . Farone, S. (2001). Further evidence of the association between behavioral inhibition and social anxiety in children. American Journal of Psychiatry, 158, 1673–1679.

The Nature, Diagnosis, and Treatment of Neuroticism 379 Biederman, J., Rosenbaum, J. F., Bolduc-Murphy, E. A., Faraone, S. V., Chaloff, J., Hirshfeld, D. R., & Kagan, J. (1993). A 3-year follow-up of children with and without behavioral inhibition. Journal of the American Academy of Child and Adolescent Psychiatry, 32, 814–821. Biederman, J., Rosenbaum, J., Hirshfeld, D., & Faraone, S. (1990). Psychiatric correlates of behavioral inhibition in young children of parents with and without psychiatric disorders. Archives of General Psychiatry, 47, 21–26. Blair, K., Shaywitz, J., Smith, B. W., Rhodes, R., Geraci, M., Jones, M., . . . Pine, D. S. (2008). Response to emotional expressions in generalized social phobia and generalized anxiety disorder: Evidence for separate disorders. American Journal of Psychiatry, 165, 1193–1202. doi:10.1176/appi.ajp.2008.0707106 Boelen, P., Vrinssen, I., & van Tulder, F. (2010). Intolerance of uncertainty in adolescents: Correlations with worry, anxiety, social anxiety, and depression. Journal of Nervous and Mental Disease, 198, 194–200. Borkovec, T. D., Abel, J. L., & Newman, H. (1995). Effects of psychotherapy on comorbid conditions in generalized anxiety disorder. Journal of Consulting and Clinical Psychology, 63, 479–483. Boswell, J. F., Farchione, T. J., Sauer-Zavala, S. E., Murray, H. W., Fortune, M., & Barlow, D. H. (2013). Anxiety sensitivity and interoceptive exposure: A transdiagnostic construct and change strategy. Behavior Therapy, 44, 417–431. Boswell, J. F., Thompson-Hollands, J., Farchione, T. J., & Barlow, D. H. (2013). Intolerance of uncertainty: A common factor in the treatment of emotional disorders. Journal of Clinical Psychology, 69, 630–645. doi:10.1002/jclp.21965 Bouton, M., Mineka, S., & Barlow, D. H. (2001). Modern learning theory perspective on the etiology of panic disorder. Psychological Review, 108, 4–32. Brown, K., & Ryan, R. (2003). The benefits of being present: Mindfulness and its role in psychological well-being. Journal of Personality and Social Psychology, 84, 822–848. Brown, K., Weinstein, N., & Creswell, D. (2011). Trait mindfulness modulates neuroendocrine and affective responses to social-evaluative threat. Psychoneuroendocrinology, 37, 2037–2041. Brown, T. A. (2007). Temporal course and structural relationships among dimensions of temperament and DSM–IV anxiety and mood disorder constructs. Journal of Abnormal Psychology, 116, 313–328. Brown, T. A., Antony, M. M., & Barlow, D. H. (1995). Diagnostic comorbidity in panic disorder: Effect on treatment outcome and course of comorbid diagnoses following treatment. Journal of Consulting and Clinical Psychology, 63, 408–418. Brown, T. A., & Barlow, D. H. (1995). Long-term outcome in cognitive-behavioral treatment of panic disorder: Clinical predictors and alternative strategies for assessment. Journal of Consulting and Clinical Psychology, 63, 754–765. Brown, T. A., & Barlow, D. H. (2002). Classification of anxiety and mood disorders. In D. H. Barlow (Ed.), Anxiety and its disorders: The nature and treatment of anxiety and panic (2nd ed., pp. 292–327). New York, NY: Guilford Press. Brown, T. A., & Barlow, D. H. (2005). Dimensional versus categorical classification of mental disorders in the fifth edition of Diagnostic and Statistical Manual of Mental Disorders and beyond: Comment on the special section. Journal of Abnormal Psychology, 114, 551–556. Brown, T. A., & Barlow, D. H. (2009). A proposal for a dimensional classification system based on the shared features of the DSM-IV anxiety and mood disorders: Implications for assessment and treatment. Psychological Assessment, 21, 256–271. Brown, T. A., Campbell, L. A., Lehman, C. L., Grisham, J. R., & Mancill, R. B. (2001). Current and lifetime comorbidity of the DSM-IV anxiety and mood disorders in a large clinical sample. Journal of Abnormal Psychology, 110, 49–58.

380 David H. Barlow, et al. Brown, T. A., Chorpita, B. F., & Barlow, D. H. (1998). Structural relationships among dimensions of the DSM-IV anxiety and mood disorders and dimensions of negative affect, positive affect, and autonomic arousal. Journal of Abnormal Psychology, 107, 179–192. Brown, T. A., & McNiff, J. (2009). Specificity of autonomic arousal to DSM-IV panic disorder and post-traumatic stress disorder. Behaviour Research and Therapy, 47, 487–493. Brown, T. A., & Rosellini, A. J. (2011). The direct and interactive effects of neuroticism and life stress on the severity and longitudinal course of depression. Journal of Abnormal Psychology, 120, 844–856. Brozina, K., & Abela, J. (2006). Behavioral inhibition, anxious symptoms, and depressive symptoms: A short-term prospective examination of the diathesis-stress model. Behaviour Research and Therapy, 44, 1337–1346. Bullis, J. R., Fortune, M. R., Farchione, T. J., & Barlow, D. H. (2013). Long-term outcome of the unified protocol for the transdiagnostic treatment of emotional disorders. Manuscript submitted for publication. Butler, L., & Nolen-Hoeksema, S. (1994). Gender differences in responses to depressed mood in a college sample. Sex Roles, 30, 331–346. Calmes, C., & Roberts, J. (2007). Repetitive thought and emotional distress: Rumination and worry as prospective predictors of depressive and anxious symptomatology. Cognitive Therapy and Research, 31, 343–356. Campbell-Sills, L., Barlow, D., Brown, T., & Hofman, S. (2006). Acceptability and suppression of negative emotion in anxiety and mood disorders. Emotion, 6, 587–595. Campbell-Sills, L., Liverant, G. I., & Brown, T. A. (2004). Psychometric evaluation of the behavioral inhibition/behavioral activation scales in a large sample of outpatients with anxiety and mood disorders. Psychological Assessment, 16, 244–254. Carl, J. R., Gallagher, M. W., Sauer-Zavala, S. E., Bentley, K. H., & Barlow, D. H. (2014). A preliminary examination of the effects of the unified protocol on temperament. Comprehensive Psychiatry, 55, 1426–1434. Cash, M., & Whittingham, K. (2010). What facets of mindfulness contribute to psychological well-being and depressive, anxious, and stress-related symptomatology? Mindfulness, 1, 177–182. Cheavens, J., & Heiy, J. (2011). The differential roles of affect and avoidance in major depressive and borderline personality disorder symptoms. Journal of Social and Clinical Psychology, 30, 441–457. Cheavens, J., Rosenthal, M., Daughters, S., Novak, J., Kosson, D., & Lynch, T. (2005). An analogue investigation of the relationships among perceived parental criticism, negative affect, and borderline personality disorder features: The role of thought suppression. Behaviour Research and Therapy, 43, 257–268. Chorpita, B. F., Albano, A. M., & Barlow, D. H. (1998). The structure of negative emotions in a clinical sample of children and adolescents. Journal of Abnormal Psychology, 107, 74–85. Chorpita, B. F., Brown, T. A., & Barlow, D. H. (1998). Perceived control as a mediator of family environment in etiological models of childhood anxiety. Behavior Therapy, 29, 457–476. Clark, D. M. (1986). A cognitive approach to panic. Behaviour Research and Therapy, 24, 461–470. Clark, L. A. (2005). Temperament as a unifying basis for personality and psychopathology. Journal of Abnormal Psychology, 114, 505–521. Clark, L. A. (2009). Stability and change in personality disorder. Current Directions in Psychological Science, 18, 27–31. Clark, L. A., Vittengl, J., Kraft, B., & Jarrett, D. (2003). Separate personality traits from states to predict depression. Journal of Personality Disorders, 17, 152–172.

The Nature, Diagnosis, and Treatment of Neuroticism 381 Clark, L. A., & Watson, D. (1991). Tripartite model of anxiety and depression: Psychometric evidence and taxonomic implications. Journal of Abnormal Psychology, 103, 103–116. Clark, L. A., & Watson, D. (2008). Temperament: An organizing paradigm for trait psychology. In O. P. John, R. R. Robins, & L. A. Pervin (Eds.), Handbook of personality: Theory and research (3rd ed.). New York, NY: Guilford Press. Clark, L. A., Watson, D., & Mineka, S. (1994). Temperament, personality, and the mood and anxiety disorders. Journal of Abnormal Psychology, 103, 103–116. Collimore, K., McCabe, R., Carelton, N., & Asmundson, G. (2008). Media exposure and dimensions of anxiety sensitivity: Differential associations with PTSD symptom clusters. Journal of Anxiety Disorders, 22, 1021–1028. Cox, B., Enns, M., Walker, J., Kjernisted, K., & Pidlubny, S. (2001). Psychological vulnerabilities in patients with major depression versus panic disorder. Behaviour Research and Therapy, 39, 567–573. Cox, B., Taylor, S., & Enns, M. (1999). Fear of cognitive dyscontrol in relation to depressive symptoms: Comparisons between original and alternative measures of anxiety sensitivity. Journal of Behavior Therapy and Experimental Psychiatry, 30, 301–311. Creswell, D., Way, B., Eisenberger, N., & Leiberman, M. (2007). Neural correlates of dispositional mindfulness during affect labeling. Psychosomatic Medicine, 69, 560–565. Cuijpers, P., Smit, F., Penninx, B. W. J. H., de Graaf, R., ten Have, M., & Beekman, A. T. F. (2010). Economic costs of neuroticism: A population based study. Archives of General Psychiatry, 67, 1086–1093. Digman, J. (1990). Personality structure: Emergence of the five-factor model. Annual Review of Psychology, 41, 417–440. Eaton, N., Krueger, R., & Oltmanns, T. (2011). Aging and the structure and long-term stability of the internalizing spectrum of personality and psychopathology. Psychology and Aging, 26, 987–993. Ellard, K. K., Fairholme, C. P., Boisseau, C. L., Farchione, T. J., & Barlow, D. H. (2010). Unified protocol for the transdiagnostic treatment of emotional disorders: Protocol development and initial outcome data. Cognitive and Behavioral Practice, 17, 88–101. doi:10.1016/j.cbpra.2009.06.002 Etkin, A., Prater, K. E., Hoeft, F., Menon, V., & Schatzberg, A. F. (2010). Failure of anterior cingulate activation and connectivity with the amygdala during implicit regulation of emotional processing in generalized anxiety disorder. American Journal of Psychiatry, 167, 545–554. doi:10.1176/appi.ajp.2009.09070931 Etkin, A., & Wager, T. D. (2007). Functional neuroimaging of anxiety: A meta-analysis of emotional processing in PTSD, social anxiety disorder, and specific phobia. American Journal of Psychiatry, 164, 1476–1488. doi:10.1176/appi. ajp.2007.07030504 Eysenck, H. J. (1947). Dimensions of personality. Oxford, England: Kegan Paul. Eysenck, H. J. (1961). The handbook of abnormal psychology. New York, NY: Basic Books. Eysenck, H. J. (1981). A model for personality. New York, NY: Springer-Verlag. Eysenck, H. J., & Eysenck, S. B. G. (1975). Manual of the Eysenck Personality Questionnaire (adult and junior). London, England: Hodder & Stoughton. Farchione, T. J., Fairholme, C. P., Ellard, K. K., Boisseau, C. L., Thompson-Hollands, J., Carl, J., . . . Barlow, D. H. (2012). Unified protocol for the transdiagnostic treatment of emotional disorders: A randomized controlled trial. Behavior Therapy, 43, 666–678. Freud, S. (1924). Collected papers. New York, NY: International Psychoanalytic Press. Fu, C. H. Y., Williams, S. C. R., Cleare, A. J., Brammer, M. J., Walsh, N. D., Kim, J., . . . Bullmore, E. T. (2004). Attenuation of the neural response to sad faces in major depression by antidepressant treatment. Archives of General Psychiatry, 61, 877–889. Gallagher, M. W., Payne, L. P., White, K. S., Shear, K. M., Woods, S. W., Gorman, J. M., & Barlow, D. H. (2013). Mechanisms of change in cognitive-behavioral therapy for panic

382 David H. Barlow, et al. disorder: The unique effects of self-efficacy and anxiety sensitivity. Behavior Research and Therapy, 51, 767–777. Gershuny, B. S., & Sher, K. J. (1998). The relation between personality and anxiety: Findings from a 3-year prospective study. Journal of Abnormal Psychology, 107, 252–262. Goldberg, L. (1993). The structure of phenotypic personality traits. American Psychologist, 48, 26–34. Gray, J. A. (1982). The neuropsychology of anxiety. New York, NY: Oxford University Press. Gray, J. A. (1991). The neuropsychology of temperament. In J. Strelau & A. Angleitner (Eds.), Explorations in temperament: International perspectives in theory and measurement (pp. 105–128). New York, NY: Plenum. Gray, J. A., & McNaughton, N. (1996). The neuropsychology of anxiety: A reprise. In D. A. Hope (Ed.), Nebraska Symposium on Motivation: Vol. 43. Perspectives on anxiety, panic, and fear (pp. 61–134). Lincoln: University of Nebraska Press Griffith, J. W., Zinbarg, R. E., Craske, M. G., Mineka, S., Rose, R. D., Waters, A. M., & Sutton, J. M. (2010). Neuroticism as a common dimension in the internalizing disorders. Psychological Medicine, 40, 1125–1136. Harmer, C. J., Mackay, C. E., Reid, C. B., Cowen, P. J., & Goodwin, G. M. (2006). Antidepressant drug treatment modifies the neural processing of nonconscious threat cues. Biological Psychiatry, 59, 816–820. Harmer, C. J., O’Sullivan, U., Favaron, E., Massey-Chase, R., Ayres, R., Reinecke, A., . . . Cowen, P. J. (2009). Effect of acute antidepressant administration on negative affective bias in depressed patients. American Journal of Psychiatry, 166, 1178–1184. Hayes, S. C., Luoma, J. B., Bond, F. W., Masuda, A., & Lillis, J. (2006). Acceptance and commitment therapy: Model, processes and outcomes. Behaviour Research and Therapy, 44, 1–25. Hayes, S. C., Wilson, K. G., Gifford, E. V., Follette, V. M., & Strosahl, K. (1996). Experiential avoidance and behavioral disorders: A functional dimensional approach to diagnosis and treatment. Journal of Consulting and Clinical Psychology, 64, 1152–1168. Hayward, C., Killen, J. D., Kraemer, H. C., & Taylor, C. B. (1998). Linking self-reported childhood behavioral inhibition to adolescent social phobia. Journal of the American Academy of Child and Adolescent Psychiatry, 37, 1308–1316. Heimberg, R. G., Liebowitz, M. R., & Hope, D. A. (1998). Cognitive behavioral group therapy versus phenelzine therapy for social phobia: 12-week outcome. Archives of General Psychiatry, 55, 1133–1141. Helson, R., Jones, C., & Kwan, V. S. (2002). Personality change over 40 years of adulthood: Hierarchical linear modeling analyses of two longitudinal samples. Journal of Personality and Social Psychology, 83, 752–766. Hirshfeld-Becker, D. R., Biederman, J. F., Henin, A., Faraone, S. V., Davis, S., Harrington, K., & Rosenbaum, J. (2007). Behavioral inhibition in preschool children at risk is a specific predictor of middle childhood social anxiety: A five-year follow-up. Journal of Developmental and Behavioral Pediatrics, 28, 225–233. Hirshfeld-Becker, D. R., Rosenbaum, J. F., Biederman, J. F., Bolduc, E. A., Faraone, S. V., Snidman, N., . . . Kagan, J. (1992). Stable behavioral inhibition and its association with anxiety disorder. Journal of the American Academy of Child and Adolescent Psychiatry, 31, 103–111. Hoehn-Saric, R., Schlund, M. W., & Wong, S. H. Y. (2004). Effects of citalopram on worry and brain activation in patients with generalized anxiety disorder. Psychiatry Research, 131, 11–21. doi:10.1016/j.pscychresns.2004.02.003 Hofmann, S. G., Heering, S., Sawyer, A. T., & Asnaani, A. (2009). How to handle anxiety: The effects of reappraisal, acceptance, and suppression strategies on anxious arousal. Behaviour Research and Therapy, 47, 389–394. doi:10.1016/j.brat.2009.02.010

The Nature, Diagnosis, and Treatment of Neuroticism 383 Holmes, A. J., Lee, P. H., Hollinshead, M. O., Bakst, L., Roffman, J. L., Smoller, J. W., & Buckner, R. I. (2012). Individual differences in amygdala-medial prefrontal anatomy link negative affect, impaired social functioning, and polygenetic depression link. Journal of Neuroscience, 32, 18087–18100. Hong, R. Y. (2007). Worry and rumination: Differential associations with anxious and depressive symptoms and coping behavior. Behaviour Research and Therapy, 45, 277–290. Hudson, J. I., & Pope, H. G. (1990). Affective spectrum disorder: Does antidepressant response identify a family of disorders with a common pathophysiology? American Journal of Psychiatry, 147, 552–564. John, O. P. (1990). The “Big Five” factor taxonomy: Dimensions of personality in the natural language and in questionnaires. In L. A. Pervin (Ed.), Handbook of personality: Theory and research (pp. 66–100). New York, NY: Guilford Press. John, O. P., & Srivastava, S. (1999). The Big Five trait taxonomy: History, measurement, and theoretical perspectives. In L. A. Pervin & O. P. John (Eds.), Handbook of personality: Theory and research (2nd ed., pp. 102–138). New York, NY: Guilford Press. Jylhä, P., & Isometsä, E. (2006). The relationship of neuroticism and extraversion to symptoms of anxiety and depression in the general population. Depression and Anxiety, 23, 281–289. Kabat-Zinn, J. (1982). An outpatient program in behavioral medicine for chronic pain patients based on the practice of mindfulness meditation: Theoretical considerations and preliminary results. General Hospital Psychiatry, 4, 33–47. Kagan, J. (1989). Temperamental contributions to social behavior. American Psychologist, 44, 668–674. Kagan, J. (1994). Galen’s prophecy: Temperament in human nature. New York, NY: Basic Books. Kagan, J., & Snidman, N. (1991). Infant predictors of inhibited and uninhibited profiles. Psychological Science, 2, 40–44. Kasch, K. L., Rottenberg, J., Arnow, B. A., & Gotlib, I. H. (2002). Behavioral activation and inhibition systems and the severity and course of depression. Journal of Abnormal Psychology, 111, 589–597. Kashdan, T. B., Breen, W. E., Afram, A., & Terhar, D. (2010). Experiential avoidance in idiographic, autobiographical memories: Construct validity and links to social anxiety, depressive, and anger symptoms. Journal of Anxiety Disorders, 24, 528–534. Keightley, M. L., Seminowicz, D. A., Bagby, R. M., Costa, P. T., Fossati, P., & Mayberg, H. S. (2003). Personality influences limbic-cortical interactions during sad mood induction. NeuroImage, 20, 2031–2039. Kendell, R. E. (1975). The role of diagnosis in psychiatry. Oxford, England: Blackwell Scientific. Kennedy, S. J., Rapee, R. M., & Edwards, S. L. (2009). A selective intervention program for inhibited preschool-aged children of parents with an anxiety disorder: Effects on current anxiety disorders and temperament. Journal of the American Academy of Child and Adolescent Psychiatry, 48, 602–609. doi:10.1097/CHI.0b013e31819f6fa9 Kessler, R. C., Berglund, P., & Demler, O. (2003). The epidemiology of major depressive disorder: Results from the National Comorbidity Survey Replication (NCS-R). Journal of the American Medical Association, 289, 3095–3105. Kessler, R. C., Cox, B. J., Green, J. G., Ormel, J., McLaughlin, K. A., Merikangas, K. R., . . . Zaslavsky, A.M. (2011). The effects of latent variables in the development of comorbidity among common mental disorders. Depression and Anxiety, 28, 29–39. Kessler, R. C., Gruber, M., Hettema, J. M., Hwang, I., Sampson, N., & Yonkers, K. A. (2008). Comorbid major depression and generalized anxiety disorders in the National Comorbidity Survey follow-up. Psychological Medicine, 38, 365–374.

384 David H. Barlow, et al. Kessler, R. C., Nelson, C. B., McGonagle, K. A., Lui, J., Swartz, M., & Blazer, D. G. (1996). Comorbidity of DSM-III-R major depressive disorder in the general population: Results from the National Comorbidity Survey. British Journal of Psychiatry, 168, 17–30. Kessler, R. C., Stang, P. E., Wittchen, H. U., Ustun, T. B., Roy-Byrne, P. P., & Walters, E. E. (1998). Lifetime panic-depression comorbidity in the National Comorbidity Survey. Archives of General Psychiatry, 55, 801–808. Kotov, R., Watson, D., Robles, J. P., & Schmidt, N. B. (2007). Personality traits and anxiety symptoms: The multilevel trait predictor model. Behaviour Research and Therapy, 45, 1485–1503. Krueger, R. F. (1999). The structure of common mental disorders. Archives of General Psychiatry, 56, 921–926. Lahey, B. B. (2009). Public health significance of neuroticism. American Psychologist, 64, 241–256. Lee, J. K., Orsillo, S. M., Roemer, L., & Allen, L. B. (2010). Distress and avoidance in generalized anxiety disorder: Exploring relationships with intolerance of uncertainty and worry. Cognitive Behaviour Therapy, 39, 126–136. Lewis, A. (1970). The ambiguous word “anxiety.” International Journal of Psychiatry, 9, 62–79. Lilienfeld, S. O. (1999). Anxiety sensitivity and the structure of personality. In S. Taylor (Ed.), Anxiety sensitivity: Theory, research and treatment of the fear of anxiety (pp. 149–182). Mahwah, NJ: Erlbaum. Lorberbaum, J. P., Kose, S., Johnson, M. R., Arana, G. W., Sullivan, L. K., Hamner, M. B., . . . George, M. S. (2004). Neural correlates of speech anticipatory anxiety in generalized social phobia. NeuroReport, 15, 2701–2705. Lyubomirsky, S., & Nolen-Hoeksema, S. (1995). Effects of self-focused rumination on negative thinking and interpersonal problem solving. Journal of Personality and Social Psychology, 69, 176–190. Lyubomirsky, S., Tucker, K., Caldwell, N., & Berg, K. (1999). Why ruminators are poor problem solvers: Clues from the phenomenology of dysphoric rumination. Journal of Personality and Social Psychology, 77, 1041–1060. Maack, D. J., Tull, M. T., & Gratz, K. L. (2012). Experiential avoidance mediates the association between behavioral inhibition and posttraumatic stress disorder. Cognitive Therapy and Research, 36, 407–416. MacLeod, C., & Mathews, A. (2012). Cognitive bias modification approaches to anxiety. Annual Review of Clinical Psychology, 8, 189–217. Maller, R. G., & Reiss, S. (1992). Anxiety sensitivity in 1984 and panic attacks in 1987. Journal of Anxiety Disorders, 6, 214–247. Manos, R. C., Kanter, J. W., & Busch, A.M. (2010). A critical review of assessment strategies to measure the behavioral activation model of depression. Clinical Psychology Review, 30, 547–561. Mata, J., Thompson, R. J., Jaeggi, S. M., Buschkuehl, M., Jonides, J., & Gotlib, I. (2012). Walk on the bright side: Physical activity and affect in major depressive disorder. Journal of Abnormal Psychology, 121, 297–308. doi:10.1037/a0023522 Mathews, A., & McLeod, C. (2005). A cognitive vulnerability to emotional disorders. Annual Review of Clinical Psychology, 1, 167–195. Mayberg, H. S., Liotti, B. M., Brannan, S. K., McGinnis, S., Mahurin, R. K., Jerabek, P. A., . . . Fox, P. T. (1999). Reciprocal limbic-cortical function and negative mood: Converging PET findings in depression and normal sadness. American Journal of Psychiatry, 156, 675–682.

The Nature, Diagnosis, and Treatment of Neuroticism 385 McCabe, C., Mishor, Z., Cowen, P. J., & Harmer, C. J. (2010). Diminished neural processing of aversive and rewarding stimuli during selective serotonin reuptake inhibitor treatment. Biological Psychiatry, 67, 439–445. McCrae, R. R., & Costa, P. T. (1987). Validation of the five-factor model of personality across instruments and observers. Journal of Personality and Social Psychology, 52, 81–90. McHugh, R. K., & Barlow, D. H. (2010). The dissemination and implementation of evidence-based psychological treatments: A review of current efforts. American Psychologist, 65, 73–84. McLaughlin, K. A., Mennin, D. S., & Farach, F. J. (2007). The contributory role of worry in emotion generation and dysregulation in generalized anxiety disorder. Behaviour Research and Therapy, 45, 1735–1752. doi:10.1016/j.brat.2006.12.004 McNally, R. J. (1996). Methodological controversies in the treatment of panic disorder. Journal of Consulting and Clinical Psychology, 64, 88–91. Mennin, D. S., Heimberg, R. G., Turk, C. L., & Fresco, D. M. (2005). Preliminary evidence for an emotion dysregulation model of generalized anxiety disorder. Behaviour Research and Therapy, 43, 1281–1310. doi:10.1016/j.brat.2004.08.008 Merikangas, K. R., Zhang, H., & Aveneoli, S. (2003). Longitudinal trajectories of depression and anxiety in a prospective community study. Archives of General Psychiatry, 60, 993–1000. Meyer, B., Johnson, S. L., & Winters, R. (2001). Responsiveness to threat and incentive in bipolar disorder: Relations of the BIS/BAS scales with symptoms. Journal of Psychopathology and Behavioral Assessment, 23, 133–143. Mineka, S., Watson, D., & Clark, L. A. (1998). Comorbidity of anxiety and unipolar mood disorders. Annual Review of Psychology, 49, 377–412. Moore, S. A., Zoellner, L. A., & Mollenholt, N. (2008). Are expressive suppression and cognitive reappraisal associated with stress-related symptoms? Behaviour Research and Therapy, 46, 993–1000. doi:10.1016/j.brat.2008.05.001 Mroczek, D., & Spiro, A. (2003). Modeling intraindividual change in personality traits: Findings from the Normative Aging Study. Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 58, 153–165. Murphy, S. E., Yiend, J., Lester, K. J., Cowen, P. J., & Harmer, C. J. (2009). Short-term serotonergic but not noradrenergic antidepressant administration reduces attentional vigilance to threat in healthy volunteers. International Journal of Neuropsychopharmacology, 12, 169–179. Naragon-Gainey, K. (2010). Meta analysis of the relations of anxiety sensitivity to the depressive and anxiety disorders. Psychological Bulletin, 136, 128–150. Naragon-Gainey, K., Gallagher, M. W., & Brown, T. A. (2013). Stable “trait” variance of temperament as a predictor of the temporal course of depression and social phobia. Journal of Abnormal Psychology, 122, 611–623. Nathan, P. E., & Gorman, J. M. (Eds.). (2007). A guide to treatments that work (3rd ed.). London, England: Oxford University Press. Nemeroff, C. (2013). Psychoneuroimmunoendocrinology: The biological basis of mind-body physiology and pathophysiology. Depression and Anxiety, 30, 285–287. doi:10.1002/da.22110 Newman, M., & Llera, S. (2011). A novel theory of experiential avoidance in generalized anxiety disorder: A review and synthesis of research supporting a contrast avoidance model of worry. Clinical Psychology Review, 31, 371–382. Nolen-Hoeksema, S. (1991). Responses to depression and their effects on the duration of depressive episodes. Journal of Abnormal Psychology, 100, 569–582.

386 David H. Barlow, et al. Nolen-Hoeksema, S. (2000). The role of rumination in depressive disorders and mixed anxiety/depressive symptoms. Journal of Abnormal Psychology, 109, 504–511. Nolen-Hoeksema, S., Larson, J., & Grayson, C. (1999). Explaining the gender difference in depressive symptoms. Journal of Personality and Social Psychology, 77, 1061–1072. Nolen-Hoeksema, S., Wisco, B. E., & Lyubomirsky, S. (2008). Rethinking rumination. Perspectives on Psychological Science, 3, 400–424. Norton, G. R., Rockman, G. E., Ediger, J., Pepe, C., Goldberg, S., Cox, B. J., & Asmundson, G. J. G. (1997). Anxiety sensitivity and drug choice in individuals seeking treatment for substance abuse. Behaviour Research and Therapy, 35, 859–862. O’Connor, D. B., O’Connor, R. C., & Marshall, R. (2007). Perfectionism and psychological distress: Evidence of the mediating effects of rumination. European Journal of Personality, 21, 429–452. Oltmanns, T. F., & Balsis, S. (2011). Personality disorders in later life: Questions about the measurement, course, and impact of disorders. Annual Review of Clinical Psychology, 7, 321–349. Paquette, V., Levesque, J., Mensour, B., Leroux, J. M., Beaudoin, G., Bourgouin, P., & Beauregard, M. (2003). “Change the mind and you change the brain”: Effects of cognitive-behavioral therapy on the neural correlates of spider phobia. NeuroImage, 18, 401–409. Paulesu, E., Sambugaro, E., Torti, T., Danelli, L., Ferri, F., Scialfa, G., . . . Sassaroli, S. (2010). Neural correlates of worry in generalized anxiety disorder and in normal controls: A functional MRI study. Psychological Medicine, 40, 117–124. doi:10.1017/ S0033291709005649 Phan, K. L., Fitzgerald, D. A., Nathan, P. J., & Tancer, M. E. (2006). Association between amygdala hyperactivity to harsh faces and severity of social anxiety in generalized social phobia. Biological Psychiatry, 59, 424–429. doi:10.1016/j.biopsych.2005.08.012 Pickett, S. M., Lodis, C. S., Parkhill, M. R., & Orcutt, H. K. (2012). Personality and experiential avoidance: A model of anxiety sensitivity. Personality and Individual Differences, 53, 246–250. doi:10.1016/j.paid.2012.03.031 Plehn, K., & Peterson, R. A. (2002). Anxiety sensitivity as a predictor of the development of panic symptoms, panic attacks, and panic disorder: A prospective study. Journal of Anxiety Disorders, 16, 455–473. doi:10.1016/S0887–6185(02)00129–9 Porto, P. R., Oliveira, L., Mari, J., Volchan, E., Figueira, I., & Ventura, P. (2009). Does cognitive behavioral therapy change the brain? A systematic review of neuroimaging in anxiety disorders. Journal of Neuropsychiatry and Clinical Neurosciences, 21, 114–125. doi:10.1176/appi.neuropsych.21.2.114 Purdon, C. (1999). Thought suppression and psychopathology. Behaviour Research and Therapy, 37, 1029–1054. doi:10.1016/S0005–7967(98)00200–9 Quilty, L. C., Meusel, L. A., & Bagby, R. M. (2008). Neuroticism as a mediator of treatment response to SSRIs in major depressive disorder. Journal of Affective Disorders, 111, 67–73. doi:10.1016/j.jad.2008.02.006 Rachman, S., & de Silva, P. (1978). Abnormal and normal obsessions. Behaviour Research and Therapy, 16, 233–248. doi:10.1016/0005–7967(78)90022–0 Rapee, R. M., Kennedy, S., Ingram, M., Edwards, S., & Sweeney, L. (2005). Prevention and early intervention of anxiety disorders in inhibited preschool children. Journal of Consulting and Clinical Psychology, 73, 488–497. doi:10.1037/0022–006X.73.3.488 Rapee, R. M., Kennedy, S. J., Ingram, M., Edwards, S. L., & Sweeney, L. (2010). Altering the trajectory of anxiety in at-risk young children. American Journal of Psychiatry, 167, 1518–1525. doi:10.1176/appi.ajp.2010.09111619

The Nature, Diagnosis, and Treatment of Neuroticism 387 Rasmussen, M. K., & Pidgeon, A.M. (2011). The direct and indirect benefits of dispositional mindfulness on self-esteem and social anxiety. Anxiety, Stress and Coping: An International Journal, 24, 227–233. doi:10.1080/10615806.2010.515681 Rassin, E., Muris, P., Schmidt, H., & Merkelbach, H. (2000). Relationship between thought action fusion, thought suppression, and obsessive-compulsive symptoms: A structural equation model approach. Behaviour Research and Therapy, 38, 889–897. doi:10.1016/ S0005–7967(99)00104–7 Rassovsky, Y., Kushner, M. G., Schwarze, N. J., & Wangensteen, O. D. (2000). Psychological and physiological predictors of response to carbon dioxide challenge in individuals with panic disorders. Journal of Abnormal Psychology, 109, 616–623. doi:10.1037 //0021–843X.109.4.616 Reardon, J. M., & Williams, N. L. (2007). The specificity of cognitive vulnerabilities to emotional disorders: Anxiety sensitivity, looming vulnerability, and explanatory style. Journal of Anxiety Disorders, 21, 625–643. doi:10.1016/j.janxdis.2006.09.013 Reiss, S. (1991). Expectancy model of fear, anxiety, and panic. Clinical Psychology Review, 11, 141–153. doi:10.1016/0272–7358(91)90092–9 Reiss, S. (1997). Trait anxiety: It’s not what you think it is. Journal of Anxiety Disorders, 11, 201–214. doi:10.1016/S0887–6185(97)00006–6 Reiss, S., Peterson, R. A., Gursky, D. M., & McNally, R. J. (1986). Anxiety sensitivity, anxiety frequency, and the prediction of fearfulness. Behaviour Research and Therapy, 24, 1–8. doi:10.1016/0005–7967(86)90143–9 Roberts, B. W., & Mroczeck, D. (2008). Personality trait change in adulthood. Current Directions in Psychological Science, 17, 31–35. doi:10.1111/j.1467–8721.2008.00543.x Roberts, B. W., Walton, K. E., & Viechtbauer, W. (2006). Patterns of mean-level change in personality traits across the life course: A meta-analysis of longitudinal studies. Psychological Bulletin, 132, 1–25. doi:10.1037/0033–2909.132.1.1 Robinson, J. L., Kagan, J., Reznick, J. S., & Corley, R. (1992). The heritability of inhibited and uninhibited behavior: A twin study. Developmental Psychology, 28, 1030–1037. doi:10.1037/0012–1649.28.6.1030 Roemer, L., Salters, K., Raffa, S., & Orsillo, S. M. (2005). Fear and avoidance of internal experiences in GAD: Preliminary tests of a conceptual model. Cognitive Therapy and Research, 29, 71–88. doi:10.1007/s10608–005–1650–2 Rosellini, A. J. (2013). Initial development and validation of a dimensional classification system for the emotional disorders (Unpublished doctoral dissertation). Boston University, Boston, MA. Rosellini, A. J., Lawrence, A., Meyer, J., & Brown, T. A. (2010). The effects of extraverted temperament on agoraphobia in panic disorder. Journal of Abnormal Psychology, 119, 420–426. doi:10.1037/a0018614 Sarin, S., Abela, J., & Auerbach, R. (2005). Response styles theory of depression: A test of specificity and causal mediation. Cognition & Emotion, 19, 751–761. doi:10.1080/02699930441000463 Sauer, S. E., & Baer, R. A. (2009). Responding to negative internal experience: Relationships between acceptance and change-based approaches and psychological adjustment. Journal of Psychopathology and Behavioral Assessment, 31, 378–386. doi:10.1007/ s10862–009–9127–3 Sauer-Zavala, S., Boswell, J. F., Gallagher, M. W., Bentley, K. H., Ametaj, A., & Barlow, D. H. (2012). The role of negative affectivity and negative reactivity to emotions in predicting outcomes in the unified protocol for the transdiagnostic treatment of emotional disorders. Behaviour Research and Therapy, 50, 551–557. doi:10.1016/j.brat.2012.05.005

388 David H. Barlow, et al. Schmidt, N. B., Keough, M. E., Timpano, K. R., & Richey, J. A. (2008). Anxiety sensitivity profile: Predictive and incremental validity. Journal of Anxiety Disorders, 22, 1180–1189. doi:10.1016/j.janxdis.2007.12.003 Schwartz, C. E., Snidman, N., & Kagan, J. (1999). Adolescent social anxiety as an outcome of inhibited temperament in childhood. Journal of the American Academy of Child and Adolescent Psychiatry, 38, 1008–1015. doi:10.1097/00004583–199908000–00017 Segal, Z. V., Williams, J. M. G., & Teasdale, J. D. (2002). Mindfulness-based cognitive therapy for depression: A new approach to preventing relapse. New York, NY: Guilford Press. Segerstrom, S. C., Tsao, J. C. I., Alden, L. E., & Craske, M. G. (2000). Worry and rumination: Repetitive thought as a concomitant and predictor of negative mood. Cognitive Therapy and Research, 24, 671–688. doi:10.1023/A:1005587311498 Selby, E. A., Anestis, M. D., & Joiner, T. E. (2008). Understanding the relationship between emotional and behavioral dysregulation: Emotional cascades. Behaviour Research and Therapy, 46, 593–611. doi:10.1016/j.brat.2008.02.002 Shahar, B., & Herr, N. R. (2011). Depressive symptoms predict inflexibly high levels of experiential avoidance in response to daily negative affect: A daily diary study. Behaviour Research and Therapy, 49, 676–681. doi:10.1016/j.brat.2011.07.006 Shin, L. M., & Liberzon, I. (2010). The neurocircuitry of fear, stress, and anxiety disorders. Neuropsychopharmacology, 35, 169–191. doi:10.1038/npp.2009.83 Shin, L. M., Wright, C. I., Cannistraro, P. A., Wedig, M. M., McMullin, K., Martis, B., . . . Rauch, S. L. (2005). A functional magnetic resonance imaging study of amygdala and medial prefrontal cortex responses to overtly presented fearful faces in posttraumatic stress disorder. Archives of General Psychiatry, 62, 273–281. doi:10.1001/archpsyc.62.3.273 Slater, E. (1943). The neurotic constitution: A statistical study of two thousand neurotic soldiers. Journal of Neurology and Psychiatry, 6, 1–16. doi:10.1136/jnnp.6.1–2.1 Small, B. J., Hertzog, C., Hultsch, D. F., & Dixon, R. A. (2003). Stability and change in adult personality of 6 years: Findings from the Victoria Longitudinal Study. Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 58, 166–176. doi:10.1093/ geronb/58.3.P166 Soskin, D. P., Carl, J. R., Alpert, J., & Fava, M. (2012). Antidepressant effects on emotional temperament: Toward a biobehavioral research paradigm for major depressive disorder. CNS Neuroscience & Therapeutics, 18, 441–451. doi:10.1111/j.1755–5949.2012. 00318.x Speisman, R. B., Kumar, A., Rani, A., Foster, T. C., & Omerod, B. K. (2012). Daily exercise improves memory, stimulates hippocampal neurogenesis and modulates immune and neuroimmune cytokines in aging rats. Brain, Behavior, and Immunity, 28, 25–43. Spijker, J., de Graf, R., Oldehinkel, A. J., Nolen, W. A., & Ormel, J. (2007). Are the vulnerability effects of personality and psychosocial functioning on depression accounted for by subthreshold symptoms? Depression and Anxiety, 24, 472–478. doi:10.1002/da.20252 Straube, T., Mentzel, H. J., & Miltner, W. H. (2006). Neural mechanisms of automatic and direct processing of phobogenic stimuli in specific phobia. Biological Psychiatry, 59, 162–170. doi:10.1016/j.biopsych.2005.06.013 Strupp, H. (1973). The future of research in psychotherapy. In H. Strupp (Ed.), Psychotherapy: Clinical, research, and theoretical issues (pp. 733–756). Lanham, MD: Jason Aronson. doi:10.1037/11523–027 Tang, T. Z., DeRubeis, R. J., Hollon, S. D., Amsterdam, J., Shelton, R., & Schalet, B. (2009). Personality change during depression treatment: A placebo-controlled trial. Archives of General Psychiatry, 66, 1322–1330. doi:10.1001/archgen psychiatry.2009.166 Taylor, S. (1999). Anxiety sensitivity: Theory, research and treatment of the fear of anxiety. Mahwah, NJ: Erlbaum.

The Nature, Diagnosis, and Treatment of Neuroticism 389 Tellegen, A. (1985). Structures of mood and personality and their relevance to assessing anxiety, with an emphasis on self-report. In A. H. Tuma & J. D. Maser (Eds.), Anxiety and the anxiety disorders (pp. 681–706). Hillsdale, NJ: Erlbaum. Tillfors, M., Furmark, T., Marteinsdottir, I., & Fredrikson, M. (2002). Cerebral blood flow during anticipation of public speaking in social phobia: A PET study. Biological Psychiatry, 52, 1113–1119. doi:10.1016/S0006–3223(02)01396–3 Tolin, D. F., Abramowitz, J. S., Brigidi, B. D., & Foa, E. B. (2003). Intolerance of uncertainty in obsessive-compulsive disorder. Journal of Anxiety Disorders, 17, 233–242. doi:10.1016/ S0887–6185(02)00182–2 Tomarken, A. J., Dichter, G. S., Freid, C., Addington, S., & Shelton, R. C. (2004). Assessing the effects of bupropion SR on mood dimensions of depression. Journal of Affective Disorders, 78, 235–241. doi:10.1016/S0165–0327(02)00306–3 Tsao, J. C. I., Lewin, M. R., & Craske, M. G. (1998). The effects of cognitive-behavior therapy for panic disorders on comorbid conditions. Journal of Anxiety Disorders, 12, 357–371. doi:10.1016/S0887–6185(98)00020–6 Tsao, J. C. I., Mystkowski, J. L., & Zucker, B. G. (2002). Effects of cognitive-behavioral therapy for panic disorder on comorbid conditions: Replication and extension. Behavior Therapy, 33, 493–509. doi:10.1016/S0005–7894(02)80013–2 Tull, M. T., Gratz, K. L., Salters, K., & Roemer, L. (2004). The role of experiential avoidance in posttraumatic stress symptoms and symptoms of depression, anxiety, and somatization. Journal of Nervous and Mental Disease, 192, 754–761. doi:10.1097/01. nmd.0000144694.30121.89 Tull, M. T., & Roemer, L. (2007). Emotion regulation difficulties associated with the experience of uncued panic attacks: Evidence of experiential avoidance, emotional nonacceptance, and decreased emotional clarity. Behavior Therapy, 38, 378–391. doi:10.1016/j. beth.2006.10.006 Turk, C. L., Heimberg, R. G., Luterek, J. A., Mennin, D. S., & Fresco, D. M. (2005). Emotion dysregulation in generalized anxiety disorder: A comparison with social anxiety disorder. Cognitive Therapy and Research, 29, 89–106. doi:10.1007/s10608–005– 1651–1 Tyrer, P. J. (1989). Classification of neurosis. Chichester, England: John Wiley & Sons. Watson, D. (2005). Rethinking mood and anxiety disorders: A quantitative hierarchical model for DSM-V. Journal of Abnormal Psychology, 114, 522–536. doi:10.103 7/0021–843X.114.4.522 Watson, D., & Clark, L. A. (1993). Behavioral disinhibition versus constraint: A dispositional perspective. In D. M. Wegner & J. W. Pennebaker (Eds.), Handbook of mental control (pp. 506–527). New York, NY: Prentice Hall. Watson, D., Clark, L. A., & Carey, G. (1988). Positive and negative affectivity and their relation to anxiety and depressive disorders. Journal of Abnormal Psychology, 97, 346–353. doi:10.1037//0021–843X.97.3.346 Wegner, D. M., Schneider, D. J., Carter, S. R., & White, T. L. (1987). The paradoxical effects of thought suppression. Journal of Personality and Social Psychology, 53, 5–13. doi:10.1037//0022–3514.53.1.5 Weiss, N. H., Tull, M. T., Davis, L. T., Dehon, E. E., Fulton, J. J., & Gratz, K. L. (2012). Examining the association between emotion regulation difficulties and probable posttraumatic stress disorder within a sample of African Americans. Cognitive Behaviour Therapy, 41, 5–14. doi:10.1080/16506073.2011.621970 Widiger, T. A., & Crego, C. (2013). Diagnosis and classification. In I. B. Weiner, G. Stricker, & T. A. Widiger (Eds.), Handbook of psychology: Clinical psychology (2nd ed., pp. 3–18). Hoboken, NJ: John Wiley & Sons.

390 David H. Barlow, et al. Widiger, T. A., Verheul, R., & van den Brink, W. (1999). Personality and psychopathology. In L. A. Pervin & O. P. John (Eds.), Handbook of personality: Theory and research (2nd ed., pp. 347–366). New York, NY: Guilford Press. Zinbarg, R. E., & Barlow, D. H. (1996). Structure of anxiety and the anxiety disorders: A hierarchical model. Journal of Abnormal Psychology, 105, 181–193. doi:10.1037/0021–843X.105.2.181

The Ascendance of Evidence-Based Psychological Treatments

Article 18

Health Care Policy, Psychotherapy Research, and the Future of Psychotherapy David H. Barlow Boston University

In a quest for accountability in the delivery of health services, health care policymakers in both government and private sectors are creating clinical practice guidelines, many of which heavily emphasize medical and pharmacological approaches. Yet, there are now sufficient data to support the efficacy of psychotherapeutic procedures for a wide variety of specific disorders, and it seems now is the appropriate time to communicate these findings to health care policymakers and the public at large. At the same time, data must be developed on the delivery of these interventions by frontline clinicians in the settings where they practice. The emergence of practice research networks may accomplish this latter goal. Long before managed care pierced the hearts and minds of psychologists and other health professionals with radical and wrenching change and very real excesses and abuses, the U.S. government decided that there needed to be more accountability in the provision of health care services. This development was a natural consequence of the spiraling costs of health care and the government’s increasing involvement in paying for health care services—payments that have been administered by various government agencies, such as the Health Care Financing Administration. To this end, in 1989, Congress created the Agency for Health Care Policy and Research (AHCPR). This independent agency within the Public Health Service was organized to be on a level with the National Institutes of Health (NIH). However, this agency has only one overriding purpose: to determine the effectiveness (including the cost-effectiveness) of intervention strategies for specific disorders, with the aim of increasing the quality and reducing the costs of health care. One major mechanism for accomplishing this goal is to promulgate this information in the form of clinical practice guidelines. These guidelines actually begin to articulate, on the basis of existing evidence, the optimal strategies for assessing and treating a variety of problems (e.g., lower back pain) presenting to clinical settings. Psychologists became most aware of these efforts following the publication of a clinical practice guideline on major depressive disorder

394 David H. Barlow

(Depression Guideline Panel, 1993; Muñoz, Hollon, McGrath, Rehm, & VandenBos, 1994). The principal motivation for creating this new government agency, during a time when government is supposed to be shrinking, was a series of in-depth reports to Congress demonstrating that the quality of health care (including behavioral health care) was very uneven. In fact, this congressional foray into influencing and regulating health care delivery has been gathering momentum for decades, most notably in the passage of the Health Maintenance Organization Act in 1973 and in the approval of the concept of diagnostic related groups in 1982. With these developments, government decided it was important to assess the efficacy of interventions and other management strategies for physical and behavioral disorders, and it wasn’t long until health care policymakers, both in other branches of government and in the private sector, followed suit. For example, NIH, through its “consensus conference” mechanism, has published and disseminated information on the efficacy (or the inefficacy) of interventions for a variety of disorders, such as panic disorder (Wolfe & Maser, 1994). As VandenBos (1993) pointed out, as a result of these policy changes, the range of services ultimately included in the evolving system of national health care will be smaller and more selective, and the choice of services will be based on two factors. The first factor will be effectiveness—what is the evidence that certain interventions will relieve suffering or enhance functioning for specific health care problems? The second factor will be cost—from the point of view of public health policy, how expensive is it to implement the policy or to deliver the treatment? But it is in the private sector, in the form of managed care, where efforts to formulate clinical practice guidelines have been abused most often. In this context, almost all mental health professionals have been subjected to brief, anonymous guidelines that do nothing more than limit the number of sessions in an ill-concealed attempt to reduce costs. Even worse, the specific nature of these guidelines is usually unavailable and, therefore, not subject to critical review. In any case, clinical practice guidelines, particularly government guidelines, are rapidly proliferating and assuming the force of law in many states, in that clinicians following these guidelines are not subject to malpractice litigation. The ominous consequences for those who do not follow these guidelines are all too apparent. But the construction of these guidelines is not limited to government agencies; clinical practice guidelines published by the American Psychiatric Association, and appearing at the rate of approximately one or two per year, are also being widely disseminated and are influencing policymakers (Zarin, Pincus, & Mclntyre, 1993). Thus far, guidelines on eating disorders, depression, substance use disorders, and bipolar disorder, among others, have appeared (e.g., American Psychiatric Association, 1995; Zarin et al., 1993). Furthermore, the appearance of clinical practice guidelines is not limited to the United States. In fact, using some of the same meta-analytic methodology and criteria utilized by the AHCPR to construct clinical practice guidelines, the World Health Organization periodically publishes lists of essential treatments for specific disorders. For more than a decade, Australia and New Zealand have jointly

Health Care Policy, Psychotherapy Research, and the Future of Psychotherapy 395

undertaken an extensive quality assurance project (e.g., The Quality Assurance Project, 1983, 1984, 1985). The products of this project are clinical practice guidelines, which are updated periodically, that have been widely adopted by health care policymakers in those countries. Finally in the United Kingdom, the National Health Service recently commissioned a report on psychological interventions, with sufficient evidence for efficacy to be utilized in government health care settings (Roth & Fonagy, 1995). The appearance of these guidelines cannot as yet be considered good news for those who call themselves psychotherapists. Anyone who has read the clinical practice guideline on the treatment of major depressive disorder (Depression Guideline Panel, 1993) has been concerned with the strong emphasis on pharmacological approaches and the relative deemphasis of psychotherapeutic approaches. Reasons for this disparity have been addressed at some length elsewhere (Barlow; 1994; Muñoz et al., 1994) and are not belabored here. The American Psychiatric Association’s guidelines understandably emphasize the effectiveness of pharmacological and other medical approaches to behavioral disorders. Principal reasons for this emphasis include the strong body of evidence from randomized clinical trials underscoring the efficacy of pharmacological approaches with specific disorders, at least in the short term. If left unchecked, these developments ultimately will threaten the status of psychotherapeutic interventions as a principal approach in any health care system, relegating psychotherapy to an afterthought or for the purpose of promoting adherence to medical regimens. Other problematic developments for psychotherapeutic interventions include developing opposition from some of the very consumers psychotherapists are seeking to serve. For instance, one powerful and effective consumer group, the National Alliance for Mental Illness (NAMI), has taken the position that serious psychological disorders, usually defined as most (but not all) disorders identified in the Diagnostic and Statistical Manual of Mental Disorders, 4th Edition (DSM-IV; American Psychiatric Association, 1994), are biologically based and, therefore, should be treated medically and that research on psychotherapeutic approaches may be a waste of time. NAMI and other consumer groups are playing an enormously important role in educating policymakers and the public at large about the economic impact and suffering associated with behavioral disorders. But the message often conveyed is that “true disorders,” as opposed to some minor protestations of the worried well, are not going to respond in any real way to nonmedical approaches such as psychotherapy. Efficacy of Psychological Interventions In light of the medicalization of mental disorders in clinical practice guidelines emphasizing drug treatments, what evidence is there to support the efficacy of psychotherapy? Fortunately, by any current standards of science used in the construction of clinical practice guidelines by government and other agencies, the evidence is now incontrovertible that there are effective psychological interventions for a large number of (but by no means all) psychological disorders (Barlow, 1994; Barlow & Lehman, 1996; Nathan & Gorman, in press;

396 David H. Barlow

Seligman, 1994). Numerous studies and subsequent meta-analyses have demonstrated that any number of specific psychotherapeutic approaches, either alone or, in some cases, in combination with pharmacological approaches, are more effective than credible alternative psychological interventions containing nonspecific factors serving as “psychological placebos.” More importantly, in many cases, the psychotherapeutic procedures are as effective as or more effective than pharmacological approaches with proven efficacy, or they greatly enhance the effects of drugs. These psychological treatments are by no means limited to cognitive-behavioral interventions but range across family systems interventions, interpersonal psychotherapy, and brief psychodynamic interventions for a variety of disorders ranging from anxiety disorders, to schizophrenia, to autism. To take just a few examples, there now are treatments with proven efficacy for the anxiety disorders, including panic disorder with agoraphobia, generalized anxiety disorder, social phobia, specific phobia, obsessive-compulsive disorder, and posttraumatic stress disorder, that meet all existing government standards of evidence (Barlow & Lehman, 1996). There are several highly effective treatments for depression, including cognitive-behavioral approaches and interpersonal psychotherapy (Frank & Spanier, 1995; Jacobson & Hollon, 1996), and these treatments also seem to be effective in preventing the recurrence of major depressive episodes (Frank, Kupfer, Wagner, McEachran, & Cornes, 1991), a topic to which I shall return below. Furthermore, these treatments, when delivered by mental health professionals in primary care settings, are far more effective than usual and customary care provided by primary care physicians, even when these physicians are aware of the diagnosis (Schulberg, Block, & Madonia, 1995). This study articulates nicely with the recent retrospective survey by Consumer Reports (Seligman, 1995) that “suggested” the superiority of doctoral-level mental health clinicians to primary care physicians. That is, the study provides necessary evidence from a randomized, controlled clinical trial on likely reasons for this observation. There are at least two effective treatments for eating disorders: cognitive-behavioral therapy and interpersonal psychotherapy (e.g., Fairburn, Jones, Peveler, Hope, & O’Connor, 1993). Interestingly, interpersonal psychotherapy for eating disorders does not focus directly on eating problems but rather on interpersonal relations that provide the context for eating disorders. Yet, in Fairburn et al.’s study, one year following the end of treatment, interpersonal psychotherapy was just as effective as cognitive-behavioral approaches, and both were more effective than behavior therapy alone. For bipolar disorder, several preliminary studies indicate that the effects of medication in the prevention of recurrence of episodes can be substantially enhanced with the addition of structured family therapy (Miklowitz, Simoneau, Sachs-Ericsson, Warner, & Suddath, 1996; I. W. Miller, Keitner, Epstein, Bishop, & Ryan, 1991). Much the same can be said for family therapy with schizophrenia, on the basis of the pioneering work by Falloon, Goldstein, Hogarty. and others (e.g., Falloon, Boyd, & McGill, 1985; Hogarty, McEvoy, & Munetz, 1988), with some evidence now indicating that family therapy is more successful than drugs alone at preventing relapse for periods of up to eight years following treatment (Tarrier, Barrowclough, Porceddu, & Fitzpatrick, 1994).

Health Care Policy, Psychotherapy Research, and the Future of Psychotherapy 397

Complex psychological treatment is the only strategy with any evidence for effectiveness for borderline personality disorder (Linehan, Armstrong, Suares, Allmon, & Heard, 1991), and follow-up studies indicate that these gains are maintained (Linehan, Heard, & Armstrong, 1993). Furthermore, as one might expect, for a chronic personality disorder, this treatment is long-term psychotherapy lasting at least one year, probably longer in the typical case. However, if further trials replicate this result, no one in a position of creating mental health policy could argue with either the treatment effectiveness or the cost-effectiveness of this long-term therapy, as compared with treatment as usual, in terms of its effect on the utilization of health care services. It is also known that psychotherapeutic procedures with proven efficacy exist for physical disorders such as cardiovascular disease and cancer (e.g., Fawzy, Fawzy, Arndt, & Pasnau, 1995; Frasure-Smith, 1991; Spiegel, Bloom, Kraemer, & Gottheil, 1989) and that these procedures can extend and even save lives. Incontrovertible evidence now exists that psychological interventions change brain function and, very likely, brain structure as at least part of their mechanism of action (e.g., Barlow & Durand, 1995; Baxter et al., 1992; Ironson et al., 1992). Where analyses have been done, as in the case of eating disorders, psychological interventions are highly cost-effective. For example, Mitchell, Peterson, and Agras (1999) determined that up to 22 sessions of psychosocial treatment for eating disorders could be delivered at a cost of only 50%–80% of the total cost of 13 sessions of pharmacotherapy, assuming that the patient stayed on a serotonin-specific reuptake inhibitor (fluoxetine, the currently preferred pharmacological treatment for bulimia) for one year. Additional analyses of cost-effectiveness, which involve complex determinations of effects on functioning as well as cost, are also likely to be favorable for psychosocial treatments (i.e., N. E. Miller & Magruder, in press). Firm evidence also exists that psychotherapists can successfully treat a number of additional problems, such as insomnia (Murtagh & Greenwood, 1995) and a variety of childhood disorders (e.g., Houts, Berman, & Abramson, 1994; Wells & Egan, 1988). This is only a brief and partial overview of this topic. For more complete summaries, see Barlow (1994), Nathan and Gorman (in press), Roth and Fonagy (1995), and Seligman (1994). It is also noteworthy (and provides excellent public relations) that Consumer Reports (1995) recently recognized and reported on some of these specific interventions, such as cognitive therapy for depression. Of course, with all disorders for which there are “effective” treatments, both drug and psychological, substantial numbers of patients fail to improve or improve only partially, as Wilson (1996b) recently articulated in the case of eating disorders. Continued aggressive efforts at treatment development and evaluation are necessary, and there is no reason for complacency. Nevertheless, results reviewed above would seem reason for optimism among practicing psychotherapists. Treatments with proven efficacy exist for a large number of psychological disorders, and these interventions are finally making their way into consensus reports on effective treatment and into clinical practice guidelines (e.g., Depression Guideline Panel, 1993; Wolfe & Maser, 1994). Yet, the beginnings of these efforts, combined with the appearance of

398 David H. Barlow

preliminary lists of psychotherapeutic procedures meeting the types of evidentiary standards used by governments around the world in constructing clinical practice guidelines, have aroused objections in some circles (e.g., Chambless et al., 1996; Roth & Fonagy, 1995; Task Force on Promotion and Dissemination of Psychological Procedures, 1995). Objections to Empirically Supported Treatments and the Role of Psychotherapy Research Objections to identifying empirically supported psychosocial interventions with an evidentiary base equivalent to pharmacological or other treatments seem to cluster into two general categories. The first set of objections alleges that the scientific method, as it is known, is incapable of proving the effectiveness or ineffectiveness of psychotherapy and that some alternative approach to psychotherapy research is preferable. This point of view usually revolves around the adequacy or inadequacy of particular philosophies of science that might underlie psychotherapy research, or other implicit flaws. Indeed, interesting alternative philosophical approaches have been proposed that might ultimately supplement the knowledge of human behavior in context (e.g., Biglan & Hayes, in press; Leigland, 1995; Polkinghorne, 1983). But arguments about philosophy, although popular in academic circles, are unlikely to affect Congress or health care policymakers at every level of government who long ago made the decision that health care professionals must be accountable and demonstrate that what they do works. Furthermore, whether it is NIH, the National Institute of Occupational Safety and Health, the National Institute of Education, or the Food and Drug Administration, these rules of evidence have been well worked out and rely on empirical demonstrations of relief of dysfunction or enhancement of functioning. It is unlikely that they would make an exception for psychotherapy. Although methods for evaluating psychotherapy are flawed and subject to considerable improvement (see below), antiscience arguments have not been, and will not be, influential. A second set of objections centers on inadequacies in the scientific method that leave important questions concerning the effects of psychotherapy unanswered (e.g., Barlow, Hayes, & Nelson, 1984; Garfield, in press; Hayes, Barlow, & Nelson-Grey, in press). From this viewpoint, the beginning of research data supporting the efficacy of psychotherapy is generally welcomed, but some conclude that it might be premature to designate specific psychosocial interventions as effective until more is known about the generality of effectiveness, the feasibility of applying the intervention, and the mechanisms of action of the treatments under consideration. Also at issue is the generalizability of highly structured, manualized psychological interventions used in clinical trials to practice settings, where psychotherapy is often applied more flexibly and adapted to meet the perceived needs of the patient, who may present with multiple problems. These are very serious objections that cannot be ignored. In fact, most clinical practice guidelines, even those as sophisticated as the AHCPR’s guideline on depression (Depression Guideline Panel, 1993), focus very narrowly on randomized clinical trials conducted in clinical research centers comparing treatment to no treatment or some good alternative such as

Health Care Policy, Psychotherapy Research, and the Future of Psychotherapy 399

placebo. Results are often reported for the average patient seen in one type of setting, with insufficient description of individual variation. Little attention is paid to the type of research that would examine the applicability of these treatments, whether drug or psychosocial, to the clinical settings in which they are usually applied, an area of research that falls under the broad heading of services research. Recognizing the substantial influence of emerging clinical practice guidelines as well as their inevitability, the American Psychological Association appointed a task force to create a document against which to measure the adequacy of these emerging guidelines and to move the field beyond a narrow reliance on tightly controlled, randomized clinical trials. This task force was jointly sponsored by the Committee for the Advancement of Professional Practice, the Board of Professional Affairs, and the Board of Scientific Affairs of the American Psychological Association. The final document was entitled Template for Developing Guidelines: Interventions for Mental Disorders and Psychosocial Aspects of Physical Disorders (American Psychological Association, Task Force on Psychological Intervention Guidelines, 1995). The purpose of this template, which was approved as a policy of the American Psychological Association in 1995, is to articulate the types of research studies necessary to designate any particular therapeutic intervention as sufficiently effective in treating a given disorder to be included in a clinical practice guideline. Thus, this document does not in itself advocate guidelines, but, rather, recognizing their inevitability, serves as a means of evaluating existing guidelines and, in many instances, illustrates their inadequacy. The template was modeled after the highly respected Standards for Educational and Psychological Testing (American Psychological Association, 1985), which set out the basic scientific principles for constructing an adequate psychological test. The template requires that clinical practice guidelines be evaluated on the basis of two simultaneous considerations, or axes. The first axis requires that committees or panels constructing guidelines conduct a rigorous assessment of the scientific evidence, with the goal of measuring the efficacy of any given intervention. The term efficacy usually refers to the results of a systematic evaluation of the intervention in a controlled clinical research context. Considerations relevant to the internal validity of these conclusions are usually highlighted. As noted above, it is this type of evidence that has formed the basis of most well-constructed clinical practice guidelines to date. The second axis specifies that the guidelines consider the applicability and feasibility of the intervention in the local setting where the intervention is delivered, as well as determine the generalizability of an intervention with established efficacy. The term effectiveness has sometimes been used to describe the second axis, but the term clinical utility was chosen by the American Psychological Association, Task Force on Psychological Intervention Guidelines (1995). An overview of the types of evidence designated as important by the template is provided in Table 18.1. Research addressing these concerns, as mentioned above, is usually referred to as services research, and it is this second axis where data are usually lacking. Because of this gap, some clinical scientists have concluded that it may be premature to identify and disseminate information on effective psychological treatments.

Internal Validity (Efficacy) Clinical Utility (External Validity) 1. Better than alternative therapy (randomized controlled trials . . . 1. Feasibility RCT’s) A) Patient acceptability (cost, pain, duration, side effects, etc.) B) Patient choice in face of relatively equal efficacy 2. Better than non-specific therapy (RCT’s) C) Probability of compliance 3. Better than no therapy (RCT’s) D) Ease of disseminability—e.g., number of practitioners with 4. Quantified clinical observations competence, requirements for training, opportunities for 5. Strongly positive clinical consensus training, need for costly technologies or additional support 6. Mixed clinical consensus personnel, etc. 7. Strongly negative clinical consensus 2. Generalizability 8. Contradictory evidence A) Patient characteristics Note: Confidence in treatment efficacy is based on both: a) the 1) Cultural background issues absolute and relative efficacy of the treatment and 2) Gender issues b) the quality of the studies on which the judgment is made, as 3) Development issues well as their replicability. 4) Other relevant patient characteristics B) Therapist characteristics C) Issues of robustness when applied in practice settings with different time frames, etc. D) Contextual factors regarding setting in which treatment is delivered 3. Costs and Benefits A) Costs of Delivering intervention to individual and society B) Costs of withholding intervention to individual and society. NOTE: Confidence in clinical utility as reflected on these 3 dimensions should be based on systematic and objective methods and strategies for assessing these characteristics of treatment as they are applied in actual practice. In some cases, randomized controlled trials will exist. More often, data will be in the form of quantified clinical observations [clinical replication series] or other strategies such as health economic.

Note. From “Template for Developing Guidelines: Interventions for Mental Disorders and Psychosocial Aspects of Physical Disorders,” by American Psychological Association Board of Professional Affairs Task Force on Psychological Intervention Guidelines, 1995. Approved by the APA Council of Representatives, February 1995, Washington D.C. Copyright © 1996 by the American Psychological Association.

Increasing confidence in efficiency

Table 18.1 Overview of Template for Constructing Psychological intervention Guidelines

Health Care Policy, Psychotherapy Research, and the Future of Psychotherapy 401

The template also notes that current guidelines are disorder-based, because all professional and government agencies involved in these issues now look at evidence in terms of specific psychological and drug treatments for identified problems or disorders. In other words, when talking about efficacy or clinical utility, it is important to know whether one is talking about adjustment disorders, schizophrenia, panic disorder, or autism. Treatment by any clinician would necessarily differ from disorder to disorder or from problem to problem. Therefore, the question “Do drugs work?” or “Does psychotherapy work?” without further specification of the type of drug or psychological intervention, and the disorder that provides the context for the intervention, is simply not relevant to those constructing clinical practice guidelines, as noted by the American Psychological Association, Task Force on Psychological Intervention Guidelines (1995). Of course, services research examining broad patterns of service delivery, where specific questions such as “How long does the average patient usually stay in psychotherapy?” are addressed, have a very different and useful goal. But, these data will have to be disaggregated at some point to be useful in clinical practice guidelines, because the answers will be very different if the patient is undergoing an adjustment to life circumstances or suffering from bipolar disorder. The inability to disaggregate the results is another limiting factor in the retrospective survey of patients receiving psychotherapy published by Consumer Reports (1995), although, presumably, many of the doctoral-level mental health professionals whose clients were surveyed were using effective drug or psychosocial treatments, as was evident in Schulberg et al.’s (1995) study. In any case, although this report provides good public relations for psychotherapy, as noted above, it is of limited use to health care policymakers since it was based on retrospective self-report in a highly self-selected sample. As Seligman (1995) pointed out, prospective studies of the applicability of psychotherapy in clinical settings with appropriate comparison groups will almost always be required to produce credible results on the clinical utility of certain psychosocial approaches, and the Consumer Reports study pointed to some interesting questions that should be asked. Emerging Research on Clinical Utility In any case, a number of specific issues require substantial additional attention in the context of research on clinical utility. These include client variables, therapist variables, the therapeutic process and mechanisms of action, and the relative utility of the increasing use of structured or manualized psychosocial treatments to guide clinical practice, to name just a few. These issues will continue to be explored and debated as clinicians strive for a more satisfactory and applicable clinical science (e.g., Barlow, in press; Garfield, in press; Hayes, 1995; Wilson, in press), but it is sufficient for the current purposes to simply outline some of the major issues. It is very likely that client variables, including personality factors and comorbidity, interact in a very strong way with the effects of psychological (and pharmacological) interventions. Data are just beginning to appear on this topic (e.g., Beutler & Clarkin, 1990; Chambless et al., 1996; Garfield, 1994). But it would be inaccurate to assume that these client variables necessarily interfere

402 David H. Barlow

with the effects of specific interventions. For example, in the research clinic of myself and my colleagues, the impact of comorbidity on outcome was examined in a large group of patients with panic disorder as a principal diagnosis; they were treated with a cognitive-behavioral procedure for panic disorder developed in our setting (Barlow & Craske, 1994; Barlow, Craske, Cerny, & Klosko, 1989). Much to our surprise, pretreatment comorbidity (or the lack of it) had no effect on the outcome or treatment for panic disorder three months and two years following treatment. It is also widely assumed that clients included in efficacy studies differ systematically from those who are excluded for one reason or another. However, in one study specifically examining this issue in clients with social phobia, Juster, Heimberg, and Engelberg (1995) found that clients excluded from studies for various reasons, and those who refused to enter the study, differed systematically at pretreatment from those included on a variety of variables, including socioeconomic status and social support. Despite these differences, both those excluded and those included in the clinical trials evidenced comparable clinical gain from treatment. It is unlikely that these effects will hold across the full range of psychological disorders and for all patterns of client variables, but research is proceeding in an area that is becoming known as aptitude-treatment interaction. For example, Beutler et al. (1991) began to examine a client variable termed reactance (resistance) that may have important implications for the style in which treatment is administered. Blatt (1995) provided some interesting preliminary data on the implications of the trait of perfectionism for treatment outcome in depression. But data on generalizability to many other client variables, such as minority populations, are sorely lacking. Hopefully, research on client variables will expand considerably. Recent advances in profiling individual responses to treatment on a wide range of outcomes should prove particularly useful in this regard (e.g., Howard, Moras, Brill, Martinovich, & Lutz, 1996). Most people would assume that therapist variables. including competence and experience, would be positively related to outcome. But this assumption has been very difficult to demonstrate. In fact, one clinical utility study from a managed care setting found that experience of the therapist was inversely related to outcome (Hiatt & Hargrave, 1995)! However, because neither the disorder nor the mode of treatment was specified or disaggregated in any meaningful way, it is difficult to draw conclusions. For this reason, some have proposed that doctoral-level practitioners may no longer be needed in large numbers in the absence of differential treatment effectiveness because it is more cost-effective to use lesser trained professionals (e.g., Christenson & Jacobson, 1994). However, other data are beginning to appear that may tell a different story. For example, Frank et al. (1991) assessed the relationship between quality of interpersonal therapy, as determined by objective ratings, and the prevention of recurrence of depressive episodes in clients with recurrent unipolar depression. Those clients who received high-quality interpersonal therapy delivered in a skilled manner were far more likely to stay well than was a comparable group receiving lower quality therapy. Interestingly, Frank and Spanier (1995) noted that staying focused on the interpersonal

Health Care Policy, Psychotherapy Research, and the Future of Psychotherapy 403

therapy protocol seemed to best reflect competence in this study. In any case, patients receiving this higher quality therapy only once a month were nearly as successful as patients continuing on full-dose antidepressant medication in preventing recurrence of depressive episodes for three years (46% for those taking medication vs. 39% for those receiving interpersonal therapy; Frank et al., 1990, 1991). Patients receiving lower quality therapy were no different than patients receiving a placebo in terms of their recurrence of depressive episodes (although this post hoc analysis needs prospective replication). Similarly, in a study mentioned above, Schulberg et al. (1995) found doctoral-level mental health professionals were far more effective than primary care physicians in treating major depressive disorder. It is likely that additional studies will underscore the importance of therapist competence. Regarding more structured or manualized therapies, most assume, myself included, that adapting these treatments in a flexible manner to the client and setting will provide better outcomes than applying the intervention rather rigidly, as is often the case in efficacy studies. In fact, Jacobson et al. (1989) found no differences in outcome in clients following marital therapy delivered in a highly structured manner versus more flexibly, although the flexible delivery showed a nonsignificant trend to be slightly better at a six-month follow-up. In another study that specifically tested this hypothesis with 120 phobic patients, the results were, counterintuitively, quite the opposite in that standardized therapy proved to be significantly more successful than flexible therapy (Schulte, Kunzel, Pepping, & Schulte-Bahrenberg, 1992). Wilson (1996a) outlined an important research agenda examining the use and the misuse of these structured programs in clinical practice. Conclusions Only the first plateau has been reached in the development of effective psychological interventions for a variety of problems. There now are a number of studies with high internal validity demonstrating the efficacy of psychotherapeutic procedures as compared with some credible alternatives for a variety of problems in the average patient. In so doing, these interventions have been demonstrated to be significantly superior to credible psychosocial control conditions in the majority of cases, thereby ruling out nonspecific or common factors as the principal mechanism of action (Barlow, 1994; Nathan & Gorman, in press; Roth & Fonagy, 1995; Seligman, 1994). Of course, common factors will always enhance the effects of psychosocial or drug treatments, but it no longer seems reasonable to assume they are the only active ingredient or even the major ingredient. Furthermore, psychotherapy research has produced a distinct trend away from schools of psychotherapy toward more eclectic approaches drawing on procedures from a variety of psychotherapeutic orientations (Barlow, 1995). Thus, in this day and age, psychosocial procedures are seldom based on just theory, or on the charisma of the founder of the school of psychotherapy, but rather on evidence, often emanating from clinical practice, that certain specific strategies or techniques are effective. Thus, cognitive-behavioral therapy has discovered the “unconscious” (Kihlstrom, Barnhardt, & Tataryn, 1992), and most psychotherapists have recognized the

404 David H. Barlow

wisdom of engaging the interpersonal system of the patient that provides the context in which the presenting problem occurs. This empirically derived psychotherapy integration is likely to prove more fruitful than a theoretically derived integration. Various agencies are continuing to document these scientific developments. Division 12 of the American Psychological Association has appointed a standing committee (the Task Force on Psychological Interventions) to revise and update its report on empirically supported treatment as well as the methodology for developing this information, and the first revision is now complete (Chambless et al., 1996). At the same time, this task force cautions against using this report in any way as some kind of a standard for determining reimbursable services. A separate Division 12 task force constituted by Past-President Gerald P. Koocher and chaired by Suzanne Bennett-Johnson is focusing on empirically supported interventions and preventive programs for children. As noted above, the National Health Service in the United Kingdom has commissioned a similar but far more detailed report to guide policy on psychological interventions in the United Kingdom (Roth & Fonagy, 1995). Data on clinical utility are sorely lacking, although studies are beginning to appear on issues such as generality across clients, therapists, and settings, as well as feasibility and cost-effectiveness. This process should receive a considerable boost from the creation of practice research networks, such as those recently constructed by the American Psychiatric Association as well as those under consideration by the American Psychological Association. In the case of the American Psychiatric Association, a thousand clinicians have agreed to have their practices closely monitored in terms of clients seen, interventions, and outcomes so that the applicability of new drugs or psychosocial treatments can be evaluated in the settings in which they are actually offered. This network is called the Association’s Clinical Outcomes Research Network (American Psychiatric Association, Guidelines Steering Committee and the Office of Research, 1993; McIntyre, 1994). These data, in turn, can be related to local issues, and successful innovations utilized by some clinicians in adapting assessment or intervention procedures to their local setting can be communicated widely to others or evaluated more fully in clinical research centers. Thus, practitioners acting as local scientists and using, perhaps, individual patient profiling technology as articulated by Howard et al. (1996), as well as the most powerful psychological treatments, may take the lead in guiding clinical researchers in the production of important data on clinical utility. This would fulfill a goal articulated previously (e.g., Barlow et al., 1984) and, more recently, elegantly elaborated by Stricker and Trierweiler (1995). In this way, practitioners and researchers will finally sit down together to reach the common goal of providing the most effective service to clients. Finally, some would say it is too early to conclude that psychotherapy works and that there are effective psychological interventions because clinicians do not yet know enough about the clinical utility of their procedures. But, from a public policy point of view, it is important to recognize that managed care companies, government policymakers, and others are writing and publishing clinical practice guidelines, and these guidelines are assuming the force of law in many states, as noted above. Colleagues in medicine have not been

Health Care Policy, Psychotherapy Research, and the Future of Psychotherapy 405

as reticent to provide data on effective drug treatments, even though studies evaluating drugs are, if anything, more narrowly restricted to randomized, controlled clinical trials or efficacy studies than are those evaluating psychotherapy. For example, few drug studies last beyond a four- to eight-week time frame, and even fewer examine generality across clients, therapists, and settings in a systematic way or examine other factors relevant to clinical utility. Yet, these drugs receive approval by the Food and Drug Administration and are widely disseminated. In my view, it is time to tell the world that, although there is still much to learn, there now are psychosocial treatments with proven efficacy that, in the hands of skilled mental health professionals, can relieve human suffering and enhance human functioning. References American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington D.C.: Author. American Psychiatric Association. (1995). Practice guideline for the treatment of patients with substance use disorders: Alcohol, cocaine, opioids. American Journal of Psychiatry, 152(Suppl. 11), 5–59. American Psychiatric Association, Guidelines Steering Committee and the Office of Research. (1993). Development of a psychiatric practice research network: Overview of the purpose and a proposed approach. Washington D.C.: Author, Office of Research. American Psychological Association. (1985). Standards for educational and psychological testing. Washington D.C.: Author. American Psychological Association, Task Force on Psychological Intervention Guidelines. (1995). Template for developing guidelines: Intervention for mental disorders and psychosocial aspects of physical disorders. Washington D.C.: Author. Barlow, D. H. (Ed.). (1993). Clinical handbook of psychological disorders (2nd ed.). New York: Guilford Press. Barlow, D. H. (1994). Psychological interventions in the era of managed competition. Clinical Psychiatry: Science and Practice, 1, 109–122. Barlow, D. H. (in press). The effectiveness of psychotherapy: Science and policy. Clinical Psychology: Science and Practice. Barlow, D. H., & Craske, M. G. (1994). Mastery of your anxiety and panic, II. Albany, NY: Graywind. Barlow, D. H., Craske, M. G., Cerny, J. A., & Klosko, J. S. (1989). Behavioral treatment of panic disorder. Behavior Therapy, 20, 261–262. Barlow, D. H., & Durand, V. M. (1995). Abnormal psychology: An integrative approach. Pacific Grove, CA: Brooks/Cole. Barlow, D. H., Hayes, S. C., & Nelson, R. O. (1984). The scientist-practitioner: Research and accountability in clinical and educational settings. New York: Pergamon Press. Barlow, D. H., & Lehman. C. (1996). Advances in the psychosocial treatment of anxiety disorders: Implications for national health care. Archives of General Psychiatry, 53, 727–735. Baxter, L. R., Jr., Schwartz, J. M., Bergman, K. S., Szuba, M. P., Guze, B. H., Mazziotta, J. C., Alazraki, A., Selin, C. E., Ferng, H. K., Munford. P., & Phelps, M. E. (1992). Caudate glucose metabolic rate changes with both drug and behavior therapy for obsessive-compulsive disorder. Archives of General Psychiatry, 49, 681–689. Beutler, L. E., & Clarkin, J. F. (1990). Systematic treatment selection: Toward targeted therapeutic interventions. New York: Brunner/Mazel.

406 David H. Barlow Beutler, L. E., Engle, D., Mohr, D., Daldrup, R. J., Bergan, J., Meredith, K., & Merry, W. (1991). Predictors of differential response to cognitive, experiential, and self-directed psychotherapeutic procedures. Journal of Consulting and Clinical Psychology, 59, 333–340. Biglan, A., & Hayes, S. C. (in press). Should the behavioral sciences become more pragmatic? The case for functional contextualism in search on human behavior. Applied and Preventive Psychology: Current Scientific Perspectives. Blatt, S. J. (1995). The destructiveness of perfectionism: Implications for the treatment of depression. American Psychologist, 50, 1003–1020. Chambless, D. L., Sanderson, W. C., Shoham, V., Johnson, S. B., Pope, K. S., Crits-Christoph, P., Baker, M., Johnson, B., Woody, S. R., Sue, S., Butler, L., Williams, D. A., & McCurry, S. (1996). An update on empirically validated therapies. The Clinical Psychologist, 49, 5–18. Christensen, A., & Jacobson, N. S. (1994). Who (or what) can do psychotherapy: The status and challenge of nonprofessional therapies. Psychological Science, 5, 8–14. Consumer Reports. (1995, November). Mental health: Does therapy help? 734–739. Depression Guideline Panel. (1993). Depression in primary care: Vol. 2. Treatment of major depression (Clinical practice guideline. No. 5; AHCPR Publication No. 93–0551). Rockville, MD: U.S. Department of Health and Human Services, Public Health Service, Agency for Health Care Policy and Research. Fairburn, C. G., Jones, R., Peveler, R. C., Hope, R. A., & O’Connor, M. (1993). Psychotherapy and bulimia nervosa: Longer term effects of interpersonal psychotherapy, behavior therapy, and cognitive behavior therapy. Archives of General Psychiatry, 50, 419–428. Falloon, I. R. H., Boyd, J. L., & McGill, C. W. (1985). Family management in the precaution of morbidity of schizophrenia: Clinical outcome of a two-year longitudinal study. Archives of General Psychiatry, 42, 882–896. Fawzy, F. I., Fawzy, N. W., Arndt, L., A., & Pasnau, R. O. (1995). Critical review of psychosocial interventions in cancer care. Archives of General Psychiatry, 52, 100–113. Frank, E., Kupfer, D. J., Perel, J. M., Cornes, C., Jarrett, D. B., Mallinger, A. G., Thase, M. E., McEachran, A. B., & Grochocinski, V. J. (1990). Three-year outcomes for maintenance therapies in recurrent depression. Archives of General Psychiatry, 47, 1093–1099. Frank, E., Kupfer, D. J., Wagner, E. F., McEachran, A. B., & Cornes, C. (1991). Efficacy of interpersonal psychotherapy as a maintenance treatment of recurrent depression. Archives of General Psychiatry, 48, 1053–1059. Frank, E., & Spanier, C. (1995). Interpersonal psychotherapy for depression: Overview, clinical efficacy, and future directions. Clinical Psychology: Science and Practice, 2, 349–369. Frasure-Smith, N. (1991). In-hospital symptoms of psychological stress as predictors of long-term outcome after acute myocardial infarction in men. American Journal of Cardiology, 67, 1221–1227. Garfield, S. L. (1994). Research on client variables in psychotherapy. In A. E. Bergin & S. L. Garfield (Eds.), Handbook of psychotherapy and behavior change (4th ed., pp. 190–228). New York: Wiley. Garfield, S. L. (in press). Some problems associated with “validated” forms of psychotherapy. Clinical Psychology: Science and Practice. Hayes, S. C. (1995). What do we want from scientific standards of psychological practice. In S. C. Hayes, V. M. Follette, R. M. Dawes, & K. E. Grady (Eds.), Scientific standards of psychological practice: Issues and recommendations (pp. 49–66). Reno, NV: Context Press. Hayes, S. C., Barlow, D. H., & Nelson-Grey, R. O. (in press). The scientist-practitioner: Research and accountability in the era of managed care (2nd ed.). Needham Heights, MA: Allyn & Bacon. Health Maintenance Organization Act of 1973, Pub. L. No. 93–222, 87 Stat. 914, Title 42 (1973, December 29).

Health Care Policy, Psychotherapy Research, and the Future of Psychotherapy 407 Hiatt, D., & Hargrave, G. E. (1995). The characteristics of highly effective therapists in managed behavioral provider networks. Behavioral Healthcare Tomorrow, 4, 19–22. Hogarty, G. E., McEvoy, J. P., & Munetz., M. (1988). Dose of fluphenazine, familial expressed emotion, and outcome in schizophrenia: Results of a two-year controlled study. Archives of General Psychiatry, 45, 797–805. Houts, A. C., Berman, J. S., & Abramson, H. (1994). Effectiveness of psychological and pharmacological treatments for nocturnal enuresis. Journal of Consulting and Clinical Psychology, 62, 737–745. Howard, K. I., Moras, K., Brill, P. L., Martinovich, Z., & Lutz, W. (1996). Evaluation of psychotherapy: Efficacy, effectiveness, and patient progress. American Psychologist, 51, 1059–1064. Ironson, G., Taylor, C. B., Boltwood, M., Bartzokis, T., Dennis, C., Chesney, M., Spitzer, S., & Segall, G. M. (1992). Effects of anger on left ventricular ejection fraction in coronary artery disease. American Journal of Cardiology, 70, 281–285. Jacobson, N. S., & Hollon, S. D. (1996). Cognitive-behavior therapy versus pharmacotherapy: Now that the jury’s returned its verdict, it’s time to present the rest of the evidence. Journal of Consulting and Clinical Psychology, 64, 74–80. Jacobson, N. S., Schmaling, K. B., Holtzworth-Munroe, A., Katt, J. L., Wood, L. F., & Follette, V. M. (1989). Research structured vs. clinically flexible versions of social learning-based marital therapy. Behaviour Research and Therapy, 27, 173–180. Juster, H. R., Heimberg, R. G., & Engelberg, B. (1995). Self selection and sample selection in a treatment study of social phobia. Behaviour Research and Therapy, 33, 321–324. Kihlstrom, J. F., Bernhardt, T. M., & Tataryn, D. J. (1992). The psychological unconscious: Found, lost, and regained. American Psychologist, 47, 788–791. Leigland, S. (1995). Discussion of Hayes “On the relation between clinical practice and psychological science,” In S. C. Hayes, V. M. Follette, R. M. Dawes, & K. E. Grady (Eds.), Scientific standards of psychological practice: Issues and recommendations (pp. 67–71). Reno, NV: Context Press. Linehan, M. M., Armstrong, H. E., Suarez, A., Allmon, D., & Heard, H. L. (1991). Cognitive–behavioral treatment of chronically parasuicidal borderline patients. Archives of General Psychiatry, 48, 1060–1064. Linehan, M. M., Heard, H. L., & Armstrong, H. E. (1993). Naturalistic follow-up of a behavioral treatment for chronically parasuicidal borderline patients. Archives of General Psychiatry, 50, 971–974. McIntyre, J. (1994, March 18). Practice research network: Pilot stage. Psychiatric News. Miklowitz, D. J., Simoneau, T. L., Sachs-Ericsson, N., Warner, R., & Suddath, R. (1996). Family risk indicators in the course of bipolar affective disorder. In C. Mundt, M. J. Goldstein, K. Hahlweg, & P. Fiedler (Eds.), Interpersonal factors in the origin and course of affective disorders (pp. 204–215). London: Gaskell Books. Miller, I. W., Keitner, G. I., Epstein, N. B., Bishop, D. S., & Ryan, C. E. (1991, November). Families of bipolar patients: Dysfunction, course of illness, and pilot treatment study. Paper presented at the annual meeting of the Association for the Advancement of Behavior Therapy, New York. Miller, N. E., & Magruder, K. M. (Eds.), (in press). The cost effectiveness of psychotherapy: A guide for practitioners, researchers and policymakers. New York: Wiley. Mitchell, J. E., Peterson, C. B., & Agras, S. (in press). Cost-effectiveness of psychotherapy for eating disorders. In N. E. Miller & K. M. Magruder (Eds.), The cost effectiveness of psychotherapy: A guide for practitioners, researchers and policymakers. New York: Wiley. Muñoz, R. F., Hollon, S. D., McGrath, E., Rehm, L. P., & VandenBos, G. R. (1994). On the AHCPR Depression in Primary Care guidelines: Further considerations for practitioners. American Psychologist, 49, 42–61.

408 David H. Barlow Murtagh, D. R. R., & Greenwood, K. M. (1995). Identifying effective psychological treatments for insomnia: A meta-analysis. Journal of Consulting and Clinical Psychology, 63, 79–89. Nathan, P. E., & Gorman, J. M. (Eds.), (in press). Psychotherapies and drugs that work. A review of the outcome studies. New York: Oxford University Press. Polkinghorne, D. (1983). Methodology for the human sciences: Systems of inquiry. Albany: State University of New York Press. The Quality Assurance Project. (1983). A treatment outline for depressive disorders. Australian and New Zealand Journal of Psychiatry, 17, 129–148. The Quality Assurance Project. (1984). Treatment outlines for the management of schizophrenia. Australian and New Zealand Journal of Psychiatry, 18, 19–38. The Quality Assurance Project. (1985). Treatment outlines for the management of anxiety states. Australian and New Zealand Journal of Psychiatry, 19, 138–151. Roth, A., & Fonagy, P. (1995, February). Research on the efficacy and effectiveness of the psychotherapies (National Health Service Report). Report to the Department of Health. London: National Health Services. Schulberg, H. C., Block, M., & Madonia, M. (1995, August). Treating major depression in primary care practice: Eight-month clinical outcomes. Paper presented at the 103rd Annual Convention of the American Psychological Association, New York. Schulte, D., Kunzel, R., Pepping, G., & Schulte-Bahrenberg, T. (1992). Tailor-made versus standardized therapy of phobic patients. Advances in Behaviour Research and Therapy, 14. 67–92. Seligman, M. E. P. (1994). What you can change and what you can’t. New York: Knopf. Seligman, M. E. P. (1995). The effectiveness of psychotherapy: The Consumer Reports study. American Psychologist, 50, 965–974. Spiegel, D., Bloom, J. R., Kraemer, H. C., & Gottheil, E. (1989). Effect of psychosocial treatment on survival of patients with metastatic breast cancer. Lancet, 14, 888–891. Stricker, G., & Trierweiler, S. J. (1995). The local clinical scientist: A bridge between science and practice. American Psychologist, 50, 995–1002. Tarrier, N., Barrowclough, C., Porceddu, K., & Fitzpatrick, E. (1994). The Salford Family Intervention Project: Relapse rates of schizophrenia at five and eight years. British Journal of Psychiatry, 165, 829–832. Task Force on Promotion and Dissemination of Psychological Procedures. (1995). Training in and dissemination of empirically-validated psychological treatments. The Clinical Psychologist, 48, 3–23. VandenBos, G. R. (1993). U.S. mental health policy: Proactive evolution in the midst of health care reform. American Psychologist, 48, 283–290. Wells, K. C., & Egan, J. (1988). Social learning and systems family therapy for childhood oppositional disorder: Comparative treatment outcome. Comprehensive Psychiatry, 29, 138–146. Wilson, G. T (1996a). Manual-based treatments: The clinical application of research findings. Behaviour Research and Therapy, 34, 295–314. Wilson, G. T. (1996b). Treatment of bulimia nervosa: When CBT fails. Behaviour Research and Therapy, 34, 197–212. Wilson, G. T. (in press). Empirically-validated treatments: Reality and resistance. Clinical Psychology: Science and Practice. Wolfe, B. E., & Maser, J. D. (1994). Treatment of panic disorder. A consensus development conference. Washington D.C.: American Psychiatric Press. Zarin, D. A., Pincus, H. A., & McIntyre, J. S. (1993). Practice guidelines. American Journal of Psychiatry, 150, 175–177.

Article 19

The Dissemination and Implementation of Evidence-Based Psychological Treatments: A Review of Current Efforts R. Kathryn McHugh and David H. Barlow Boston University

Recognizing an urgent need for increased access to evidenced-based psychological treatments, public health authorities have recently allocated over $2 billion to better disseminate these interventions. In response, implementation of these programs has begun, some of it on a very large scale, with substantial implications for the science and profession of psychology. But methods to transport treatments to service delivery settings have developed independently without strong evidence for, or even a consensus on, best practices for accomplishing this task or for measuring successful outcomes of training. This article reviews current leading efforts at the national, state, and individual treatment developer levels to integrate evidence-based interventions into service delivery settings. Programs are reviewed in the context of the accumulated wisdom of dissemination and implementation science and of methods for assessment of outcomes for training efforts. Recommendations for future implementation strategies will derive from evaluating outcomes of training procedures and developing a consensus on necessary training elements to be used in these efforts. In recent years, health care policy has incorporated evidence-based practice as a central tenet of health care delivery (Institute of Medicine, 2001). Despite the promise to raise standards of care, evidence-based practice has encountered barriers common to all knowledge diffusion efforts (Rogers, 2003), and the dissemination of empirically supported medical and psychological interventions has been slow. A report by the Institute of Medicine (2001) argued that the remarkable disconnect between medical research and practice represents “not just a gap, but a chasm” (p. 1). Moreover, reports by the U.S. Surgeon General (U.S. Public Health Service, 1999) and the President’s New Freedom Commission on Mental Health (2004) have specifically

410 R. Kathryn McHugh and David H. Barlow

highlighted a lack of access to evidence-based mental health care. In the context of the development of evidence-based psychological treatments (EBPTs; Barlow, 2008; Kazdin & Weisz, 2003; Nathan & Gorman, 2007) with the potential to improve clinical outcomes across mental health and substance abuse treatment settings, research has confirmed low levels of successful dissemination in both clinical practice settings (e.g., Goisman, Warshaw, & Keller, 1999; Stewart & Chambless, 2007) and graduate and internship training programs (e.g., Crits-Christoph, Frank, Chambless, Brody, & Karp, 1995; Weissman et al., 2006). Motivated by the continued lack of widespread availability of EBPTs, both public and private funding mechanisms for dissemination and implementation efforts have emerged. Government agencies with a marked sense of urgency have created financial and regulatory incentives and mandates promoting a shift to evidence-based practice and are driving these efforts with large financial commitments totaling several billion dollars. Support for these efforts continues to build. As a consequence, a number of new programs have emerged rather quickly, and some have begun programmatic activity. The programs under way are heterogeneous in their structure, aims, and scope and are in relatively early stages of implementation and evaluation. Furthermore, many of these programs have developed independently, and some funding initiatives carry an urgent mandate precluding deliberate (but time-consuming) consultation across programs on best practices. Thus, at this critical juncture, an evidence base for the dissemination and implementation of EBPTs is lacking, and no clear consensus has emerged on best practices for these initiatives. Our aim in the current article is to review leading dissemination and implementation programs at the national, state, and individual treatment developer levels in order to examine strategies used by these programs, commonalities among them, and the extent to which they target the usual and customary barriers to adoption of emerging knowledge. We do not revisit controversies surrounding the identification or appropriateness of EBPTs (Hofmann & Weinberger, 2007); rather, we focus on the status and adequacy of efforts currently under way that have already attracted billions of dollars in funding, with more to follow. We begin with a brief discussion of dissemination and implementation science and a two-part conceptualization of the necessary components of clinician training; we then review programs in this context and on the methods they use for assessing outcomes of training. We provide a checklist of the procedures currently being used by these programs for training and for assessing outcomes. From this checklist one can begin to glean where consensus is emerging (didactic and training procedures) and where little consensus yet exists (outcomes assessment of training, patient outcomes, and procedures for sustainability). Finally, we provide conclusions and recommendations based on this review. For this article, we use the following definitions of key terms. Adoption is defined as the decision by a clinician or clinical system to learn and implement a treatment. Dissemination is defined as an effort to facilitate initial adoption, and implementation is defined as the process of transferring the treatment to

The Dissemination and Implementation of Evidence-Based Psychological Treatments 411

R. Kathryn McHugh Photo by Frank Monkiewicz

the clinical setting (e.g., training). Given the clear overlap between dissemination and implementation, many of the programs reviewed include components of both and are discussed as such. Disseminating and Implementing Innovations Difficulty in disseminating new technology is not unique to health care; the slow adoption of innovation has been consistently noted in fields as diverse as agriculture, education, and communication (see Rogers, 2003). The successful adoption of innovation often follows an s-shaped curve characterized by slow initial use that builds more rapidly over time until a “tipping point” is reached (Gladwell, 2000). Once adoption occurs (i.e., a setting chooses to begin utilization of an EBPT), the process of implementation presents another set of challenges to the long-term use and sustainability of the innovation (see Fixsen, Naoom, Blasé, Friedman, & Wallace, 2005). Indeed, sustainability of a seemingly straightforward procedure such as hand washing in hospitals has proven particularly difficult to achieve (Gawande, 2007). Many barriers can arise throughout the dissemination and implementation process, and achieving

412 R. Kathryn McHugh and David H. Barlow

success requires the management of several of these barriers, including negative perceptions of the innovation, challenges to implementing a new procedure in an existing system, and the potential for drift in utilization over time. Perhaps the greatest challenge to these efforts relative to EBPTs is training clinicians to competently administer treatments. The implementation of EBPTs may be particularly difficult relative to the implementation of other types of innovations (e.g., medication prescribing practices) because of the complex and nuanced nature of psychological therapies. Successful training of clinicians in EBPTs requires a balance of both didactic training, defined as the methods used for information transfer such as written materials and workshops, and competence training, defined as the process of acquiring skills necessary to administer a treatment skillfully and with fidelity. Traditional means of translating research results into clinical practice rest on the assumption that clinicians will adopt and administer treatments on the basis of published research findings and attendance at didactics alone. However, evidence suggests that didactic training alone, in the form of workshops or basic training materials, is insufficient to create sustainable change in clinician practices (for a review, see Oxman, Thomson, Davis, & Haynes, 1995). Recent studies examining the efficacy of training programs highlight the importance of competence training in addition to didactics (e.g., Crits-Christoph et al., 1998; Miller, Yahne, Movers, Martinez, & Pirritano, 2004). Competence training, typically involving some form of supervision or coaching, has proven a much more elusive target than didactic training; however, some important attempts to better define and evaluate competence are now beginning (e.g., Roth & Pilling, 2007). Moreover, little is known about the degree to which the achievement of competence following training will be maintained over time. Indeed, drift is a major problem in dissemination efforts generally, and thus the evaluation and maintenance of treatment fidelity may be a core component of ongoing training efforts (see McHugh, Murray, & Barlow, 2009). Evaluation of the success of efforts to train clinicians will need to rely on clear definitions of the didactic knowledge and competence benchmarks required for completion of training. Review of Current Efforts In the following sections, we review a sample of dissemination and implementation programs. The discussion of each program includes the following components: description and motivating circumstances, review of didactic and competence training procedures, and methods used for assessing outcomes. Each section concludes with a brief discussion of how the program is targeting usual and customary barriers to adoption. National Programs

Several national-level programs have been initiated to implement an extensive rollout of treatments to a wide range of service providers. The three national initiatives reviewed below are the Improving Access to Psychological Therapies

The Dissemination and Implementation of Evidence-Based Psychological Treatments 413

David H. Barlow

program in the United Kingdom and the Veterans Health Administration and National Child Traumatic Stress Network initiatives in the United States. Improving Access to Psychological Therapies Program. The Improving Access to Psychological Therapies (IAPT) program is the most extensive and centralized effort in the dissemination and implementation of EBPTs to date (see Clark et al., 2009). In 2007, the Department of Health in the United Kingdom announced a large-scale investment in the National Health Service to improve the availability of psychological treatments through providing funding, training, and structure for EBFT dissemination. The Department of Health committed to gradually building funding from 2007 to 2010 to a total of £300 million (approximately 435 million U.S. dollars), with plans to continue funding beyond that time. The National Institute for Health and Clinical Excellence (NICE) treatment guidelines—developed collaboratively by the Department of Health, stakeholders, and health care experts to provide evidence-based treatment recommendations—serve as the evidence base for service selection and provision. Consistent with the NICE guidelines, the IAPT uses a stepped care model in which the results of an initial assessment determine the relevant level of care, such as self-help, computerized therapy, or intervention with a clinician. Training and implementation are organized at the unit of a primary care trust, the unit responsible for health care provision to a particular geographical area under the National Health Service. Implementation begins with the engagement of local stakeholders to facilitate fit to the system’s needs and to

414 R. Kathryn McHugh and David H. Barlow

evaluate potential barriers to adoption. The IAPT funds trainings to ensure that adequate resources are available and emphasizes the balance between didactic and competence components. Modular training programs occurring over the course of a full year combine didactic presentations, discussion-based groups, and independent study with role-playing and/or simulation exercises (e.g., IAPT, 2008). In addition, clinicians are supervised in patient care during training. A recent publication provided a definition of competence in cognitive-behavioral therapy (CBT) that was based on the efforts of an expert panel of clinical researchers who extrapolated from empirically supported treatment manuals the types of skills necessary to successfully administer treatments (Roth & Pilling, 2007). This model guides the training program and emphasizes building both basic and disorder-specific skills. The competences framework also provides guidance on the selection of disorder-specific treatment manuals for training. Assessment of training progress examines both didactic knowledge and skills related to competence across all relevant training areas (e.g., assessment, treatment, diversity) using validated instruments, in addition to written exams measuring didactic knowledge, standardized role plays are assessed by supervisors. To facilitate ongoing training and maintenance of fidelity to interventions, sites are required to identify staff to receive additional supervisory training to serve as trainers of future staff. Certification occurs upon completion of the training and all of its elements, including a minimum required number of direct patient contact hours with supervision. The implementation of the IAPT program has involved the progressive inclusion of sites over time, beginning with two primary care pilot sites within the National Health Service in Newham and Doncaster. Together, these two sites serve a population of over 500,000 people. Preliminary results from these pilot sites are promising. In addition to demonstrating feasibility in increasing referrals and access to treatment (over 5,000 referred and almost 2,000 receiving treatment across both sites in approximately one year) and instituting a standardized outcome monitoring system with strong rates of completion, these sites showed clinical outcomes that were comparable to those in research studies (50%–60% recovery rates and effect sizes for outcome measures ranging from 0.98 to 1.26). Improvements in patient employment rates (as an index of functioning) were also found (Clark et al., 2009). As the IAPT expands to new sites, ongoing program evaluation will be conducted to examine progress toward the initiative’s goals. Demonstrating the largest system-wide commitment to disseminating and implementing EBPTs to date, the IAPT is a pioneering effort backed by substantial resources and government support. Training and assessment of outcomes of training, as well as patient outcomes are particular strengths of this effort. Training consists of a full year of ongoing didactic and competence-based learning opportunities, structured similarly to a full-time academic course. These learning opportunities focus on concrete definitions of competence and utilize ongoing evaluation to ensure that trainees arc acquiring the relevant knowledge and skills. The IAPT utilizes patient outcomes as the ultimate marker of the program’s success. This measure of program success differs from that used in many other programs and may be a particular strength as it

The Dissemination and Implementation of Evidence-Based Psychological Treatments 415

provides a direct marker of the program’s ultimate goal. The data available to date support the feasibility of collecting outcomes data at a high rate as well as positive symptom and quality-of-life outcomes in line with those achieved in efficacy trials. The longer term sustainability of the program’s ability to maintain clinician adherence and competence and continue to improve or maintain positive patient outcomes is currently unclear given that this program is in an early stage of implementation. The Veterans Health Administration. The Veterans Health Administration (VHA) is the largest organized system of health care in the United States. In response to the publication of the President’s New Freedom Commission on Mental Health report in 2004, a workgroup was created to conduct a needs assessment specific to the VHA system. This work resulted in the development of the Mental Health Strategic Plan in 2004, which called for the integration and improvement of mental health care within the VHA and a commitment to support over 200 initiatives in this area. Among the goals of this plan was the implementation of dissemination efforts with a focus on the translation of EBPTs into clinical practice through training clinicians within the VHA system. Funding for this initiative was $316 million in fiscal year (FY) 2007 and $380 million in FY 2008. The process of identifying treatments to be disseminated first involves identifying a need area in the system, for which EBPTs are then identified through consideration of the VHA/Department of Defense best practice guidelines. Clinician preferences are accommodated through offering training in multiple EBPTs for a given need area. For example, trainings have been conducted for both cognitive-processing therapy (CPT; Resick, Monson, & Chard, 2007) and prolonged exposure therapy (PE; Foa, Hembree, & Rothbaum, 2007) for the treatment of posttraumatic stress disorder (PTSD) and for both CBT and acceptance and commitment therapy (ACT) for the treatment of depression and anxiety. All VHA centers are required to have staff trained in EBPTs; however, the clinical use of FBPTs is not mandated, and the decision on whether to utilize a particular intervention remains with the clinician and the patient. Overall, the goal of the VHA program is for EBPTs to reflect the usual standard of care and for this standard to be available to all patients within the system (B. E. Karlin, personal communication, August 6, 2008). Efforts to disseminate EBPTs are coordinated by the VHA Central Office for rollout throughout the system; for each treatment initiative, a coordinating site within the system is identified on the basis of expertise in that area. These sites, often led by experts in the respective treatment being offered, assist in the implementation of EBPTs under direction from the VHA Central Office. The overall structure of training typically involves an intensive workshop, including a didactic component, and experiential strategies such as small-group activities and role plays. Subsequent to the workshop, participants return to their sites and begin treating patients, and some ongoing consultation is required to facilitate competence. Completion of a training program requires participation in both the initial workshop and ongoing consultation activities. The frequency and duration of ongoing consultation vary depending on the particular treatment, often occurring weekly for six months. The consultant monitors treatment administration and provides ongoing feedback

416 R. Kathryn McHugh and David H. Barlow

and guidance to clinicians as they learn the intervention. In addition, patient outcomes are collected, such as symptom-change and quality-of-life measures. Furthermore, like the IAPT model, a “train the trainer” model emphasizes the development of supervisory skills within treatment teams to facilitate ongoing use of and adherence to the intervention. Given its particular relevance to the VHA system, PTSD was identified as a primary area for intervention, and thus a nationwide training for CPT was the earliest funded initiative through this program (Resick, Foa, Ruzek, & Karlin, 2008). Trainings were coordinated through the treatment developers at the National Center for PTSD in Boston, Massachusetts. In the first phase of this initiative, training materials, including manuals, workshop materials, and a training video library, were developed, and a conference was conducted to train experts to run workshops and provide ongoing consultation. During the second phase, CPT trainers conducted 22 official regional CPT workshops, which trained 839 VHA clinicians in CPT. There were 44 additional CPT workshops conducted by CPT trainers outside of the initiative, which resulted in the training of an additional 1,350 VHA and DOD clinicians in 2007–2008. Overall, a total of 1,488 VHA and DOD clinicians participated in a two day CPT training during this time period. Since this time, an additional round of CPT training has been successfully completed, as well as trainings in PE, CBT, and ACT. These training workshops were followed by case consultation, offered 25 hours per week (with a maximum of eight clinicians per call), via the VHA National Teleconferencing System, in order to achieve certification, clinicians must attend a two-day workshop, complete either four CPT individual cases or two group cases per the CPT protocol, and actively participate in at least 1.0 consultation calls during which they receive support and consultation to implement the CPT protocol successfully. Clinicians must also submit sample progress notes, fidelity measures, and a treatment summary as part of the certification process. A preliminary program evaluation survey suggested that more than 75% of clinicians surveyed had treated at least one patient using CPT, with a mean of approximately eight patients receiving the intervention per clinician, or approximately 6,000 patients (P. A. Resick, personal communication, May 6, 2008). Like the IAPT, the VHA effort has several advantages because of its large scope and structural support, such as the ability to roll out trainings to large numbers of clinicians. Funding is provided for trainings to facilitate completion, and enthusiasm is often built within the system by using advocates within sites to champion efforts. The VHA program emphasizes flexibility in not mandating the use of EBPTs and thus allows for an emphasis on patient preference and clinician judgment in the decision regarding whether to utilize a treatment. Flexibility may be particularly facilitative of adoption rates (see Rogers, 2003); however, the impact of this strategy on adherence/fidelity and subsequently on outcomes remains unclear at this time. For example, attendance at consultation calls may be variable, and the literature provides little guidance on how much supervision is necessary to achieve competence. Thus, the percentage of clinicians who satisfactorily complete training is not clear. Given the use of multiple implementation models, outcomes data will eventually provide guidance on the most effective model within the VHA system.

The Dissemination and Implementation of Evidence-Based Psychological Treatments 417

In addition to offering ongoing training through the Central Office, the VHA system utilizes a “train the trainer” model to maintain sustainability, and the effectiveness of this strategy will need to be evaluated over time. The National Child Traumatic Stress Network. The mission of the Substance Abuse and Mental Health Services Administration (SAMHSA) is to improve access to and quality of clinical care through facilitating the dissemination of evidence-based practices. SAMHSA is currently funding several major programs; among these is the National Child Traumatic Stress Network (NCTSN), funded by the Center for Mental Health Services. This is a broad-reaching initiative that involves collaboration among more than 50 universities and community treatment facilities in the development, evaluation, and dissemination of EBPTs for traumatized children. Its budget allocation was $29 million for FY 2007 and $33 million for FY 2008. Implementation efforts within the NCTSN have utilized a learning collaborative (LC) model, which aims to train clinicians in clinical competence and implementation capability in delivering EBPTs for trauma and to facilitate long-term sustainability of adoption (see Amaya-Jackson & DeRosa, 2007; Markiewicz, Ebert, Ling, Amaya-Jackson, & Kisiel, 2006). The LC model was adapted from the Breakthrough Series Collaborative model (Institute for Healthcare Improvement, 2003), which has been used across heterogeneous medical settings for the dissemination of best care practices. This model places particular emphasis on a scientific approach to change characterized by cycles of goal setting, implementation, and assessment. In the LC model, a project begins with the identification of a general clinical need, the development of training materials, and the identification of faculty trainers with expertise in clinical and implementation training. Treatments for dissemination are selected by the NCTSN on the basis of their empirical support and evidence for applicability for use among diverse patient populations. Groups consisting of supervisors, clinicians, and their administrative leaders can apply to participate in the training on the basis of their particular system needs; a needs/ readiness assessment is then conducted for each group. The training process is described in detail in the Learning Collaborative Information Packet and the NCTSN Learning Collaborative Toolkit (Markiewicz et al., 2006: NCTSN, 2007). The didactic portion of training includes pre-learning-session Web-based videos and readings, followed by three two-day workshops (separated by 9–12 months) that emphasize guided learning activities, role plays, and breakouts. In the time between these trainings, groups implement the treatment and assess progress. During this time, several forms of supervision and/or consultation are available (e.g., expert consultation, Web conferences). Active learning and collaboration among clinical teams is emphasized in this model to encourage motivation and to facilitate implementation among the groups trained. Training of clinicians to competence is assessed through consultation calls and collection of patient outcomes; specific procedures for competence assessment (e.g., supervisor rating of clinician’s skill at utilizing the protocol) and fidelity assessment (e.g., use of a supervisor-completed fidelity checklist) vary depending on the specific LC. Preliminary outcome data for this model suggest success in facilitating the adoption of EBPTs. In a pilot study of the implementation of trauma-focused

418 R. Kathryn McHugh and David H. Barlow

cognitive-behavioral therapy (TF-CBT; Deblinger & Heflin, 1996) at 12 sites, with 11 providing one-year follow-up data, all sites continued to provide the intervention to patients, and more than 70% of the sites increased the availability of TF-CBT to patients through providing additional training (Amaya-Jackson, Ebert, Forrester, & Deblinger, 2008). Staff turnover was reported as a major barrier to sustainability; however, even sites reporting high turnover were able to maintain posttraining levels of implementation, and many sites actually reported spread of the intervention to other locations or affiliates. The NCTSN targets perceptions of EBPTs by involving stakeholder groups at all stages of the process, from research through implementation; thus, barriers can be identified and addressed early in the dissemination process. Like many other dissemination efforts, the NCTSN targets early adopters, who are motivated to apply to receive training based on an identified need area. Training utilizes both didactic and competence components and provides a trial period following initiation of training during which barriers to implementation can be identified and then addressed at the second training session; this model has been used successfully in quality improvement in medical settings (e.g., to reduce adverse drug events; Leape et al., 2000). Indeed, the utilization of a well-established model for quality improvement is a particular strength of this program. Ongoing evaluation of number of patients treated, treatment dropout, and clinical outcomes as well as of treatment adherence (e.g., through ongoing monitoring by supervisors) is emphasized, and thus a range of outcomes is assessed, similar to the situation in the IAPT. A lack of sustainability has been specifically targeted as a potential barrier to ongoing use of services, so efforts have been made to enhance sustainability through a focus on long-term maintenance of funding support, ongoing consultation opportunities, and the use of a “train the trainer” model; as in other programs, the effectiveness of sustainability strategies is unknown at this early stage. State Programs

In recent years, state mental health systems have also increased efforts toward implementing evidence-based practice. Initiatives have been heterogeneous given the variety of structures involved in state mental health systems and have ranged in scope from small (e.g., implementing one EBPT) to system-wide. Two of the leading state initiatives, those of Hawaii and New York, are reviewed here. Hawaii. Hawaii is one of the leading states in the implementation of evidence-based practice and quality improvement in mental health care. In the mid-1990s, the state began a major restructuring of mental health care services for children in response to the settlement of a civil lawsuit alleging a failure of the state to provide sufficient services to children with disabilities. The need for improved treatment outcomes within this new system was identified, highlighting the importance of the implementation of evidence-based practice into children’s mental health care in the state (Chorpita et al., 2002). Early implementation efforts focused on needs assessment and identification of standards for evidence-based practice. Assessments involving community stakeholders were conducted to evaluate specific need areas and potential

The Dissemination and Implementation of Evidence-Based Psychological Treatments 419

barriers to the implementation of EBPTs. Furthermore, efforts were made to build support for systemic change, including the provision of materials relevant to EBPTs (e.g., research findings, clinical materials) and presentations to stakeholder groups to address concerns and misperceptions about potential changes. The Empirical Basis to Services Task Force was created by the Child and Adolescent Mental Health Division (CAMHD) of the Hawaii Department of Health in 1999, and it was charged with creating a system for identifying treatments and developing practice guidelines. The group based the identification of treatments largely on the guidelines utilized by the Society of Clinical Psychology (Division 12) of the American Psychological Association (APA; e.g., APA, 1995, which was based on the work of the APA Task Force on Psychological Intervention Guidelines, and Chambless et al. 1998); however, specific needs and priorities of the community were considered, such as the acceptability of treatments to patients (as reflected by dropout rates), length of treatment, and the difficulty of training treatment providers (Chorpita & Daleiden, 2007). The provision of training has occurred both through statewide trainings in treatments targeting specific need areas and through initiatives by the University of Hawaii. CAMHD has coordinated treatments in several specific EBPTs, such as multisystemic therapy (MST; Henggeler, Sehoenwald, Borduin, Rowland, & Cunningham, 1998), which has been disseminated throughout the state system, and, more recently, functional family therapy (FFT; Alexander, Pugh, Parsons, & Sexton, 2000) and multidimensional treatment foster care (MTFC; Chamberlain & Reid, 1998). In addition, a major innovation arising from this program is the examination of specific evidence-based treatment procedures or practice elements (e.g., social skills training, graded exposure) utilized in existing evidence-based treatment protocols (Chorpita & Daleiden, 2007; Chorpita, Daleiden, & Weisz, 2005). Recently, CAMHD began statewide trainings in these practice elements that can be combined in idiographic ways to match the needs of specific patients following guidelines provided by the CAMHD Clinical Services Office regarding how to combine components of evidence-based procedures to build a treatment plan. Although trainings vary according to the intervention and the sponsoring body (e.g., CAMHD; MST Services, LLC; MTFC Consultants), the typical structure incorporates both didactic and competence components. More broadly, information dissemination is facilitated by an innovative Web-based system, which includes detailed information summarizing the research literature and provides a structure for clinicians to gather information relevant to their particular needs (e.g., patient characteristics, treatment setting). For the package trainings, didactic trainings in the form of workshops that include role-playing activities are followed by ongoing supervision and coaching, in addition, competence training has focused on the use of both experts and peer groups for supervision and consultation. Initiatives often follow didactic components with a minimum of six months of phone consultation. The structure of the training, consultation, and assessment varies depending on the particular program (e.g., see MST section later in this article). The implementation of the practice elements training is relatively new, and thus the more comprehensive competence component is not currently in place.

420 R. Kathryn McHugh and David H. Barlow

Program evaluation through CAMHD has been emphasized in the EBPT movement in Hawaii. Preliminary data suggest that time in treatment has decreased and rate of improvement has increased, which suggests that the program has had initial success (Dalciden, Chorpita, Donkervoet, Arensdorf, & Brogan, 2006). One EBPT that has seen extensive rollout in Hawaii is MST. An open trial examining outcomes among 254 youths receiving MST showed positive results demonstrating the feasibility and acceptability of the treatment in the care setting and found substantial improvement in clinical outcomes, although rates of improvement were somewhat tower than those in efficacy trials (Tolman, Mueller, Daleiden, & Stumpf, 2007). Since 2001, approximately 300 patients annually have been treated with the MST protocol. Moreover, in addition to evaluation of clinical outcomes, clinicians are asked to report their use of treatment practices in order to provide feedback to providers and case managers and to inform the evidence available for the use of treatment practices (Higa-McMillan, Daleiden, Pestle, & Mueller, 2008). The Hawaii system targets several barriers to adoption. Significant focus on fit is accomplished by attending to the specific characteristics and needs of the system in selecting and adapting treatments. This includes evaluation of system needs by involving various stakeholders at the outset in building support for and anticipating barriers to implementation. The innovative practice elements component is notable because it reduces complexity by eliminating the large number of treatment manuals in favor of a specific set of principles that can be applied across problem areas. Furthermore, training may be simplified relative to training in several individual treatment protocols, and flexibility is afforded by the rollout of both practice elements and package treatments (e.g., MST), A drawback, however, is that clinicians must make a number of individual decisions on which modules are indicated, and the reliability of these decisions is not yet clear. Sustainability is emphasized through the use of ongoing supervision models, often through the training of on-site supervisors (i.e., “train the trainer”). Outcomes data on sustainability are not yet available. New York. The New York State Office of Mental Health gathered several focus groups starting in 2001 to evaluate the potential implementation of EBPTs into state services (Carpinello, Rosenberg, Stone, Schwager, & Felton, 2002). Various stakeholders were represented at the meetings, where needs and potential barriers were discussed and initial support building occurred for the improvement of mental health care services for both children and adults. Several programs have been initiated in New York since these meetings, including implementation of EBPTs such as FFT in clinical settings. One of the major initiatives within New York State has been Achieving the Promise for Children, Youth and Families, a $62 million initiative to improve services for children’s mental health. The “Achieving the Promise” initiative consists of several components aimed at facilitating improvement in mental health care services for children. These components include efforts aimed to improve assessment, train clinicians to administer EBPTs, identify community advocates for programs, and implement incentives for the use of EBPTs (see Bruns & Hoagwood, 2008). The Evidence-Based Treatment Dissemination Center (EBTDC), a collaboration between the Office of Mental Health and Columbia University, was developed

The Dissemination and Implementation of Evidence-Based Psychological Treatments 421

through this initiative and serves as a coordinating center for training in interventions. A steering committee identifies the priorities for training on the basis of the needs of the system and available evidence. Trainings offered through the EBTDC acknowledge the importance of both didactic and competence training. The general model consists of a Web-based training and a three-day didactic workshop followed by one year of phone consultation every two weeks for one hour in a small group format. Furthermore, evaluation of diagnosis, symptoms, and functioning at pre-, mid-, and posttreatment is emphasized as critical to successful adoption. The Office of Mental Health also contracts with external expert trainers, usually treatment developers, to conduct implementation efforts. For example. FFT has begun to be implemented in New York State through collaboration with FFT, Inc., the training organization for this intervention. This initiative also reflects an emphasis on competence training, with on-site start-up and follow-up trainings in the first six months followed by continued phone consultation and monitoring for two years. To receive certification, clinicians are required to complete all didactic training components, participate in at least 75% of calls, and complete outcome evaluations. A major training initiative through the EBTDC involved training over 300 clinicians in CBT for trauma and depression throughout New York State between June and October of 2006 (Gleacher et al., 2007). Agencies applied for admission into the trainings on the basis of fit to the program (e.g., number of patients requiring a particular service). Trainings consisted of an initial workshop followed by biweekly group phone consultation for one year, during which a minimum number of case presentations and full completion of a manual-based course of treatment with the use of outcome measures was required. Supervisors at each site received additional consultation to facilitate training and adherence; however, formal monitoring of fidelity was not conducted. In total, over 400 clinicians completed training, and although fewer than half of clinicians reported completing a full course of treatment (48% for depression, 46% for trauma), the majority of clinicians reported that they used parts of the manual and intended to use the treatment in the future (Gleacher et al., 2007). Given the effort’s success in training a large number of clinicians, the training was offered again in 2007. Overall, the EBTDC has trained approximately 400 clinicians annually, but the number of patients receiving the full or a partial treatment protocol is not available. The efforts in New York involve several initiatives that may differ in their specific procedures for implementation. The EBTDC provides systemic support through funding for widespread training in EBPTs. Trainings are unique in using a longer period of consultation calls (one year) than is used in most other programs. However, despite the large number of clinicians who enroll in this program, many do not complete the training requirements. Given the lack of availability of fidelity measures, the degree to which competence is reached among those who complete training standards or those who partially complete training is unclear. Strategies to improve clinician retention and evaluate fidelity may help to better understand the effectiveness of this program. Furthermore, plans for sustaining change, such as funding for clinical centers and ongoing provision of training, will need to be evaluated over time to determine their success at preventing drift.

422 R. Kathryn McHugh and David H. Barlow

Programs from Treatment Developers

Some of the most successful dissemination efforts have been those pursued by treatment developers. Two programs that have demonstrated particular success are reviewed below: multisystemic therapy and dialectical behavior therapy. Multisystemic Therapy. MST is an empirically supported treatment approach for antisocial behavior in adolescents (Kazdin & Weisz, 1998) that has been widely disseminated and adopted both locally (30 states in the United States) and internationally (eight countries; Schoenwald, Heiblum, Saldana, & Henggeler. 2008). MST Services, LLC, a private, for-profit corporation, was started in 1996 to provide structured implementation training to community treatment providers. Efforts to implement MST in service provision settings have been described in some detail (e.g., Edwards, Schoenwald, Henggeler, & Strother, 2001; Schoenwald et al., 2008). The implementation process for a site begins with an in-depth assessment of needs and barriers. Sites include both public and private mental health service settings that provide a range of types of services (e.g., outpatient, home-based). MST trainers and community leaders collaborate to identify the fit of MST to the clinical needs of the community and to build support for adoption. Factors including financial resources, sustainability, compatibility with organizational beliefs and goals, and potential infrastructure barriers are all considered, and the results of this assessment are used to guide the implementation process. Following a decision to adopt, an extensive training program is conducted (Edwards et al., 2001). Didactic training (including both education and experiential components) consists of an initial five-day, on-site training, quarterly booster training, and provision of written materials to the program. Competence training is a critical component of this process and consists of both weekly on-site supervision led by a trained staff supervisor and weekly phone consultation led by an off-site expert supervisor. In-house supervisors receive training to lead regular group and/or individual supervision. In addition, ongoing consultation with an MST expert is used to monitor supervisor and clinician fidelity to the treatment model and to assist with the learning process. The frequency and duration of both in-house supervision and external consultation are often high early in implementation of MST and then may decrease over time. Given the importance of fidelity to treatment outcomes for MST (Schoenwald, Sheidow, Letourneau, & Liao, 2003), adherence to the treatment is emphasized and monitored through the use of monthly, empirically validated adherence measures of clinician practices completed by the patient’s family, which are scored using an Internet-based system that provides immediate feedback to clinicians. Furthermore, supervisors are responsible for also monitoring clinician adherence. Supervisor adherence to the supervisory guidelines is also regularly monitored by MST consultants, who rate and provide feedback on their adherence to the supervisory procedures. Thus, monitoring is facilitated through both the regular, standardized measurement of adherence and booster trainings that provide opportunity to manage difficulties with implementation.

The Dissemination and Implementation of Evidence-Based Psychological Treatments 423

Several strategies have been undertaken to assess clinical outcomes, including parent and clinician report and the use of archival data regarding criminal outcomes (Schoenwald, 2008), For example, unlike results in Hawaii, in a large study of the transportability of MST, clinical outcomes similar to those found in efficacy trials were noted (Schoenwald et al., 2003). Furthermore, clinician adherence has been shown to be significantly associated with clinical outcome (see Schoenwald, 2008). Moreover, evaluation of the implementation model has suggested that poorer outcomes are seen in the absence of ongoing consultation and fidelity checks (Henggeler, Melton, Brondino, Scherer, & Hanley, 1997). Efforts to implement MST have addressed barriers in a manner very consistent with the strengths of implementing an individual treatment. The fit to the clinical service environment is strongly emphasized through the use of extensive needs assessment and involvement from stakeholders and advocates. Furthermore, training is extensive and includes a particularly lengthy competence component relative to training in other programs, particularly regarding the emphasis on fidelity. The use of a Web-based system to facilitate fidelity is unique to this program and may provide a particularly cost-effective method for maximizing fidelity and preventing drift. Sustainability of changes is maximized through training of on-site supervisors and provision of ongoing monitoring and booster trainings. Preliminary results suggest that this model has been particularly successful; further evaluation will provide additional information on the success of these strategies. Dialectical Behavior Therapy. Dialectical behavior therapy (DBT) is a type of CBT developed for the treatment of borderline personality disorder (Linehan, 1993). It has demonstrated both efficacy (e.g., Linehan, Comtois, & Murray, 2006) and effectiveness (e.g., Kroger et al., 2006) and has been widely disseminated both within the United States and internationally. From its initial development, the transportability of DBT has been a focus of its developers. The barriers to successful implementation in community settings were evaluated during the process of treatment development and were used to inform the treatment manual (see Linehan, 1993). A not-for-profit organization (Behavioral Tech. LLC) was developed to coordinate the dissemination and training of DBT. Dissemination occurs both at the state level and for interested groups of clinical providers. Linehan described the latter type of dissemination as an “early adopter model” in which interested groups who demonstrate high levels of readiness and interest seek training (M. M. Linehan, personal communication, July 30, 2008). More traditional dissemination methods, including the sale of instructional videos and treatment descriptions, have been utilized widely by clinicians. Trainings occur on several levels of intensity ranging from basic training to advanced intensive training. Once basic skills have been obtained through an introductory workshop or review of DBT materials, clinicians are eligible to participate in intensive trainings. These trainings arc designed for treatment teams who participate in two five-day workshops separated by several months of implementation and evaluation. The first workshop includes components of other intensive workshops (e.g., lectures, group activities), and the second focuses more on consultation and evaluation of the implemented treatment services.

424 R. Kathryn McHugh and David H. Barlow

Ongoing consultation is available to sites as needed to facilitate maintenance of fidelity and to address barriers through problem solving. The use of treatment teams that facilitate ongoing adherence and monitoring of outcomes is emphasized as critical to the administration of DBT. Certification is currently not available but is in the planning stages. In the implementation of a DBT program, there is a strong emphasis on monitoring outcomes for sustained fidelity and quality improvement (Comtois et al., 2007). Linehan described this process as critical to the long-term success of training efforts to maximize effectiveness and to prevent drift (M. M. Linehan, personal communication, July 30, 2008). The team-based approach may provide both support and a means to facilitate continued fidelity to the treatment model and evaluation procedures. Assessment of fidelity is based on the judgment of the treatment team. Adoption of DBT has been widespread, and the demand from providers and mental health systems remains high. Almost 2,500 clinicians have received training through Behavioral Tech from 2003 to 2007 alone, and DBT has been introduced to 31 states and 12 countries (Linehan, Manning, & Ward-Ciesielski, 2008). Monitoring of patient outcomes (e.g., hospitalizations) within Behavioral Tech has suggested feasibility of the implementation model and success of adoption into clinical settings. The results of this monitoring, along with feedback from clinical teams regarding barriers to implementation, have been used to continually improve the training model. Barriers to implementation have been vetted and targeted throughout the development and evaluation of DBT, and the focus has been on training and sustainability. Trainings have been adapted to maximize the achievement of competence and to allow for the best fit for the clinical system. The training model, similar to the one used in the NCTSN, allows for a period of implementation between trainings, which is different from the usual one-time didactic training followed by a period of supervision and implementation. This allows the management of both expected and unanticipated barriers by providing an opportunity for additional intensive consultation after the first period of implementation. Monitoring of clinical outcomes has been emphasized for both initial training and ongoing implementation to prevent drift and maximize effectiveness. The use of such monitoring in controlled investigations may provide more information on the effectiveness of each component of this program. Discussion In this article we have reviewed some of the leading efforts by federal, state, and private organizations to disseminate and implement EBPTs and have examined the strategies currently used to facilitate the successful transfer of interventions into service provision settings. More in-depth descriptions of these efforts are available (see Clark et al., 2009; Schoenwald, 2008) or in progress (McHugh & Barlow, in press). As we noted earlier, many of these programs were created with a sense of urgency that precluded the development of a consensus on, or even knowledge of, procedures required to achieve success; but at this early stage in these efforts, there is a clear need

The Dissemination and Implementation of Evidence-Based Psychological Treatments 425

for a consensus on the best procedures for successful adoption and implementation of EBPTs. On the basis of this review and the currently available data and in consideration of the tenets of dissemination and implementation science, we found several procedures emerging as important in these efforts. From these leading efforts, we have extracted a variety of procedures that, at this time, are used by at least some groups. Table 19.1 lists the procedures identified in this review and the utilization of these procedures by each of the initiatives reviewed above (Hawaii was not included in the table because the “package” treatments have been implemented by outside organizations and thus utilize a range of procedures). Given the particular importance of training to the success of these efforts, much attention is focused on the nature of training. Table 19.1 also includes components of needs assessment, evaluation, and practices to facilitate sustainability, Organizational factors, which are also a critical component of these efforts, are not discussed in detail; discussion of such factors can be found elsewhere (e.g., Fixsen et al., 2005). TABLE 19.1 Procedures for Comprehensive Assessment and Training in Leading Dissemina-

tion Programs

Program Standard Needs and barrier assessment Agency driven Early adopter Heterogeneous stakeholder involvement Structured needs assessment Training structure Spaced training Booster trainings/ advanced training Didactic training Training materials Workshop

IAPT

VHA

NCTSN

New York State

DBT

MST

X

X X X

X X

X

X X X

X X X

X X X

X

X

X

X

X

X

Xa X

X

X X

X X

X

X

X

X

X

X

X

X

X

X

X

X

X (Continued )

TABLE 19.1 (Continued)

Program Standard

IAPT

VHA

NCTSN

New York State

DBT

Web-based individual training Assessment of knowledge Competence training No. of patients required to be seen Supervision In-person supervision (individual or group) Telephone consultation (individual or group) Tape feedback Fidelity to treatment Duration of expert supervision (months) Outcomes collected Patient outcomes No. of patients receiving services/ service outcomes? Pre-post symptoms Ongoing symptom monitoring Impairment/ quality of life Assessment of competence

X

X

X

X

X

X

X

X

MST

X

8b

2–4

1c

3

Xd

X

X

X

X

X

X

X

X

X

X

X

X X

X X

X

X

X X

12

varies

9–12

ongoing

ongoing

X

12

X

X

X

X

X

X

X

X

X

X

Xe

X

Program Standard

IAPT

VHA

Validated instrument Clinician assessed Supervisor assessed Patient assessed Individual feedback to clinicians of assessment data Certification Clinician/ training outcome No. of clinicians trained Clinician attrition Percentage who achieve competence Sustainability Train the trainerg Structured long-term consultation (> 1 year) Peer consultation network

X

X

X

X

X

NCTSN

New York State

DBT

X

MST

X X

X

X

X

X

X

X

X

Xf

unavailable > 1,500

unavailable > 1,200

> 2,500

unavailable

unavailable < 30%

unavailable unavailable unavailable N/A

unavailable unavailable unavailable unavailable N/A

N/A

X

X

X

X

X X

X X

X

X

X X

Note. IAPT = Improving Access to Psychological Therapies program; VHA = Veterans Health Administration; NCTSN = National Child Traumatic Stress Network; DBT = dialectical behavior therapy; MST = multisystemic therapy; N/A = not applicable. IAPT offers a year-long full time course. b Minimum of 200 hours of assessment and treatment. Must be completed with fidelity. d Must initiate program and begin implementing program, but there is no minimum number of patients. e Monitoring is based on identified treatment goals. f Programs are certified, not individual clinicians. g Includes training on-site supervisors to provide ongoing supervision to staff and does not preclude mandated expert training for new staff members (e.g., MST). a c

428 R. Kathryn McHugh and David H. Barlow

It is important to note that studies evaluating the efficacy of dissemination and implementation programs in general, and of procedures specifically, are in very early stages. Moreover, these and other programs are moving forward at a rapid pace, and thus efforts to continuously improve these procedures arc ongoing. The standards listed in Table 19.1 reflect our interpretation of the components of programs used by the leading efforts at this time. From this framework, it is unclear what combination of procedures may prove to be critical to the success of these efforts. It is possible that only some of the procedures listed in the table are necessary to achieve high levels of adoption, competence, and sustainability or that additional procedures not yet introduced may prove to be important. In addition, from a cost-efficacy perspective, it will be important to both maximize cost savings and ensure that sufficient funding and effort are invested to implement the procedures needed for meaningful and sustainable change. For example, given the importance of fidelity to outcomes, the failure to include fidelity monitoring within these efforts may attenuate the outcomes achieved (McHugh et al., 2009). Future Directions Given the relative lack of data regarding the efficacy of specific dissemination and implementation procedures for increasing access to EBPTs, evaluation is a particularly important topic for future research. For example, few programs have information on the number of clinicians (or clinical teams) who fail to reach competence standards after initiating training, which may have major implications for future initiatives. In order to determine the best practice for these efforts, examination of outcomes in ongoing efforts is needed. Such examination should not be limited to clinical outcomes and should also include potential mediators of successful adoption, such as the duration of supervision necessary to achieve competence, the best procedures to facilitate sustainability (e.g., “train the trainer” models), and the most effective means of gaining stakeholder support and addressing negative perceptions about EBPTs among late adopters. The programs reviewed in this article are some of the leaders in these efforts, and using the knowledge gained from these and other ongoing initiatives is critical to informing future dissemination and implementation efforts. To this end, we suggest that all programs begin to assess both training outcomes and clinical outcomes using the procedures presented in Table 19.1. Thus, training outcomes would routinely include acquisition of didactic knowledge (currently present in all leading programs) as well as objective assessment of fidelity including clinician competence and number and percentage of clinicians who complete training, achieve competence, and sustain competence (currently present in only a few leading programs). Clinical outcomes, including number of individuals receiving the intervention, problem remediation, and reductions in impairment and increases in quality of life, all benchmarked to efficacy studies, would comprise the more important result. Governments, public health authorities, and individuals suffering from psychological problems around the world are demanding increased access to psychological treatments, and the urgency of this demand has gotten ahead of

The Dissemination and Implementation of Evidence-Based Psychological Treatments 429

the determination of best practices to achieve it. Scientists and clinicians must work together to meet this demand with the same sense of urgency expressed by health care policymakers and the individuals we serve. References Alexander, J. F., Pugh, C., Parsons, B. V., & Sexton, T. L. (Eds.). (2000). Functional family therapy (2nd ed.). In D. S. Elliott (Series Ed.), Blueprints for violence prevention (BP-003). Boulder, CO: Center for the Study and Prevention of Violence, Institute of Behavioral Science, University of Colorado. Amaya-Jackson, L., & DeRosa, R. R. (2007). Treatment considerations for clinicians in applying evidence-based practice to complex presentations in child trauma. Journal of Traumatic Stress, 20, 379–390. doi:10.1002/jts.20266 Amaya-Jackson, L., Ebert, L., Forrester, A., & Deblinger, E. (2008, March). Fidelity to the Learning Collaborative Model: Essential elements of a methodology for the adoption and implementation of evidence-based practices. Paper presented at the meeting of the National Child Traumatic Stress Network, Anaheim, CA. American Psychological Association. (1905). Template for developing guidelines: Interventions for mental disorders and psychosocial aspects of physical disorders. Washington D.C.: American Psychological Association. Barlow, D. H. (Ed.), (2008). Clinical handbook of psychological disorders (4th ed.). New York: Guilford Press. Bruns, H. J., & Hoagwood, K. E. (2008). State implementation of evidence-based practice for youths, part I: Responses to the state of the evidence. Journal of the American Academy of Child & Adolescent Psychiatry, 47, 369–373. doi: 10.1097/CHI.0b013c816485f4 Carpinello, S. E., Rosenberg, L., Stone, J., Schwager, M., & Felton, C. J. (2002). Best practices: New York State’s campaign to implement evidence-based practice for people with serious mental disorders. Psychiatric Services, 53, 153–155. Chamberlain, P., & Reid, J. B, (1998). Comparison of two community alternatives to incarceration for chronic juvenile offenders. Journal of Consulting and Clinical Psychology, 66, 624–633. doi: 10.1037/0022-006X.66.4.624 Chambless, D. L., Baker, M. J., Baucom, D. H., Beutler, L. E., Calhoun, K. S., Crits-Christoph, P., . . . Woody, S. R. (1998). Update on empirically validated therapies, II. The Clinical Psychologist, 51(1), 3–15. Chorpita, B. F., & Daleiden, E. L. (2007). 2007 biennial report: Effective psychosocial interventions for youth with behavioral and emotional needs. Retrieved from Hawaii Department of Health, Child and Adolescent Mental Health Division: http://hawaii.gov/health/ mental-health/camhd/library/pdf/ebs/ebs012.pdf Chorpita, B. F., Daleiden, E. L., & Weisz, J. R. (2005). Identifying and selecting the common elements of evidence based interventions: A distillation and matching model. Mental Health Services Research, 7, 5–20. doi:10.1016/j.appsy.2005.05.002 Chorpita, B. F., Yim, L. M., Donkervoet, J. C, Arensdorf, A., Amundsen, M. J., McGee, C., . . . Morelli, P. (2002). Toward large-scale implementation of empirically supported treatments for children: A review and observations by the Hawaii Empirical Basis to Services Task Force. Clinical Psychology: Science and Practice, 9, 165–190. Clark, D. M., Layard, R., Smithies, R., Richard, D. A., Stickling, R., & Wright, B. (2009). Improving access to psychological therapy: Initial evaluation of two UK demonstration sites. Behaviour Research and Therapy, 47, 910–920. doi:10.1016/j.brat.2009.07.010 Comtois, K. A., Koons, C. R., Kim, S. A., Manning, S. Y., Bellows, E., & Dimeff, L. A. (2007). Implementing standard dialectical behavior therapy in an outpatient setting.

430 R. Kathryn McHugh and David H. Barlow In L. A. Dimmeff & K. Koerner (Eds.), Dialectical behavior therapy in clinical practice: Applications across disorders and settings (pp. 37–68). New York: Guilford Press. Crits-Christoph, P., Frank, E., Chambless, D. L., Brody, C., & Karp, J. F. (1995). Training in empirically validated treatments: What are clinical psychology students learning? Professional Psychology: Science and Practice, 26, 514–522. doi:10.1037/0735-7028/26.5.514 Crits-Christoph, R., Siqueland, L., Chittams, J., Barber, J. P., Beck, A. T., Frank, A., . . . Woody, G. (1998). Training in cognitive, supportive-expressive, and drug counseling therapies for cocaine dependence. Journal of Consulting and Clinical Psychology, 66, 484–492. doi:10.1037/0736-9735.25.3.483 Daleiden, E. L., Chorpita, B. F., Donkervoet, C., Arensdorf, A. M., & Brogan, M. (2006). Getting better and getting them better: Health outcomes and evidence-based practice within a system of care. Journal of the American Academy of Child and Adolescent Psychiatry, 45, 749–756. doi:10.1097/01.chi.0000215154.07142.63 Deblinger, E., & Heflin, A. H. (1996), Treating sexually abused children and their nonoffending parents: A cognitive-behavioral approach. Thousand Oaks. CA: Sage. Edwards, D. L., Schoenwald, S. K., Henggeler, S. W., & Strother, K. B. (2001). A multi-level perspective on the implementation of multisystemic therapy (MST): Attempting dissemination with fidelity. In G. A. Bernfield, D. P. Farringston, & A. W. Leschied (Eds.), Offender rehabilitation in practice: Implementing and evaluating effective programs (pp. 97–120). London, England: Wiley. Fixsen, D. L., Naoom, S. F., Blasé, K. A., Friedman, R. M., & Wallace, F. (2005). Implementation research: A synthesis of the literature. Tampa, FL: University of South Florida, Louis de la Parte Florida Mental Health Institute. Department of Child and Family Studies. Foa, E., Hemhree, E., & Rothbaum, B (2007). Prolonged exposure therapy for PTSD: Emotional processing of traumatic experiences, therapist guide. New York, NY: Oxford University Press. Gawande, A. (2007). Better: A surgeon’s notes on performance. New York, NY: Metropolitan Books. Gladwell, M. (2000). The tipping point: How little things can make a big difference. New York, NY: Little, Brown. Gleacher, A. A., Leviti, J. M., North, M. S., Goldman, E., Hoagwood, K., Albano, A. M., & Radigan, M. (2007, August). The Evidence-Based Treatment Dissemination Center. In T. A. Glover (Chair), Identifying and applying evidence-based mental health practices. Symposium conducted at the meeting of the American Psychological Association, San Francisco, CA. Goisman, R. M., Warshaw, M. G., & Keller, M. B, (1999). Psychosocial treatment prescriptions for generalized anxiety disorder, panic disorder and social phobia, 1991–1996. American Journal of Psychiatry, 156, 1819–1821. Henggeler, S. W., Melton, G. B., Biondino, M. J., Scherer, D. G., & Hanky, J. H. (1997). Multisystemic therapy with violent and chronic juvenile offenders and their families: The role of treatment fidelity in successful dissemination. Journal of Consulting and Clinical Psychology, 65, 821–833. doi:10/1037/0022-006X.65.5.821 Henggeler, S. W., Schoenwald, S. K., Borduin, C. M., Rowland, M. D., & Cunningham, P. B. (1998). Multisystemic treatment of antisocial behavior in children and adolescents. New York, NY: Guilford Press. Higa-McMillan, C. K., Daleiden, E. L., Pestle, S. L., & Mueller, C. W. (2008, November). Evidence-based practice and practice-based evidence: Using local and national data to encourage youth provider behavior change in a public mental health system. In L. D. Osterberg (Chair), Implementation of ESTs in clinical service settings: What do services look like following dissemination efforts? Symposium presented at the annual convention of the Association of Behavioral and Cognitive Therapies. Orlando, FL.

The Dissemination and Implementation of Evidence-Based Psychological Treatments 431 Hofmann, S. G., & Weinberger, J. (2007). The art and science of psychotherapy. New York, NY: Brunner-Routledge. Improving Access to Psychological Therapies. (2008). Improving Access to Psychological Therapies implementation plan: Curriculum for high-intensity therapies workers. Retrieved from http://www.dh.gov.uk/en/Publicationsandstatistics/Publications/ PublicationsPolicyAndGuidance/DH_083350 Institute for Healthcare Improvement. (2003). The Breakthrough Series: IHI’s collaborative model for achieving breakthrough improvement (IHI Innovation Series white paper). Boston, MA: Institute for Healthcare Improvement. Institute of Medicine. (2001). Crossing the quality chasm: A new health system for the 21st Century. Washington D.C.: Author. Kazdin, A. E., & Weisz, J. R. (1998). Identifying and developing empirically supported child and adolescent treatments. Journal of Consulting and Clinical Psychology, 66, 19–36. doi:10.1037/0022-006X.66.1.19 Kazdin, A. E., & Weisz, J. R. (2003). Evidence-based psychotherapies for children and adolescents. New York. NY: Guilford Press. Kröger, C., Schweiger, U., Sipos, V., Arnold, R., Kahl, K. G., Schunert, T., . . . Reinecker, H. (2006). Effectiveness of dialectical behaviour therapy for borderline personality disorder in an inpatient setting. Behaviour Research and Therapy, 44, 1211–1217. doi:10.1016/j. brat.2005.08.012 Leape, I. L., Kabcenell, A. L., Gandhi, T. K., Carver, P., Nolan, T. W., & Berwick, D. M. (2000). Reducing adverse drug events: Lessons from a Breakthrough Series collaborative. Joint Commission Journal on Quality Improvement, 26, 321–331. Linehan, M. M. (1993). Cognitive behavioral treatment of borderline personality disorder. New York, NY: Guilford Press. Linehan, M. M., Comtois, K. A., & Murray, A. M. (2006). Two-year randomized controlled trial and follow-up of dialectical behavior therapy vs. therapy by experts for suicidal behaviors and borderline personality disorder. Archives of General Psychiatry, 63, 757–766. Linehan, M. M., Manning, S., & Ward-Ciesielski, E. (2008, November). The dialectical behavior therapy intensive training model. In R. K. McHugh (Chair), Dissemination of empirically-supported treatments: The current state of the art. Symposium conducted at the meeting of the Association for Behavioral and Cognitive Therapies, Orlando, FL. Markiewicz, J., Ebert, L., Ling, D., Amaya-Jackson, L., & Kisiel, C. (2006). Learning collaborative toolkit. Los Angeles, CA, and Durham, NC: National Center for Child Traumatic Stress. McHugh, R. K., & Barlow D. H. (Eds). (in press). The dissemination of evidence based psychological treatments. New York, NY: Oxford University Press. McHugh, R. K., Murray, H. W., & Barlow, D. H. (2009). Balancing fidelity and adaptation in the dissemination of empirically supported treatments. The promise of transdiagnostic interventions. Behaviour Research and Therapy, 47, 946–953. doi:10.1016/j. brat.2009.07.005 Miller, W. R., Yahne, C. E., Movers, T. B. Martinez, J., & Pirritano, M. (2004). A randomized trial of methods to help clinicians learn motivation interviewing. Journal of Consulting and Clinical Psychology, 72, 1050–1062. doi:10.1037/0022-006X.72.6.1050 Nathan, P. E., & Gorman, J. M. (Eds.). (2007). A guide to treatments that work (3rd ed.). New York, NY: Oxford University Press. National Center for Child Traumatic Stress. (2007, March). Learning collaborative information packet. Retrieved from http://www.netsnet.org/netsn_assets/pdfs/ lc_info_packet_3–22–07.pdf

432 R. Kathryn McHugh and David H. Barlow Oxman, A.D., Thomson, M. A., Davis, D. A., & Haynes, R. B. (1995). No magic bullets: A systematic review of 102 trials of interventions to improve professional practice. Canadian Medical Association Journal, 152, 1423–1431. President’s New Freedom Commission on Mental Health. (2004). Report of the President’s New Freedom Commission on Mental Health. Retrieved from http://www.mentalhealthcommission.gov/reports/FinalReport/toc.html Resick, P. A., Foa, E. B., Ruzek, J. I., & Karlin, B. (2008, November). Disseminating cognitive processing therapy in VA: The advantages and challenges of a national training initiative. Paper presented at the meeting of the International Society for Traumatic Stress Studies, Chicago, IL. Resick, P. A., Monson, C. M., & Chard, K. M. (2007). Cognitive processing therapy: Veteran/ military version. Washington D.C.: Department of Veterans Affairs. Rogers, E. M. (2003). Diffusion of innovations (5th ed). New York, NY: Free Press. Roth, A. D., & Pilling, S. (2007). The competences required to deliver effective cognitive and behavioural therapy for people with depression and with anxiety disorders. Retrieved from http://www.dh.gov.uk/en/Pubiicationsandstatistics/Publications/ PublicationsPolicyAndGuidance/DH_078537 Schoenwald, S. K. (2008). Toward evidence-based transport of evidence-based treatments: MST as an example. Journal of Child and Adolescent Substance Abuse Treatment, 17, 69–91, doi:10.l080/15470650802071671 Schoenwald, S. K., Heiblum, N., Saldana, L., & Henggeler, S. W. (2008). The international implementation of multisystemic therapy. Evaluation & the Health Professions, 31, 211–225. doi:10.1177/0163278708315925 Schoenwald, S. K., Sheidow, A.J., Letourneau, E. J., & Liao, J. G. (2003). Transportability of multisystemic therapy: Evidence for multi-level influences. Mental Health Services Research, 5, 223–239. doi:10.1023/A:102622910215I Stewart, R. E., & Chambless, D. L. (2007), Does psychotherapy research inform treatment decision in private practice? Journal of Clinical Psychology, 63, 267–281. doi:10.1002/ jclp.20347 Tolman, R. T., Mueller, C. W., Daleiden, E. L., & Stumpf, R. E. (2007). Multisystemic therapy: Treatment outcome validity and predictors of outcome January 2004 December 2005. Retrieved from http://hawaii.gov/health/mentalhealth/camhd/library/pdf/rpteval/ge/ ge/ge022.pdf U.S. Public Health Service, U.S. Department of Health and Human Services, Office of the Surgeon General. (1999). Mental health: A report of the Surgeon General. Rockville, MD: Author. Retrieved from http://www.surgeongeneral.gov/library/mentalhealth/ Weissman, M. M., Verdeli, H., Gameroff, M. J., Bledsoe, S. E., Beits, K., Mufson L., . . . Wickramaratne, P. (2006). National survey of psychotherapy training in psychiatry, psychology, and social work. Archives of General Psychiatry, 63, 925–934.

Article 20

Direct-to-Consumer Marketing of Evidence-Based Psychological Interventions: Introduction Lauren C . Santucci, R. Kathryn McHugh, and David H. Barlow Boston University

The dissemination and implementation of evidence-based psychological interventions (EBPIs) to service provision settings has been a major challenge. Most efforts to disseminate and implement EBPIs have focused on clinicians and clinical systems as the consumers of these treatments and thus have targeted efforts to these groups. An alternative, complementary approach to achieve more widespread utilization of EBPIs is to disseminate directly to patients themselves. The aim of this special section is to explore several direct-to-consumer (i.e., patient) dissemination and education efforts currently underway. This manuscript highlights the rationale for direct-to-patient dissemination strategies as well as the application of marketing science to dissemination efforts. Achieving greater access to EBPIs will require the use of multiple approaches to overcome the many and varied barriers to successful dissemination and implementation. Increasing access to and receipt of evidence-based psychological interventions (EBPIs) remains one of the primary challenges of the field. In recent years, significant resources have been allocated internationally to dissemination and implementation efforts (McHugh & Barlow, 2010). Although there have been early successes of both wide-scale (e.g., Clark et al., 2009; Karlin et al., 2010) and local (e.g., Glisson et al., 2010) efforts, there continues to be an urgent need for effective care. In addition to low rates of receipt of treatment for mental disorders generally (e.g., Collins, Westra, Dozois, & Burns, 2004; Kessler et al., 2005), there is evidence of particularly low rates of receipt of evidence-based care (e.g., Becker, Zayfert, & Anderson, 2004; Goisman, Warshaw, & Keller, 1999; Haas & Clopton, 2003; Mussell et al., 2000; President’s New Freedom Commission, 2004; Santa Ana et al., 2008; Stewart & Chambless, 2007), despite strong support for the efficacy and effectiveness of

434 Lauren C. Santucci, et al.

psychological interventions (e.g., Barlow, 2004; Butler, Chapman, Forman, & Beck, 2006; Weisz, Jenson-Doss, & Hawley, 2006). The barriers to the receipt of EBPIs are many and varied, precluding the use of “one-size-fits-all” strategies for the dissemination and implementation of these interventions. Given the focus on clinician-administered interventions in service delivery, clinical providers have been the relevant consumer group for dissemination and implementation efforts. In effect, clinicians are the individuals who will ultimately adopt the intervention by receiving training and utilizing the intervention with patients. Thus, dissemination approaches have typically targeted clinical providers and clinical systems. An alternative, complementary approach is to target the potential recipients of EBPIs (patients) with dissemination efforts. The purpose of targeting efforts to patients is threefold: (a) to increase awareness of the existence of effective psychosocial treatments; (b) to build “pull demand” for EBPIs; and (c) to increase adoption of minimal contact interventions that could include Web-based interactive programming, other telehealth applications, or traditional workbooks with minimal contact (if any) from a mental health professional. Pull demand refers to a demand for a product or service from one consumer group (e.g., retailers) that subsequently increases demand from another group (e.g., wholesalers). In this case, through targeting dissemination efforts to patients, patients may begin to request EBPIs, thus potentially increasing clinical provider demand for training in and utilization of EBPIs. This approach has been implemented with great success by the pharmaceutical industry through direct-to-patient advertising of medications and “ask your doctor about (medication)” messages. Enormous resources are dedicated by the pharmaceutical industry to direct-topatient marketing of medication, with the aim of building patient requests for the medication. Such pull demand then presumably encourages clinicians to provide these medications to patients when they are requested. In contrast, this approach has not been leveraged well in the area of psychological interventions. Direct-to-patient dissemination strategies might also be used in the context of building demand for minimal contact interventions (for which the patient is the potential adopter). Marketing strategies that identify the benefits and availability of legitimate, evidence-based self-help resources and that provide patients with information including how and where to access these resources may be used to increase demand for these treatments. The reliance on face-to-face (often individual) psychological treatments in mental health care is both costly and may also be limited, as the scope of mental illness is far beyond the workforce capacity to treat it via such methods (Kazdin & Blase, 2011). Achieving greater utilization of self-help interventions increases the number of people who are receiving services for mental illness without any necessary change in workforce. Minimal (or no) contact-formatted interventions are effective for a range of conditions, such as social phobia and binge eating, and particularly for patients with mild or subsyndromal disorders (e.g., Furmark et al., 2009; Striegel-Moore et al., 2010). Additionally, increasing use of and access to

Direct-to-Consumer Marketing of Evidence-Based Psychological Interventions 435

these services may serve as a gateway to other services. Through allowing patients to “try out” therapy, patients have the opportunity to become more familiar with the approach itself while likely experiencing some symptom and functional improvement. This personal experience with therapy and its potential for positive impact may ultimately result in the consumer seeking additional services if minimal contact alone is insufficient, resulting in a de facto stepped-care approach. Research in dissemination suggests that the opportunity to complete a trial of an innovation is positively related to ultimate adoption (Rogers, 2003). Improving access to and understanding of psychological services may serve to increase mental health literacy and decrease stigma, misperceptions, and myths about the nature of these interventions. Moreover, with initial symptom improvement, patients may be more motivated to continue to pursue treatment. The aim of this special section is to highlight current efforts that utilize direct-to-patient approaches for the dissemination and implementation of EBPIs. The manuscripts in the series describe novel approaches for targeting information and interventions to patients and provide suggestions for the application of direct-to-patient strategies in the dissemination and implementation of EBPIs. Below, we provide context for these manuscripts by briefly describing traditional strategies for marketing and explicating these principles relative to the dissemination of EBPIs. Application of Marketing Principles Social marketing—or the application of commercial marketing science to programs designed to influence the behavior of the target audience to improve their personal welfare—has been used to promote behavior change in the context of high blood pressure, heart health, child survival, smoking cessation, HIV prevention, and immunization programs, among other public health efforts (for an overview, see Andreasen, 1995). Social marketing strategies are built on an understanding of the consumer and recognize that behavior change occurs only when a consumer believes it is in his or her best interest to alter his or her behavior. The articles included in this special section emphasize this attention to the consumer perspective. Both Sanders and Kirby (2012) and Metzler and colleagues (2012) highlight the benefits of direct collaboration with consumers—in this case, parents—throughout the development, evaluation, and implementation of population-level parenting interventions. Szymanski (2012) similarly describes attention to the consumer in targeting informational interventions about obsessive-compulsive disorder (OCD) to patients and families. Following from the consumer-centered approach, social marketing efforts seek to segment the market by identifying meaningfully different groups of consumers. When the product is psychological interventions, consumer segments might include children, adults, parents, teachers, or individuals with specific disorders. After learning about each segment’s preferences, values, and perceptions, marketers choose the segment that best fits the current product or, alternatively, design a product to best fit the segment one hopes to serve. A target market is then chosen from these segments, thus attempting to fit the product to the needs/wants of that market.

436 Lauren C. Santucci, et al.

Next, social marketing campaigns consider four elements central to creating behavior change within the target market, often called the Four P’s: product, price, place, and promotion (for more on this framework, see Kotler & Armstrong, 1994). In the case of EBPIs, the product—or the intervention or service being offered—follows naturally from the target market chosen. However, several questions arise when considering EBPIs as the product. First, should marketing efforts focus on specific interventions (e.g., the Triple P Positive Parenting Program; Sanders, 1999) or principles of evidence-based interventions more broadly (e.g., cognitive-behavioral therapy [CBT])? The implications of this decision are large in terms of who the stakeholders are and what potential advocate organizations such as the Association for Behavioral and Cognitive Therapies (ABCT), the American Psychological Association (APA), and the International OCD Foundation (IOCDF) can do in terms of bringing together a unified, direct-to-consumer marketing approach. Second, how do we ensure that the product being delivered is what it claims to be? Should interventions be vetted, such as is done by the National Health Service (NHS) and the Substance Abuse and Mental Health Services Administration (SAMHSA)? If so, how should this vetting occur and by whom? These questions are central to maintaining the integrity of the product. Regarding price, the utilization of EBPIs involves the monetary cost of the service itself as well as the cost of time away from home or work to attend sessions, transportation, or perceived social stigma, among others. Even if a consumer perceives the potential benefits of the intervention, these perceived costs can serve as a barrier to receipt of services. Thus, social marketing campaigns must address perceptions of both the costs and the benefits of the service. Additionally, research is increasingly pointing to CBT as a cost-efficacious alternative to pharmacotherapy, particularly once the treatment phase ends (e.g., Domino et al., 2009; McHugh et al., 2007). Highlighting this body of research through social marketing efforts might promote the use of EBPIs among consumers trying to choose between pharmacotherapy and psychological treatments. The place in which EBPIs are offered is another important consideration. Channel partners, or those who will facilitate the delivery of the product or information about the product to the consumer, should be identified. One approach is to work with organizations that consumers trust or are familiar with, training individuals within those organizations to provide information about or even deliver EBPIs. For example, the Prevention and Relationship Enhancement Program (PREP; Markman, Stanley, & Blumberg, 2001) and the IOCDF have partnered with insurance companies. Through the IOCDF partnership, insurance providers pay for clinicians on their panels to receive training in EBPIs for OCD. Once clinicians are trained, the IOCDF then refers patients to these clinicians. However, even the best programs will fail if they are not readily available to the consumers who want to act. Given the limited availability of clinicians trained to administer EBPIs, expanded models of treatment delivery are needed such that the “place” in which EBPIs are available extends beyond traditional treatment provision settings (Kazdin & Base, 2011). Fortunately, modifications have been made to existing EBPIs so that they are more accessible to consumers, such as computerized CBT

Direct-to-Consumer Marketing of Evidence-Based Psychological Interventions 437

(e.g., Khanna & Kendall, 2010; March, Spence, & Donovan, 2009; Proudfoot et al., 2004), media-based parenting interventions (as described in Sanders & Kirby, 2012; Metzler, Sanders, Rusby, & Crowley, 2012), and CBT delivered through primary care settings (e.g., Craske et al., 2009; Roy-Byrne et al., 2010). The fourth factor, promotion, represents the multiple aspects of marketing communication, including advertising, public relations/publicity, and promotion strategy (e.g., building pull demand), among others. When considering the promotion of EBPIs, one must take into account the limited funds likely available to carry out these efforts. Thus, creative and cost-effective strategies must be employed, such as those that increase consumer access to expert-based information, facilitate partnerships with community liaisons, and utilize media and social media tools. Research into the Triple P Positive Parenting Program (Sanders, 1999), for example, suggests that the media can be effective in its own right as an intervention. Indeed, two studies examining the impact of broadcast television programs on parenting practices found that the viewing public watching someone else participate in treatment on television led to reductions in disruptive child behavior and improvements in parenting confidence (Sanders & Kirby, 2012). Other creative promotional strategies may be utilized, such as the development of a Web television channel that provides the public with a repository of videos capturing therapeutic and/or preventive interventions (M. Sanders, personal communication, November 21, 2009; Santucci & McHugh, 2009). Rather than brand-related promotion, a Web television channel could provide examples of the critical ingredients of various EBPIs including what is involved and what is expected of the consumer. In making available such a resource, consumers have the opportunity to “try out” an EBPI before choosing to enter treatment while simultaneously learning what to look for in a psychological intervention, potentially increasing pull demand for these treatments. Finally, in terms of social media tools—perhaps the most feasible promotion strategy from a financial perspective—Szymanski (2012) describes the potential uses of Facebook, Twitter, and YouTube to disseminate IOCDF event announcements and OCD-related psychoeducational materials. Such resources have the benefit of being both highly compatible with daily lifestyles of patients and relatively low cost. Conclusion Dissemination and implementation efforts to date have focused primarily on clinicians and clinical systems while few efforts have attempted to disseminate information about EBPIs directly to patients. However, disseminating information directly to patients is a promising complementary strategy for improving access to EBPIs. In addition to the potential benefits relative to the adoption of EBPIs, this strategy is highly consistent with patient-centered care approaches (see Berwick, 2009). Berwick argues that the traditional “profession” approach of medicine emphasizes the authority of the clinician and the profession to be the sole assessor of its own quality. As such, the decision-making power has been firmly with the medical professional and not the patient. However, as definitions of evidence-based medicine have shifted to highlight the importance

438 Lauren C. Santucci, et al.

of patient preference (see American Psychological Association, 2005; Institute of Medicine, 2001) the role of patients in advocating for and participating in decisions about their own care becomes more important. As such, marketing of EBPIs becomes an issue not only of access to care, but also of health literacy—providing patients with the information and tools to make informed decisions about their own care. Indeed, a recent meta-analysis suggests that—despite the trends in service utilization for psychotherapy and pharmacotherapy in recent years—patients prefer psychological treatments when given the choice (McHugh, Whitton, Peckham, Welge, & Otto, in preparation). There are numerous scientific and logistical challenges to marketing EBPIs. Of particular importance will be the examination of whether traditional marketing approaches (or alternatives to these approaches) are able to achieve successful pull demand and adoption of self-help interventions. The manuscripts in this special section highlight some early successes; however, this area of study is in its infancy and much work remains to be done. Research addressing the efficacy of marketing procedures will be critical to determining whether this is a viable strategy for improving access to evidence-based care. If early promising results, such as those described in this special section, are replicated and marketing strategies prove to be a feasible and effective approach to increasing access to EBPIs, the challenge remains of how such efforts will be funded. The pharmaceutical industry is a multibillion-dollar industry without a close equivalent in psychological interventions. Thus, financial investment in this strategy will likely require a mix of private and public commitment. Moreover, there are financial implications of marketing approaches. Although the field has taken more of a social marketing perspective to the issue of marketing psychological treatments, financial incentives to providers and systems exist to increase demand for EBPIs. Thus, the ethics of such an approach (e.g., conflict of interest) must also be considered as the field explores this strategy. Dissemination and implementation are an enormous challenge to any discipline that aims to transport innovations to the public. The acknowledgment that traditional passive and low-intensity strategies to increase the utilization of EBPIs are not successful has led to major efforts to identify effective procedures for disseminating and implementing these interventions. Rapid advancement of this important public health agenda will require the utilization of multiple strategies and will benefit from multidisciplinary approaches, such as the application of social marketing and other models of technology transfer. References American Psychological Association (2005). Policy statement on evidence-based practice in psychology. Retrieved June 15, 2011, from http://www.apa.org/practice/resources/ evidence/evidence-based-statement.pdf. Andreasen, A. (1995). Marketing social change. San Francisco, CA: Jossey Bass. Barlow, D. H. (2004). Psychological treatments. American Psychologist, 59, 869–878. doi:10 .1037/0003-066X.59.9.869. Becker, C. B., Zayfert, C., & Anderson, E. (2004). A survey of psychologists’ attitudes towards and utilization of exposure therapy for PTSD. Behaviour Research and Therapy, 42, 277–292. doi:10.1016/S0005–7967(03)00138–4.

Direct-to-Consumer Marketing of Evidence-Based Psychological Interventions 439 Butler, A. C., Chapman, J. E., Forman, E. M., & Beck, A. T. (2006). The empirical status of cognitive-behavioral therapy: A review of meta-analyses. Clinical Psychology Review, 26, 17–31. doi:10.1016/j.cpr.2005.07.003. Clark, D. M., Layard, R., Smithies, R., Richards, D. A., Suckling, R., & Wright, B. (2009). Improving access to psychological therapy: Initial evaluation of two UK demonstration sites. Behaviour Research and Therapy, 47, 910–920. doi:10.1016/j.brat.2009.07.010. Craske, M. G., Roy-Byrne, P. P., Stein, M. B., Sullivan, G., Sherborne, C., & Bystritsky, A. (2009). Treatment for anxiety disorders: Efficacy to effectiveness and implementation. Behavior Research and Therapy, 47, 931–937. doi:10.1016/j.brat.2009.07.012. Domino, M. E., Foster, E. M., Vitiello, B., Kratochvil, C. J., Burns, B. J., Silva, S. G., . . . March, J. S. (2009). Relative cost-effectiveness of treatments for adolescent depression: 36-week results from the TADS randomized trial. Journal of the American Academy of Child and Adolescent Psychiatry, 48, 711–720. doi:10.1097/CHI.0b013e3181a2b319. Furmark, T., Carlbring, P., Hedman, E., Sonnenstein, A., Clevberger, P., Bohman, B., . . . Andersson, G. (2009). Guided and unguided self-help for social anxiety disorder: Randomised controlled trial. British Journal of Psychiatry, 195, 440–447. doi:10.1192/bjp. bp. 108.060996. Glisson, C., Schoenwald, S. K., Hemmelgarn, A., Green, P., Dukes, D., Armstrong, K. S., . . . Chapman, J. E. (2010). Randomized trial of MST and ARC in a two-level evidence-based treatment implementation strategy. Journal of Consulting and Clinical Psychology, 78, 537–550. doi:10.1037/a0019160. Goisman, R. M., Warshaw, M. G., & Keller, M. B. (1999). Psychosocial treatment prescriptions for generalized anxiety disorder, panic disorder and social phobia, 1991–1996. American Journal of Psychiatry, 156, 1819–1821. Haas, H. L., & Clopton, J. R. (2003). Comparing clinical and research treatments for eating disorders. International Journal of Eating Disorders, 33, 412–420. doi:10.1002/eat.10156. Institute of Medicine. (2001). Crossing the quality chasm: A new health system for the 21st century. Washington, DC: National Academy Press. Karlin, B. E., Ruzek, J. I., Chard, K. M., Eftekhari, A., Monson, C. M., Hembree, E. A., . . . Foa, E. B. (2010). Dissemination of evidence-based psychological treatments for posttraumatic stress disorder in the Veterans Health Administration. Journal of Traumatic Stress, 23, 663–673. doi:10.1002/jts.20588. Khanna, M. S., & Kendall, P. C. (2010). Computer-assisted cognitive behavioral therapy for child anxiety: Results of a randomized clinical trial. Journal of Consulting and Clinical Psychology, 78, 737–745. doi:10.1037/a0019055. Kotler, P., & Armstrong, G. (1994). Principles of marketing (6th ed.). Englewood Cliffs, NJ: Prentice Hall. March, S., Spence, S. H., & Donovan, C. L. (2009). The efficacy of an internet-based cognitive-behavioral therapy intervention for child anxiety disorders. Journal of Pediatric Psychology, 34, 474–487. doi:10.1093/jpepsy/jsn099. Markman, H. J., Stanley, S. M., & Blumberg, S. L. (2001). Fighting for your marriage: New and revised. San Francisco, CA: Jossey Bass. McHugh, R. K., & Barlow, D. H. (2010). The dissemination and implementation of evidence-based psychological treatments: A review of current efforts. American Psychologist, 65, 73–84. doi:10.1037/a0018121. McHugh, R. K., Otto, M. W., Barlow, D. H., Gorman, J. M., Shear, M. K., & Woods, S. W. (2007). Cost-efficacy of individual and combined treatments for panic disorder. Journal of Clinical Psychiatry, 68(7), 1038–1044. doi:10.4088/JCP.v68n0710. McHugh, R. K., Whitton, S. W., Peckham, A.D., Welge, J. A., & Otto, M. W. (in preparation). Patient preferences for the treatment of psychological disorders. Manuscript in preparation.

440 Lauren C. Santucci, et al. Metzler, C. W., Sanders, M. R., Rusby, J. C., & Crowley, R. (2012). Using consumer preference information to increase the reach and impact of media-based parenting interventions in a public health approach to parenting support. Behavior Therapy, 43, 257–270. Mussell, M. P., Crosby, R. D., Crow, S. J., Knopke, A. J., Peterson, C. B., Wonderlich, S. A., & Mitchell, J. E. (2000). Utilization of empirically supported psychotherapy treatments for individuals with eating disorders: A survey of psychologists. International Journal of Eating Disorders, 27, 230–237. doi:10.1002/(SICI)1098–108X(200003)27: 23.0.CO;2–0. President’s New Freedom Commission. (2004). Report of the President’s New Freedom Commission on Mental Health. Retrieved May 7, 2008, from http://www.mentalhealthcommission.gov/reports/FinalReport/toc.html. Proudfoot, J., Ryden, C., Everitt, B., Shapiro, D. A., Goldberg, D., Mann, A., . . . Gray, J. A. (2004). Clinical efficacy of computerized cognitive-behavioural therapy for anxiety and depression in primary care: Randomized controlled trial. British Journal of Psychiatry, 185, 46–54. doi:10.1192/bjp. 185.1.46. Roy-Byrne, P., Craske, M. G., Sullivan, G., Rose, R. D., Edlund, M. J., Lang, A. J., . . . Stein, M. B. (2010). Delivery of evidence-based treatment for multiple anxiety disorders in primary care: A randomized controlled trial. Journal of the American Medical Association, 303, 1921–1928. doi:10.1001/jama.2010.608. Sanders, M. R. (1999). Triple-P-Positive Parenting Program: Towards an empirically validated multilevel parenting and family support strategy for the prevention of behavior and emotional problems in children. Clinical Child and Family Psychology Review, 2, 71–90. Sanders, M. R., & Kirby, J. M. (2012). Consumer engagement and the development, evaluation, and dissemination of evidence-based parenting programs. Behavior Therapy, 43, 236–250. Santa Ana, E. J., Martino, S., Ball, S. A., Nich, C., Frankforter, T. L., & Carroll, K. M. (2008). What is usual about “treatment-as-usual”? Data from two multisite effectiveness trials. Journal of Substance Abuse Treatment, 35, 369–379. Santucci, L. C., & McHugh, R. K. (2009, November). Taking CBT, DTC: Direct-to-consumer marketing of evidence-based interventions. Panel discussion conducted at the 43rd annual convention of the Association for Behavioral and Cognitive Therapies, New York, NY. Panelists: David H. Barlow, Matthew Sanders, Howard Markman, & Jeff Szymanski. Stewart, R. E., & Chambless, D. L. (2007). Does psychotherapy research inform treatment decision in private practice? Journal of Clinical Psychology, 63, 267–281. doi:10.1002/ jclp. 20347. Striegel-Moore, R. H., Wilson, G. T., DeBar, L., Perrin, N., Lynch, F., Rosselli, F., . . . Kraemer, H. C. (2010). Cognitive behavioral guided self-help for the treatment of recurrent binge eating. Journal of Consulting and Clinical Psychology, 78, 312–321. doi:10.1037/ a0018915. Szymanski, J. (2012). Using direct-to-consumer marketing strategies with obsessivecompulsive disorder in the nonprofit sector. Behavior Therapy, 43, 251–256. Weisz, J. R., Jenson-Doss, A., & Hawley, K. M. (2006). Evidence-based youth psychotherapies versus usual clinical care. American Psychologist, 61, 671–689. doi:10.1037/ 0003–066X.61.7.671.

Article 21

Evidence-Based Psychological Treatments: An Update and a Way Forward David H. Barlow, Jacqueline R. Bullis, Jonathan S. Comer, and Amantia A. Ametaj Center for Anxiety and Related Disorders, Boston University

Enormous progress in the field of clinical science has been made over the past 50 years, with advances in our understanding of psychopathology and more sophisticated research methodology leading to the development of more efficacious psychological treatments for a variety of behavioral disorders. Despite these advances, the public health impact of well-established psychological treatments is less than it should be. After an overview of the current status of the field, we identify barriers that must be overcome to maximize the public health impact and propose that to breach these barriers we must (a) augment the efficacy of treatments, (b) broaden the impact of treatments across diagnoses to include temperamental variables, (c) attend more closely to mechanisms of action of treatments, and (d) learn the best methods for disseminating and implementing psychological interventions. We conclude by proposing new directions in both research and clinical practice to accomplish these goals. Introduction The notion that psychological practice can actually be influenced by science is a relatively new phenomenon in psychology, and indeed in all of the mental health professions, with origins in the 1960s and 1970s. Since one of us (D.H.B.) began his career in that era, it is revealing, to say the least, to reflect back on the state of clinical science at that time. The Past: How Far Have We Come? Hans Eysenck’s notorious article on the lack of effects from psychotherapy, first published in 1952 (Eysenck 1952) but reprinted more prominently in later years (e.g., Eysenck 1965), had roiled the largely psychoanalytic establishment. His findings, based on crude actuarial tables from the records of insurance companies of the day, comprise, perhaps, the first primitive quantitative review or “meta-analysis,” and although the science underlying his

442 David H. Barlow, et al.

conclusions on the relative ineffectiveness of psychotherapy was weak indeed, there were no objective findings to offer in refutation. Thus, the most usual response to Eysenck’s assertions at the time was anecdote, as notably represented by a quote from Hans Strupp (who was later himself to become one of the pioneers of psychotherapy research): “Clinical observations amply document that many patients benefit from an interpersonal relationship with a professional person when they are troubled by difficulties in living and are seeking help. To argue otherwise is simply to close one’s eyes to the facts” (1964, p. 101). Nevertheless, there was a growing belief among many, including Strupp, that we should be striving to move beyond “clinical observation” and demonstrate the effects of psychological interventions through the scientific method, but the fact was that nobody had a good idea of how to do it. The few clinical trials conducted in those years tended to be extraordinarily ambitious, with numbers of patients running well into the hundreds. Perhaps the best known of these studies is one of the first: the Cambridge Somerville Youth Study (Powers & Witmer 1951), which was designed to explore the effects of a psychosocial intervention on what might today be called conduct problems in adolescents or, at the very least, adolescents at risk for conduct problems. In this study, 650 boys were randomized to either active treatment or treatment as usual. The 10 therapists in the active treatment condition had no formal training and were told to do “whatever you think is best” for five sessions a year for up to five years. Typically, therapists focused on arranging physical exams, organizing a stint in a summer camp, or placing the boys in special education, along with a bit of counseling. The boys were not characterized in any meaningful way, nor were any measurements systematically collected, but crude outcomes, such as contact with the law or other kinds of difficulties encountered in subsequent years, revealed, not surprisingly, no differences between groups at the end of the five-year trial. Despite the initial findings, the study continued for another 30 years, replicating the finding of no differences between groups at 10, 20, and 30 years after the interventions had taken place. This study and others like it, such as a large naturalistic study from the Menninger Clinic that yielded few, if any, meaningful findings (Kernberg 1973), led to considerable despair among the psychotherapy researchers of the day. Indeed, one of the most sophisticated psychotherapy researchers of that era, Carl Rogers, well known for his early work evaluating psychological treatments for schizophrenia, advocated abandoning formal research in psychotherapy altogether in 1969, since in his view it was yielding nothing of value and had no impact on practice (Bergin & Strupp 1972). But another pioneer in our field, Gordon Paul, suggested that the question, is psychotherapy effective? was the wrong question to ask in the first place, since any test of a global treatment, such as psychotherapy, was bound to fail. He urged that clinical researchers begin defining the independent variable (therapy) more precisely and ask the question, what specific treatment is effective with a specific type of client under what circumstances? (1967, p. 112). Following Paul’s guidelines, the early work of pioneers such

Evidence-Based Psychological 443

as Joseph Wolpe and Isaac Marks, but also Allen Bergin and Hans Strupp, changed the landscape of research on psychological treatments. This was due to the promise, if not the realization at that time, of translating research from basic psychological and behavioral science to the applied arena in the service of developing and evaluating more effective interventions (Barlow 2011a, Hersen & Barlow 1976). Also during the 1970s, conceptualizations of psychopathology became more empirical and specific, facilitating the development of reliable and valid dependent variables, and both behavioral and psychodynamic treatments were described in detail, paving the way for more systematic and objectively defined independent variables (Barlow 2011a). The Present: Where Are We Now? Now several generations have passed, and those of us who were trained in that era are approaching the end of our careers in very different circumstances. At this time, governments around the world and their health care systems, faced with demonstrably inadequate health care and spiraling costs, have decided that the quality of health care should improve, that it should be evidence based, and that it is in the public’s interest that this happen (Barlow 1996, 2004; McHugh & Barlow 2012). In no area of health care has this development produced more radical change than among the mental health professions, with psychology most often leading the way in the development and evaluation of evidence-based psychological treatments. The remainder of this review provides a brief overview of how we have reached this point, describes the current status of our applied science, and identifies the barriers we must overcome if we are to continue to progress during the coming years. Advances in the Understanding of Psychopathology

Many of the earliest psychological treatments ultimately showed only limited efficacy in the clinic. Included in this group are early treatments for anxiety disorders, particularly phobias, such as Wolpe’s systematic desensitization, and even early in vivo exposure-based procedures (Barlow 1988). But with a new emphasis on the scientific process, investigators began to find out why, and the process of discovering the answers is illustrative of how new knowledge of the nature of psychopathology can influence and sharpen treatment development across other classes of disorders. For example, in the 1980s, the centrality of panic attacks to many anxiety disorders was discovered and, in particular, the realization that internal cues were just as important, if not more important, in triggering fear and anxiety than situational cues due to the process of interoceptive conditioning (Barlow 1988). Traditional conditioning theories posited that when an individual experienced a panic attack in a particular situation or place, he or she would learn that this situation was a trigger for panic. In essence, any unconditioned stimulus (UCS) paired with a panic attack can become a conditioned stimulus (CS) that signals the imminent onset of the panic response.

444 David H. Barlow, et al.

More modern conditioning research (Bouton et al. 2001) includes a focus on interoceptive conditioning, which suggests that lower levels of autonomic arousal can serve as conditioned stimuli for higher levels of arousal (Barlow 1988). Individuals can learn to associate one interoceptive experience with another interoceptive experience, which is precisely what occurs during a panic attack; small, subtle changes in heart rate or body temperature become associated with the onset of a full-blown panic attack. The pairing of interoceptive events has been demonstrated using drug stimuli: After undergoing conditioning trials that paired a pentobarbital injection with a subsequent morphine injection, rats demonstrated better tolerance of the UCS (morphine) when it was preceded by the CS (pentobarbital) than when the UCS was presented alone (Bouton et al. 2001). The realization that panic attacks could be triggered by internal cues was an enormous advancement in our understanding of the pathology underlying panic disorder, an understanding that soon contributed to more effective treatments (Barlow 1988). We also learned that avoidant behavior that prevented the full processing of fear and anxiety cues extended beyond gross situational avoidance to include subtle behavioral avoidance (e.g., avoidance of eye contact, avoidance of caffeine or alcohol), cognitive avoidance (e.g., distraction, worry, rumination), and reliance on safety signals (e.g., presence of a companion or “good luck” charms). There is now a substantial body of empirical research demonstrating that efforts to avoid (or suppress) thoughts, emotions, or physiological responses actually result in increased physiological arousal, greater autonomic instability, and more stress-related symptoms, despite the desire to down-regulate arousal (Campbell-Sills et al. 2006, Gross 1998). Studies explicitly evaluating the relationship between this emotional or “experiential” avoidance (i.e., any action to prevent full emotional experience and arousal) and psychopathology suggest that experiential avoidance increases the likelihood of substance use relapse, mediates the effect of trauma, and interacts negatively with selfregulatory strategies to result in greater psychological distress (e.g., Chawla & Ostafin 2007). The construct of experiential avoidance is actually recognized, either explicitly or implicitly, across all theoretical orientations (Blackledge & Hayes 2001, Chawla & Ostafin 2007). In our own newly developed Unified Protocol for Transdiagnostic Treatment of Emotional Disorders described below (Barlow et al. 2011), the identification and prevention of emotional avoidance is one of five core treatment strategies. Finally, there was a realization that the methods of exposure-based procedures had failed to keep up with the advances in the basic science of fear learning and extinction. This was largely due to, it turns out, a misguided focus on fear reduction within exposure-based sessions, since neither amount of fear reduction during exposure sessions nor level of fear at the end of the session predicts therapeutic outcome. This is because extinction does not occur through unlearning the CS-UCS association as previously thought. Thus, instead of requiring patients to remain in a fear-producing situation until the fear had diminished, scientists discovered that the creation of new memories associated with both external and internal cues, if sufficiently established, could override existing fear responses without necessarily eliminating them

Evidence-Based Psychological 445

(Bouton et al. 2001, Craske et al. 2008). These scientific advances and others like them have led to the refinement of existing treatments and the development of new treatments for anxiety disorders. Evidence-Based Psychological Treatments

With similar progress in other areas (e.g., Barlow 1996, 2004, 2008, 2011b), it became clear to both practitioners and health care policy makers that robust evidence-based psychological treatments exist for a variety of disorders and problems, and these treatments should be disseminated to those who could benefit from them. Table 21.1, an update of a table originally published in 2004 (Barlow 2004), presents just a partial listing of some of the disorders for which evidence exists demonstrating clear efficacy of psychological treatments compared to credible alternative treatments. It is important to note that this table presents a sampling of studies first published in two of the world’s leading health care journals, the New England Journal of Medicine and the Journal of the American Medical Association. In North America, and also to some extent around the world, health care policy is derived from evidence on health care practices first appearing on the pages of these two journals. Table 21.1 The efficacy of selected specific psychological treatments versus medication or Alternative Treatments Published in the Journal of the American Medical Association or the New England Journal of Medicine.

Disorder

Results

Reference

Stress incontinence in the elderly/women Insomnia

PT > meds + control at acute and follow-up PT > meds or placebo at acute + follow-up PT > routine medical care

Burgio et al. 1998, Goode et al. 2003 Morin et al. 1999, Sivertsen et al. 2006 Teri et al. 2003

PT > usual care or alternative treatments at follow-up (modest effects) PT alone = meds alone; PT + meds > than either alone at follow-up PT = meds at acute—both > placebo PT > meds (or PT + meds) at follow-up PT > present-centered psychotherapy PT > supportive therapy and education (effect size = 0.68)

Donta et al. 2003

Depression and physical health in Alzheimer’s patients Gulf War veterans’ illnesses

Depression Panic disorder

PTSD Tourette’s disorder

Keller et al. 2000 Barlow et al. 2000

Schnurr et al. 2007 Piacentini et al. 2010

Abbreviations: PT, psychological treatments; PTSD, posttraumatic stress disorder; meds, medication.

446 David H. Barlow, et al.

The table presents a mix of studies, with some targeting very specific conditions often found in health care settings, such as stress incontinence, and others targeting more common psychological disorders, such as anxiety and depressive disorders. Some of the specific problems, such as treating depression in patients with Alzheimer’s disease, may have a profound impact. For example, Teri et al. (2003) demonstrated that psychological interventions for this group not only improved depression, but also delayed institutionalization and improved physical health compared to usual best medical practices. Piacentini et al. (2010) recently demonstrated the efficacy of psychological treatments for Tourette’s disorder compared to more customary treatments involving supportive therapy and education. Both of these disorders are most usually treated by psychologists in dedicated clinics or community settings specializing in these problems, thus underscoring the continuing need for specialty psychological clinics where the expertise is available to focus on these problems (J. S. Corner & D. H. Barlow, manuscript submitted). But most psychologists, practicing increasingly in primary care settings as “generalists,” will focus on the more common problems, and leading evidence for the efficacy of psychological treatments for some of these disorders is also available in Table 21.1. Two examples, insomnia and schizophrenia, are highlighted here because many psychologists and other mental health clinicians, particularly in North America, are still relatively unaware of the robust evidence base for the psychological treatments of these disorders. Insomnia. The case of insomnia is especially interesting since the treatment most familiar to clinicians and the population at large is sleep medication, due to heavy direct-to-consumer advertisement by pharmaceutical companies. But in fact, a meta-analysis of pharmacological treatments for insomnia found that despite receiving approval from the US Food and Drug Administration, the majority of medications prescribed for the treatment of sleep maintenance have not consistently demonstrated significant efficacy (Rosenberg 2006). Table 21.1 describes one of the first studies testing brief psychological treatments compared to these popular sleep medications or placebo at both posttreatment and follow-up (Morin et al. 1999). Results show very dramatically the superiority of a brief cognitive-behavioral treatment compared to medication, which was in turn superior to placebo at treatment termination. More importantly, beginning at the three-month follow-up and continuing to a two-year follow-up, medication did not differ significantly from placebo, whereas the psychological treatment retained its effects. Interestingly, adding medication to cognitive-behavioral treatment did not enhance effectiveness and, if anything, interfered with the effectiveness of the psychological treatment. Since this report, a number of additional studies and reviews have confirmed the efficacy of psychological treatments for insomnia, particularly over the long term (Mitchell et al. 2012, Morin et al. 2009). As a result of these findings, the American Academy of Sleep Medicine, as early as 2006, designated this brief psychological treatment for insomnia as the first-line treatment for people with both primary and secondary chronic insomnia, including chronic hypnotic users (Lambert 2008, Morgenthaler et al. 2006). As of 2008, because of this demonstrated efficacy, the American Academy of Sleep Medicine began requiring sleep centers in the United States seeking accreditations to have these psychological treatments available. But only 136 doctoral-level US sleep

Evidence-Based Psychological 447

specialists were judged to be competent to deliver these services at that time. Problems of dissemination and implementation, which are taken up later in the review, are not limited to specific areas such as insomnia and present a major hurdle to the realization of the potential of psychological treatments in the coming decades. Schizophrenia. Schizophrenia (not represented in Table 21.1) presents another interesting example of evidenced-based psychological approaches. Pfammatter et al. (2006) conducted a meta-analysis of psychological treatments targeting four major components of schizophrenia: deficits in social skills, cognitive deficiencies associated with schizophrenia, the demonstrated toxic influence of expressed emotions in families on relapse, and the effects of psychological treatments, specifically cognitive behavioral therapy, for positive symptoms, including clusters of delusions, hallucinations, and associated phenomena. They found moderate to strong effect sizes of existing psychological treatments for each of these target problems (Pfammatter et al. 2006). Of course, the types of interventions appropriate for remediating cognitive deficits on the one hand, or positive symptoms on the other, are necessarily very different. Illustrating approaches to one of these areas, positive symptoms, Grant et al. (2012) took on the daunting task of treating low-functioning neurocognitively impaired patients with schizophrenia by adding cognitive therapy to a treatment-as-usual (TAU) standard care package that included indicated antipsychotic medications. Patients receiving cognitive therapy in addition to TAU showed a clinically significant mean improvement in global functioning from baseline to 18 months, but particularly interesting were the effects of cognitive therapy on positive symptoms as seen in Figure 21.1 (Grant et al. 2012). Values at 6, 12, and 18 months showed a marked and sustained improvement in positive symptoms from the addition of cognitive therapy compared to TAU alone. In fact, psychological treatments for schizophrenia, which are recommended as first-line treatment in the National Health Service in the United Kingdom, evidence a doubling of therapeutic benefit compared to those patients receiving medication alone (Tarrier 2008). Another study, this time examining the effects of cognitive enhancement procedures targeting cognitive deficits in schizophrenia, revealed an increasingly common outcome from studies of robust psychological treatments when properly measured; that is, demonstrable changes in brain function and even brain structure. In this case, cognitive enhancement therapy protected against frontal and temporal gray matter loss when administered early during the course of schizophrenia compared to a well-construed alternative psychological treatment termed “applied coping.” Cognitive enhancement procedures also improved long-term cognitive outcomes (Eack et al. 2010). Evidenced-based psychological treatments for these and other disorders share two common characteristics. First, they are tailored to specific forms of severe and identifiable psychopathology. This is particularly evident in the example of schizophrenia, where different psychological treatments focus on four distinct and identifiable deficits found to some degree in individuals suffering from this disorder (Addington et al. 2010). Secondly, these procedures are typically derived from psychological science, specifically behavioral and cognitive sciences, with heavy input from social cognition and interpersonal approaches. This is very different from the era of “schools” of psychotherapy

448 David H. Barlow, et al. 25

Mean SAPS Total Score

20

15

10

5

0 Baseline

ST Alone (n = 29) CT with ST (n = 31) 6

12

18*

Assessment (months)

Figure 21.1 The values at baseline are raw means; the values at 6, 12, and 18 months are adjusted means from the intent-to-treat hierarchical linear models. Note: CT, cognitive therapy; SAPS, Scale for the Assessment of Positive Symptoms; ST, standard treatment. *P = 0.04 for the mean difference based on the hierarchical linear modeling interaction of treatment condition × assessment time.

of the 1960s and 1970s, when techniques were typically derived from theoretical conceptions of personality rather than the laboratories of cognitive and behavioral sciences. Because of these unique qualities, one of us (D.H.B.) proposed (Barlow 1996, 2004, 2006) that this heterogeneous group of evidence-based interventions, targeting as it does psychopathology or psychological aspects of pathophysiology in a manner compatible with the objectives of organized health care systems, should be identified as “psychological treatments (or interventions)” to distinguish it from the broad and generic term “psychotherapy,” a valuable undertaking that ranges well beyond problems covered in any health care system, including issues of adjustment and growth. These two activities would not be distinguished in theory, technique, or even evidence, but on the problems addressed, which would serve to increase the marketability and acceptability of these approaches to health care policymakers. This terminology has now been widely adopted in, for example, the National Health Service in the United Kingdom and the Veterans Affairs Health Care System in the United States (Clark 2012, Ruzek et al. 2012). Reasons for Success. There are several reasons for the success of these interventions. First, as alluded to above, we have a greater understanding of the nature of psychopathology and pathophysiology, resulting in new, more precisely targeted treatments. Examples are evident once again in schizophrenia

Evidence-Based Psychological 449

as described above, but also in more common disorders, such as panic disorder, where discoveries of internal interoceptive triggers for fear responses (Bouton et al. 2001) facilitated the development of interoceptive exposures that directly target symptoms (Barlow 1988, Barlow & Craske 2007). Second, clinical research methodologies have improved substantially over the past several decades as illustrated above. Not only have these improvements occurred in design methodologies capable of isolating components of treatment efficacy and ruling out threats to internal validity, including the contribution of common factors and allegiance, but also in more mundane procedures such as data management, which can now be effected through direct web-based data entry from multiple sites in real time. This development makes possible coordinated multisite clinical trials typically more common in medicine, but increasingly seen in the context of mental disorders, with minimal loss of data or inaccurate data entry. In other words, the invisible college of clinical trialists working in the arena of mental disorders has developed reasonable consensus on best research methods and practices. Finally, health care systems and governments around the world, noting this strong evidence, are adopting and promoting evidence-based psychological interventions. The two most notable examples of this initiative, each with expenditures of over one billion dollars thus far, are the aforementioned program in the UK National Health Service titled “Improving Access to Psychological Therapies” (Clark 2012) and the Mental Health Strategic Plan of the US Veterans Affairs Health Care System (Ruzek et al. 2012). Two other developments have attracted attention and facilitated progress. First, psychological treatments have evolved such that the evidence base of specific procedures is more important than the theoretical origins of the procedures in schools of psychotherapy. Hence, although many evidence-based procedures are cognitive-behavioral in origin, procedures and techniques with developing evidence have been derived from myriad different approaches. Perhaps the bestknown example is motivational interviewing, which was originally validated as effective with addictive disorders and was more recently found to contribute significantly to the treatment of anxiety and mood disorders (Arkowitz et al. 2008). And yet this procedure is very clearly derived from client-centered Rogerian therapy. It is likely that schools of psychotherapy will become an anachronism in years to come and theoretical approaches will blur in the face of the development of more finely targeted treatments based on increased knowledge of psychopathology derived from the best evidence available. Second, in every survey taken, consumers prefer psychological treatments to drug treatments by a wide margin (e.g., McHugh et al. 2012). This factor will become increasingly important as dissemination and implementation efforts, currently in the early stage of development, become more widespread (McHugh & Barlow 2012). Relative Lack of Public Health Impact of Psychological Treatments

Despite these developments and outside of highly regulated health care systems such as the Veterans Affairs Health Care System, psychological interventions of

450 David H. Barlow, et al.

any kind, let alone evidence-based, are not penetrating the de facto health care delivery system in North America to the extent that they should be. In fact, recent national service use trends raise great concerns about the quality of mental health care for affected individuals generally. Specifically, in recent years we have seen a decrease in the prominence of psychological treatments in the management of mental health problems (e.g., Marcus & Olfson 2010) and a dramatic increase in the use of unsupported psychotropic interventions with unfavorable side effects. For example, among individuals receiving mental health care, we saw a significant decrease from 1998 to 2007 in the proportion of individuals receiving psychological interventions or psychotherapy, dropping from roughly 1 in 6 treated individuals to 1 in 10 treated individuals, and a simultaneous increase in the proportion of individuals receiving psychotropic medication interventions alone, with rates rising from 44% of treated individuals to almost 60% in 2007 (Marcus & Olfson 2010). There have been striking expansions in the rate of antidepressant medication prescribing (Olfson & Marcus 2009) and the prescription of off-label antipsychotic medication (Comer et al. 2011). The growth in off-label antipsychotic prescribing in outpatient mental health care is of particular concern, given the associated metabolic, endocrine, and cerebrovascular risks that have been well documented (Olfson et al. 2006). Since the introduction of second-generation antipsychotic medications in the early 1990s, these powerful medications have become increasingly common in the outpatient management of diverse clinical populations. Sedative properties associated with antipsychotic medications may help to explain their broadened use in nonpsychotic patients. Of central relevance to the field of clinical psychology, these off-label prescribing regimens are being used to treat mental health conditions for which well-established psychological treatments with empirical support exist. For example, antipsychotic prescriptions by psychiatrists for outpatient anxiety disorders increased from 10.6% to 21.3% of anxiety disorder visits (Comer et al. 2011). The largest increases in antipsychotic prescribing for anxiety disorders were among new outpatients, indicating that psychiatrists appear increasingly comfortable prescribing antipsychotic medications for anxious patients before initiating trials of other medication classes or adjusting current medications. Among the anxiety disorders, the largest increase in antipsychotic prescribing has been among visits for panic disorder. Of great concern, there have been no controlled trials evaluating the efficacy of antipsychotic treatments for panic disorder, whereas the psychological treatment of panic disorder is well established (see Craske & Barlow 2008). Such findings underscore the tremendous need for effective dissemination and implementation of psychological treatments. Barriers to Evidence-Based Psychological Treatments and a Way Forward Why have psychological treatments not been more broadly adopted in the de facto mental health delivery system in the United States? Some of the reasons for this lack of penetration are deeply imbedded in the structure of the health care delivery system in North America and elsewhere, and these policies

Evidence-Based Psychological 451

are continually undergoing scrutiny at various levels. But there are a number of barriers to the availability of psychological treatments that are more directly amenable to change by the community of mental health researchers and practitioners. These barriers are: first, the relative lack of efficacy of psychological treatments for a substantial minority of the population who are administered the treatments. Second, it is clear that most patients present with extensive comorbidity among psychological disorders and sometimes physical disorders as well. But psychological treatments administered are typically directed at just one problem or disorder at a time. Third, the very nature of our research enterprise as it has evolved utilizes a nomothetic approach that produces results reflecting the average response of large groups and ignores to some extent intersubject variability or the response of the individual patient. And yet the object of health care delivery and mental health care delivery is the well-being of the individual. The absence of a more idiographic approach to our research limits to some degree the applicability of our research findings. Concomitantly, most efforts using randomized controlled trials to establish efficacy have not mounted a parallel effort to examine purported mechanisms of action through mediation analyses or the functional analytic strategies of single-case experimental designs (Barlow 2010, Kazdin 2011, Nock 2007). But discovering active mechanisms of treatment is one of the surest ways to enhance efficacy. Finally, due in part to the relatively recent establishment of psychological treatments and the lack of concerted marketing efforts that are so much a part of drug development by large pharmaceutical corporations, efforts to disseminate and implement psychological treatments have only just begun. Each of these barriers is reviewed in turn, and it is proposed that better addressing these issues will be necessary as we move forward. Treatment Augmentation

In response to less-than-desirable treatment effects in some cases, researchers have recently turned to translational science in search of alternative methods to augment the effect of psychological treatments. Among many interesting developments, several are highlighted here. First, basic research has revealed that certain pharmacological agents, although not beneficial by themselves, seem to enhance the effects of psychological procedures. One such example within the treatment of anxiety disorders is D-cycloserine (DCS), an antibiotic that was first introduced in 1955 for the treatment of tuberculosis. DCS is a partial agonist for the N-methyl-D-aspartate receptor in the amygdala, which has been shown in animal studies to enhance the consolidation of memories that facilitate fear extinction learning (Hofmann 2007, Walker et al. 2002). In a randomized, double-blind, placebo-controlled trial, patients with panic disorder who received 50-mg doses of DCS prior to a brief evidence-based psychological treatment were significantly more likely to achieve a clinically significant change status than were participants who received a pill placebo (Otto et al. 2010). In a similarly designed study of participants with social anxiety disorder, acute administration of DCS prior to exposure therapy resulted in significantly greater reductions in social anxiety than did administration of a pill placebo (Hofmann et al. 2011). There is also preliminary support

452 David H. Barlow, et al.

for the use of DCS as an augmentation strategy for specific phobia (Ressler et al. 2004) and obsessive-compulsive disorder (Wilhelm et al. 2008). Interestingly, a recent study found that the addition of DCS to exposure therapy for posttraumatic stress disorder in veterans returning from Afghanistan and Iraq resulted in significantly poorer outcomes than the administration of a pill placebo, leading the authors to hypothesize that DCS may enhance reconsolidation of a trauma memory if fear does not sufficiently decrease during exposure (Litz et al. 2012). Thus, the parameters of DCS require further exploration before routine use in clinical practice. Oxytocin, an amino acid neuropeptide produced in the hypothalamus, is another example of a potential cognitive enhancer for exposure-based therapy. In animal research, the administration of oxytocin to rats results in twice the amount of physical contact with one another (Witt et al. 1992). When administered intranasally to humans, oxytocin has been shown to decrease anxiety and promote prosocial or approach behavior (e.g., Heinrichs et al. 2009). In a double-blind, placebo-controlled study, participants with generalized social anxiety disorder demonstrated greater activity in the medial prefrontal cortex extending into the anterior cingulate cortex while viewing an emotionally valenced interpersonal cue (sad faces) compared to a control group. But this activity reduced to levels comparable to the control group following intranasal administration of oxytocin. Interestingly, this effect was present only for negative social cues (sad faces) and not positive (happy) or neutral faces (Labuschagne et al. 2012). In another randomized, double-blind, placebo-controlled study, participants with a principal diagnosis of social anxiety disorder who received intranasal oxytocin as an adjunct to exposure therapy received more favorable appraisals of speech performance and physical appearance from judges (Guastella et al. 2008). However, oxytocin had no effect on broader patterns of symptom reduction in the short term, suggesting that further research is necessary to determine whether modifications to the frequency or amount of oxytocin administration results in more substantial clinical change. In another exciting development, neuroscientists have discovered that fear memories are encoded and consolidated with the synthesis of new proteins and that retrieval of these memories, by means of a cue, provides additional synthesis during a consolidation window (Monfils et al. 2009). In rats this reconsolidation window lasts for approximately three hours, and if extinction procedures are instituted during this window, new memories seem to be formed and the extinction process is more complete, thereby preventing reinstatement or spontaneous recovery of fear. In an important study in humans, Xue et al. (2012) tested extinction procedures among heroin addicts by cuing intense drug cravings through video clips. The investigators then tested the relative effects of extinction occurring approximately 10 minutes after the retrieval cue versus 6 hours versus extinction with no prior retrieval cue at all. Extinction of the heroin urges was substantially greater during the 10-minute window compared to the other conditions, lending some clinical validity to this new finding. If confirmed in additional disorders where exposure-based extinction processes are indicated, then clinical outcomes may well improve. Of course, these developments are just a sampling of efforts to translate findings from basic research to augment treatment outcomes.

Evidence-Based Psychological 453

Comorbidity, Temperament, and a Transdiagnostic Perspective

The fifth edition of the Diagnostic and Statistical Manual of Mental Disorders is now available, and although improvement is evident, many of the same limitations remain, including high diagnostic comorbidity (Brown & Barlow 2009). One possible explanation for such high rates of comorbidity among emotional disorders is the presence of shared etiological pathways and, more importantly, a possible shared underlying diathesis. One such example is triple vulnerability theory (Barlow 1991, 2000, 2002), which emphasizes a common etiological process among a wide range of disorders of emotion. This model focuses on three vulnerabilities—generalized biological and generalized psychological vulnerabilities, as well as a specific psychological vulnerability resulting from early learning that strongly influences the development of one set of specific focal symptoms versus another (e.g., obsessions versus panic attacks). The generalized vulnerabilities are associated with, and may largely comprise, the temperament of neuroticism—an enduring tendency to experience negative emotions (Barlow et al., manuscript submitted). Indeed, there now exists a substantial body of empirical research utilizing quantitative approaches, such as structural equation and latent variable modeling, supporting the contribution of higher-order temperamental constructs, such as neuroticism and the largely overlapping constructs of negative affectivity and behavioral inhibition, to the etiology and presentation of emotional disorders (Brown 2007, Clark 2005). In one study conducted with a large sample of outpatients, higher-order dimensions of positive and negative affect accounted for almost all of the covariance in the latent variables that corresponded to social anxiety disorder, generalized anxiety disorder, obsessive-compulsive disorder, panic disorder with agoraphobia, and unipolar depression (Brown et al. 1998). Although negative affect (neuroticism) contributes to all disorders, the strongest relationships for (low) positive affect are observed with unipolar depression and social anxiety disorder, and more recently agoraphobia (Brown et al. 1998, Rosellini & Brown 2011). There is also evidence to suggest that treatment of the patient’s principal diagnosis results in some treatment gains, at least initially for comorbid mood and anxiety disorders as well (Brown et al. 1995, Newman et al. 2010). Neuroticism also predicts disorder onset and a poorer prognosis (Barlow et al., manuscript submitted; Clark et al. 1994; Griffith et al. 2010). Taken together, the results from research exploring rates of diagnostic comorbidity, response of comorbid diagnoses to single-disorder treatment protocols, evidence of shared etiological pathways, and empirical support for the latent structure of emotional disorders comprise a strong evidence base that highlights the magnitude of shared features among anxiety and mood disorders, perhaps even eclipsing differences, and the importance of focusing on temperament, particularly neuroticism and variations in positive affect (Barlow et al., manuscript submitted). On the basis of this renewed emphasis on temperament as a unifying factor across emotional and related disorders, clinical investigators have begun focusing on the existence of transdiagnostic processes (e.g., avoidance, cognitive biases, worry and rumination, intolerance of uncertainty, sleep disturbance) that cut across diagnostic categories and comprise the core facets

454 David H. Barlow, et al.

of emotional disorders (Harvey et al. 2010, McLaughlin & Nolen-Hoeksema 2011). A treatment protocol designed to target the shared features of mood, anxiety, and related disorders, if successful, may offer a solution to the ubiquitous challenge of comorbidity while simultaneously overcoming some of the barriers to the dissemination and implementation of effective psychological treatments by reducing the number of protocols that are applicable to only a single diagnosis. One of us (D.H.B.) developed The Unified Protocol for Transdiagnostic Treatment of Emotional Disorders (Barlow et al. 2011) with the goal of addressing these challenges by distilling current, empirically supported treatments and incorporating recent advances in emotion regulation, motivational interviewing, mindfulness techniques, and exposure-based procedures (Campbell-Sills & Barlow 2007). Most notably, the Unified Protocol focuses on how an individual responds to intense emotional experiences and aims to modify unproductive strategies for down-regulating or avoiding these experiences, thereby extinguishing anxiety and distress associated with intense emotions. Additional details regarding protocol development can be found elsewhere (Campbell-Sills et al. 2012, Ellard et al. 2010, Wilamowska et al. 2010). In a randomized controlled trial of 37 patients with a principal (most severe) diagnosis of one of the anxiety disorders and accompanying comorbidity, 18 sessions of the Unified Protocol produced significant symptom reduction in depression, anxiety, and functional impairment (Farchione et al. 2012). There is also evidence to suggest that transdiagnostic treatments for anxiety disorders are effective when administered in a group format (Norton 2008) as well as with other families of disorders, such as eating disorders (Fairburn et al. 2009). Currently, our group is conducting an equivalency trial comparing the Unified Protocol to well-established, manualized disorder-specific treatment for each participant’s primary diagnosis, as well as a wait-list control condition, to evaluate whether a transdiagnostic treatment is capable of producing treatment effects on both principal and comorbid diagnoses comparable, or even superior, to our best available single-disorder treatments. Although transdiagnostic psychological treatments are a relatively new area of interest in clinical psychology, there is evidence that pharmacotherapy may also effect changes through transdiagnostic mechanisms. For example, little is known about how antidepressants mediate changes in mood or why these changes take so long to occur. Harmer et al. (2009) hypothesize that the time lag from the administration of an antidepressant to an improvement in mood is actually not due to a delay in the neuropharmacological drug action (e.g., an increase in serotonin that then produces downstream neuroadaptive effects), since there is evidence of changes in emotional processing based on sophisticated cognitive assessment as early as one week into antidepressant therapy. Accordingly, the authors propose that antidepressants likely work through reductions in negative affective biases, which in turn produce more gradual changes over time in mood as patients discover that their view of the world has changed. This suggests that antidepressants and transdiagnostic psychological interventions may, when effective, work for similar reasons, albeit following very different paths to reach this point. This finding also underscores the importance of discovering mechanisms of change in all treatments (taken

Evidence-Based Psychological 455

up in the next section). Regarding clinical utility, if further research confirms the relationship between early changes in emotional processing and subsequent improvements in mood, it may be possible in the future to predict an individual patient’s response to antidepressant therapy within the first week of treatment based on changes in emotional processing. Another example of a transdiagnostic approach is the development of modular treatment protocols. Perhaps the most successful example of modular approaches to date is the Modular Approach to Therapy for Children with Anxiety, Depression, or Conduct Problems (MATCH) (Chorpita & Weisz 2009). This effort was also driven by the poor applicability in children of specific single-problem or diagnosis treatment protocols to the everyday clinic where comorbid presentations are common (Weisz & Gray 2008). The treatment protocol consists of separate and modular treatment elements derived from evidence-based treatments for anxiety, depression, and conduct problems, and it provides clinicians with a decision flowchart for module selection as well as troubleshooting guidelines to address possible treatment difficulties (Chorpita & Weisz 2009). In a recent randomized controlled study comparing modular treatment to standard manual treatment and usual care for youth with emotional and conduct disorders, modular treatment significantly outperformed both treatment comparison conditions on multiple clinical outcome measures (Weisz et al. 2012). Unlike the Unified Protocol for emotional disorders mentioned above, or Fairburn’s transdiagnostic approach to eating disorders, this approach is not a theoretically derived set of principles emerging out of new findings pertaining to the nature of the psychopathology. Rather, MATCH identifies discrete, practical empirically based treatment modules or components and provides guidelines for identifying specific presenting problems and then matching the appropriate module. In any case, further research is necessary to explore the applicability of modular treatments in diverse clinical populations, but adapting current evidence-based treatments to follow a more modular approach may help increase implementation and effectiveness in clinical settings. Mechanisms of Action and Idiographic Approaches

The primary focus of clinical research should be on the development of effective psychological treatments, the identification of the mechanisms of change (i.e., mediators and moderators) that drive treatment response, and ultimately the specification of individual factors that increase and reduce treatment efficacy (Kazdin 2011, Nock 2007). Very few studies have examined the latter two aims for several reasons; one is the complexity of methodology and data analyses required to evaluate mediation of treatment change. It is only within the past 10 to 15 years that treatment mediation has been systemically explored, and existing studies are limited in both scope and number. A second reason is the relative lack of emphasis in extant studies on individual (idiographic) change as opposed to the average response of groups of individuals (nomothetic), and the tendency in these nomothetic studies to measure change infrequently during treatment—usually just at pretreatment and posttreatment.

456 David H. Barlow, et al.

Mechanisms of Change. It is simply not enough to establish efficacy for a particular treatment. Consistent with the theoretical or empirical identification of core strategies comprising transdiagnostic treatments, further research on treatment mechanisms should distill existing treatments into the most parsimonious models possible and allow us to maximize treatment efficacy by discarding what doesn’t work and including more of what does work (Nock 2007). In order to assess effectively for mediation of treatment changes, both clinicians and researchers must include one (preferably more) potential mechanism of change, assess proposed mechanisms and treatment outcomes continuously, and replicate any observed effects in different settings with diverse patient populations (Barlow 2010, Kazdin 2011). Many of these hypothetical mechanisms will emerge from basic research. Others will be identified serendipitously in the clinic and then verified by research. Within the emotional disorders, for example, we described basic research identifying mechanisms underlying the process of extinction and accompanying changes in brain function. We have now learned that this process, which involves the creation of new memories rather than the unlearning of specific fear and anxiety associations, is optimized during a specific window of opportunity following cued retrieval of these memories (Monfils et al. 2009, Xue et al. 2012). This information, once translated to the clinic, should prove to be very useful in augmenting the effectiveness of extinction procedures. Similarly, Harmer and colleagues (2009) have tentatively identified an important process during drug treatment of depression involving alteration of cognitive biases that might prove to be a marker for successful treatment. Confirming this mechanism should contribute substantially to the development of more effective treatments, both drug and psychological. Finally, based on contributions from basic science as well as theoretical advances, a purported mechanism of action of The Unified Protocol for Transdiagnostic Treatment of Emotional Disorders has been identified, subject to further confirmation, best described as extinguishing anxiety and distress associated with intense emotional experiences. Progress in identifying and confirming mechanisms of actions will be one of the most efficient methods for improving treatment efficacy. Are Psychological Treatments Placebos? The Dodo Bird Interpretation. As described above, better understanding of the nature of psychopathology and other targets of treatment has allowed the development of very focused interventions designed to remediate specific deficits or pathologies as diverse as cognitive deficits in schizophrenia or severe tics in Tourette’s disorder. Nevertheless, an argument has been raised that none of these studies actually demonstrate that any of these specific treatments are any better than a credible comparison condition and that any results showing efficacy are really a function of factors that are common to all treatments. In this conceptualization then, psychological treatments could be considered placebos in the sense that the therapeutic mechanism of action would not be in the specificity of the treatment for the condition targeted but rather in the more general credibility of the treatment in the eyes of both patients and therapists, and the aforementioned skill with which the therapist develops the alliance, or in other therapist factors (e.g., Wampold 2001).

Evidence-Based Psychological 457

This debate has been ongoing for decades, but the “common factor” argument has not gained traction with health care policymakers or other independent bodies around the world tasked with creating clinical practice guidelines or best practice algorithms for the treatment of psychological disorders. For example, the National Institute for Health and Clinical Excellence in the United Kingdom, charged with deciding on treatment approaches with sufficient evidence to be included in the National Health Service, and the Agency for Healthcare Research and Quality within the US Department of Human Services, as well as the Veterans Affairs Health Care System, which provides a similar function in the United States, all using sophisticated analytical methods and examining hundreds of studies, have produced detailed clinical practice guidelines identifying specific psychological interventions as recommended first-line treatments that differ from disorder to disorder. This view of the evidence also meets the test of face validity with clinicians, since few clinicians believe that you could use exactly the same procedure to treat chronic schizophrenia, specific phobia, bipolar disorder, or Tourette’s disorder, as long as the treatment is credible. To hold this view would imply that client-centered therapy, or even less highly regarded treatments such as past-life regression therapy or thought-field therapy, would work as well for any of these disorders as would the treatments specifically designed for these disorders described above. Proponents of the common factor account readily dismiss the hundreds of studies (e.g., Nathan & Gorman 2007, Weisz & Kazdin 2010) that contribute to clinical practice guidelines for one reason or another, but the most common reason is that comparison procedures are not as credible as the psychological treatments under evaluation, and furthermore that therapists evaluating these procedures often have allegiance to these treatments or have some reason to believe that they might be superior and that this influences the results. Of course, evaluation of psychological treatments can never be double blind, but recent advances in methodology as represented by the Consolidated Standards of Reporting Trials Statement outline more rigorous standards of research (Moher et al. 2001). These standards include the use of independent evaluators, adherence ratings of therapists delivering treatments, and assessment of credibility and acceptability of all treatments under investigation, among other procedures, and they have substantially mitigated threats of allegiance to internal validity (i.e., conclusions on efficacy) and make it highly unlikely that every single study would be somehow fatally flawed. In addition, at present there is no direct causal evidence supporting the effects of allegiance, although it is not an unreasonable hypothesis, and there is some post hoc correlational evidence (Leykin & DeRubeis 2009). Nevertheless, multisite studies involving sites with different allegiances designed in part to control for allegiance effects, both between medications and alternative psychological treatments, have not yielded site-by-treatment interactions that would be expected if allegiance accounted for most of the effects (Leykin & DeRubeis 2009). For example, Stangier et al. (2011) evaluated cognitive therapy and interpersonal psychotherapy as treatments for social anxiety in a multisite study comparing both to a wait-list control condition. One site had clear allegiance to cognitive therapy whereas the other was a

458 David H. Barlow, et al.

center for the study of interpersonal psychotherapy. Both treatments were efficacious compared to the wait-list control group, effects that were maintained one year after treatment, but cognitive therapy was more efficacious than interpersonal psychotherapy at both sites, reflecting the fact that there were no significant treatment-by-site interactions. Other prospective studies exist that run counter to an allegiance hypothesis. One good example is found in Shear et al.’s (2001) evaluation of a new emotion-focused treatment for panic disorder based roughly on psychodynamic concepts. The developer of this procedure, Dr. Shear, had a strong allegiance to it, had written a treatment manual she was ready to submit for publication, and had found strong evidence for efficacy compared to a waitlist condition in an initial study. But when properly tested in a prospective randomized controlled trial against a well-established psychological intervention as well as an active drug and a drug placebo condition, this treatment evidenced significantly less efficacy, roughly comparable to the drug placebo condition, at which point it was abandoned. Of course placebo effects do contribute to psychological treatments. But one interesting fact is that different disorders representing various patterns of psychopathology respond very differently to placebo interventions, whether drug or psychological, as we have demonstrated elsewhere (Huppert et al. 2004). Thus, depression with its cyclical pattern of exacerbation and remission is notoriously responsive to placebo medications in the short term, much to the chagrin of pharmaceutical companies who have difficulty demonstrating the efficacy of new antidepressant drugs compared to placebo medications. Panic disorder is also placebo responsive in the short term given its cyclical nature. In these disorders, and with this specific psychopathology, many treatments will work well initially, but the question becomes, which treatments work well in the long run and with more severe patients? In contrast, obsessive-compulsive disorder is not placebo responsive, with a placebo effect of close to zero. We have shown elsewhere that in this disorder, the contribution of specific techniques vis-à-vis other factors, such as therapist effects, is much higher, accounting for up to 60% of the variance (Huppert et al. 2007). Along these lines, more recent and sophisticated analyses examining the issue more directly have confirmed that nonspecific therapies for depression may well be efficacious at least in the short term, although less so than powerful and specific evidence-based interventions, but only in the treatment of less severe depression (Cuijpers et al. 2012). When depression is more severe, the effects of specific evidence-based interventions become stronger (Driessen et al. 2010). Furthermore, specific psychological treatments evidence an enduring effect when compared with patients treated by credible alternatives, including effective medications, and this enduring effect is even larger in magnitude than is keeping patients on continuing medication (Cuijpers et al. manuscript submitted; Hollon 2011). Thus, a possible general conclusion is that many psychological approaches, even nonspecific ones, may be effective with less severe forms of some disorders that are particularly placebo responsive, such as depression, but this would not be the case with different disorders or even more severe forms of placebo-responsive disorders (Hollon 2011, Huppert et al. 2004). In any case, to demonstrate the “dodo bird” thesis, proponents would need to do the hard work of conducting prospective randomized trials comparing

Evidence-Based Psychological 459

newly constructed “bona fide” treatments to well-established evidence-based interventions for specific disorders using data analytic procedures more appropriate to the task, such as equivalence or noninferiority analysis. It is not enough to rely on post hoc correlational reanalyses of groups of studies using meta-analytic procedures, nor is it enough or even correct to assume that studies showing no difference between specific procedures (i.e., the null hypothesis) implies equivalence. Indeed, proving the null hypothesis signifies only a failed clinical trial. On the other hand common factors do clearly contribute to the outcomes of psychological treatments even for severe patients, thus constituting one mechanism of action. We now have new evidence from a large multisite study on the treatment of panic disorder indicating that the therapeutic alliance as perceived by the patient (but not the therapist) contributes to outcome in terms of symptom reduction, both early and late in the treatment process (Huppert et al., manuscript submitted). Of course, it is certainly intuitive that patients would be more likely to follow through with the procedures assigned in this therapy, some of them producing temporary distress, in the context of a trusting relationship. Nevertheless, this finding and many others like it do indicate the importance of monitoring the therapeutic alliance during treatment. But the same study, using more sophisticated multilevel analysis, found no evidence of differential therapist effects across 23 therapists of different theoretical orientations treating 258 patients. This was in contrast to an earlier study from our group (Huppert et al. 2001), using less appropriate analytic procedures, where it was found that approximately 10% of the variance in efficacy was accounted for by variability in therapist effects. Of course, this simply indicates that training these therapists to a gold standard of competence was highly successful. This would, of course, be unlikely in more naturalistic settings where therapists would differ more widely in the delivery of evidencebased treatments. All of this underscores the need for continuous study of mechanisms of action that look at the interaction of therapists, patients, and contextual factors with the specific actions of robust evidence-based psychological treatments. Idiographic Approaches to Treatment Development and Evaluation. At present, clinical science focused on interventions is largely based on inferences made about the relations between variables through the comparison of averages between groups of individuals. This approach, also known as the nomothetic approach, has become the gold standard for establishing causal relations between independent and dependent variables since the 1980s, when the National Institutes of Health began to fund large randomized clinical trials on the treatment of behavioral disorders (Barlow 1996, 2004). Although this approach will continue to yield important findings, there are limitations in terms of the efficiency and flexibility of this methodology as well as the generalizability of findings attained (Barlow & Nock 2009). Furthermore, despite the huge influence on clinical practice exerted by large-scale, randomized controlled trials, many therapists understandably wonder about the applicability of such results to their individual patients. In fact, variability in behavior is a function of a wide range of factors, and the task of the researcher and clinician alike is to discover functional relations

460 David H. Barlow, et al.

among independent variables, or more specifically among treatments and behavior disorders, over and above the welter of environmental and biological variables affecting the patient at any given time. A limitation of a nomothetic approach is the assumption that much of the variability, including occasional deterioration in some individuals, is intrinsic to the individual or due to uncontrollable external events; as a result, complex data analytic procedures are used to look for reliable effects over and above these factors, which are treated as error (Barlow & Nock 2009). An idiographic approach—one that focuses on the processes of change in the individual or, at the very least, outcomes in individuals—is capable of supplementing the nomothetic approaches as well as supporting meaningful advances in our understanding of psychopathology. The principal idiographic methodology utilized in treatment evaluation research is the single-case experimental design approach, where well-defined treatment components serving as independent variables are systematically manipulated within an individual for the purpose of establishing causal relationships between the treatment component and aspects of the disorder being treated (Barlow et al. 2009). It is also possible to take traditional randomized controlled trials and analyze the data more idiographically. For example, Mayou et al. (2000) reanalyzed the data from a group of individuals experiencing posttraumatic stress following severe traffic accidents. These individuals had received a brief intervention called Critical Incident Stress Debriefing immediately following the accident. Overall, the data showed no effect of the treatment when compared to individuals who did not receive the treatment. But when the data were broken down based on initial scores on the Impact of Events Scale (Horowitz et al. 1979), the results showed that individuals experiencing severe stress actually did worse both immediately and over a three-year follow-up period if they received the treatment compared to those who didn’t. Idiographic methodology, particularly single-case experimental designs, is not only useful in clinical research but in applied settings as well, since the procedures are much closer to what happens in the clinic. Thus, clinicians can directly partake in clinical research by using repeated measurement and carefully documenting response as well as generating hypotheses to explain individual variation in response (e.g., McCullough 2002), which will likely become more common as health care policy pushes for greater outcome tracking of treatment progress among all clinicians (Barlow & Carl 2011). Dissemination and Implementation

As noted above, despite the development and identification of well-supported psychological treatments, problems to date with the broad availability of treatments in practice settings have seriously limited their collective public health impact. These barriers interfere with the timely provision of mental health care (Chorpita et al. 2011, Kazdin & Blase 2011). Evidence of service utilization inadequacies is portrayed in reports showing only 40% of Americans with mental disorders over the past year have received any treatment for their condition in the past 12 months (Wang et al. 2005a). For those who do receive mental health care, the median delay in treatment initiation after onset of a disorder ranges from 5 to 9 years for substance use disorders, 6 to 8 years for

Evidence-Based Psychological 461

mood disorders, and 9 to 23 years for anxiety disorders (Wang et al. 2005b). Further, those who do receive mental health care are not necessarily receiving evidence-based treatments. Regrettably, widely used approaches rarely show strong support (e.g., Ennett et al. 1994), whereas treatments showing considerable support are rarely disseminated broadly. When supported treatments are broadly disseminated, they are rarely implemented with fidelity (McHugh & Barlow 2010). In response to this state of affairs, sizable financial commitments in recent years, totaling billions of dollars, as well as considerable scholarly attention, have focused on broad dissemination and implementation efforts aimed at improving the availability and quality of services delivered throughout practice settings (Spring & Neville 2011). Dissemination efforts constitute the purposeful distribution of relevant information and materials to practitioners, and implementation efforts constitute the adoption and integration of disseminated information and materials into actual clinical practice (Beidas et al. 2012). The two processes are complementary; in order to effect meaningful and sustainable change in practice settings, each is necessary but not sufficient without the other. Among the boldest and most far-reaching dissemination and implementation efforts as noted above are the Improving Access to Psychological Therapies program (Clark 2012) and the Veterans Affairs Health Care System Mental Health Strategic Plan (Ruzek et al. 2012). Both are nationally implemented, centralized therapist-training programs that promote competent evidencebased practice for affected individuals seeking mental health services in very large health care systems. Other impressive efforts include state dissemination programs that facilitate centralized training in and use of evidence-based psychological treatments (e.g., Bruns & Hoagwood 2008) as well as programs directly pursued by the very developers of evidence-based treatments (e.g., Landes & Linehan 2012, Schoenwald 2012). As dissemination and implementation efforts move ahead, high-quality programs seem to separate into two camps: (a) the Intensive Community Training model and (b) the Public Health model. The Intensive Community Training model entails intensive training of doctoral-level mental health specialists to specific criteria. Examples include the Improving Access to Psychological Therapies program as well as the extensive training efforts emanating from the Beck Institute for Cognitive Behavior Therapy and the Dialectical Behavior Therapy Intensive Training Program (see Landes & Linehan 2012), which all entail structured didactic training followed by extended consultation with treatment experts. The Public Health model typically entails the remote teletraining and supervision of large numbers of health workers with little or no formal training in mental health care, and although the training is typically less intensive than in Intensive Community Training models, this model has the potential to reach larger numbers of affected individuals in the settings where they commonly present (e.g., primary care settings). For example, the Coordinated Anxiety Learning and Management (CALM) program uses a computer-based system to assist nonexpert mental health care providers, such as nurses, in the delivery of evidence-based treatment for a range of anxiety disorders. In a large evaluation of the program, 1,004 primary care patients with anxiety disorders across 17

462 David H. Barlow, et al.

geographically diverse primary care clinics were treated with either the CALM program or TAU, which was determined randomly at the clinic level. At 12 months, 63.7% of CALM patients showed clinical response (relative to 44.7% of TAU patients), and 51.5% of CALM patients showed remission (relative to 33.3% of TAU patients) (Roy-Byrne et al. 2010). Despite the billions of dollars collectively spent on dissemination and implementation programs, the application of dissemination science to psychological treatments is still a relatively nascent field of study, and rigorous investigations of these leading dissemination and implementation efforts are just now underway. As such, it is safe to say that the majority of the work needed to clarify optimal methods for influencing broad mental health practices remains ahead of us. Indeed, at this early stage of activity, the state of dissemination and implementation science in clinical psychology may be only slightly further along than was intervention science in the 1970s—largely correlational, naturalistic, or quasi-experimental in nature, and lacking systematic controlled evaluations. Most existing controlled evaluations have only compared dissemination efforts to no dissemination efforts at all (e.g., treatment as usual), which only provides information about the utility of a broad dissemination package, rather than specific elements that may be most associated with effectiveness or elements that may be unnecessary or even counterproductive. Moreover, many early dissemination efforts have yielded mixed and/or disappointing findings, including poor treatment adherence and sustainability. Importantly, as noted elsewhere (Schoenwald et al. 2012), despite a proliferation of scholarly commentaries on dissemination theories and large program rollouts, broadly speaking, systematic investigations of dissemination models and strategies have yet to occur on a large scale. Descriptive and correlational studies provide important preliminary support for dissemination and implementation practices, but rigorous controlled evaluations will be needed to provide the most compelling evidence of best practices in dissemination efforts. It can be tempting to conjecture how Hans Eysenck, were he alive today, would characterize our current state of affairs. Based on his early conclusions about the effectiveness of psychotherapy in the 1950s, it is quite possible that he would be skeptical of the utility of dissemination efforts. Importantly, however, it seems instructive to take a page from our field’s earlier dialogue on the effectiveness of psychotherapy—which progressed from Eysenck’s assertion that psychotherapy may be ineffective to Paul’s (1967) more accurate and nuanced recognition that our real pursuit is to identify what specific treatments are effective for which specific clients under what circumstances. Indeed, it is increasingly apparent that the question is not, are dissemination and implementation efforts effective? but rather, what specific dissemination and implementation components and methods are effective in promoting evidence-based practices among which specific therapists for which specific clients in which practice settings? Despite a great plethora of scholarly discussions about optimal models of dissemination and implementation, only a small handful of controlled evaluations, most with relatively small samples, have begun to address this more productive and nuanced latter question (e.g., Beidas et al. 2012, Lochman et al. 2009, Sholomskas et al. 2005).

Evidence-Based Psychological 463

But early efforts do reveal some consistent findings. First, for lasting and quality implementation in practice settings, didactic trainings alone are insufficient. Competency components, which include ongoing consultation or supervision with experts, are essential for trainees to acquire the necessary proficiencies to administer treatments with skill and fidelity (Beidas et al. 2012, Crits-Christoph et al. 1998, McHugh & Barlow 2012, Sholomskas et al. 2005). These competency components often have trainees consult with treatment experts while applying the disseminated techniques in their actual caseloads (Daleiden et al. 2006, Gleacher et al. 2011). Second, the matter of sustainability, which is often implied but rarely pursued actively (Lyon et al. 2011), cannot be ignored. Sustainability presents multiple challenges to the lasting impact of even the most initially successful of dissemination efforts. Even seemingly simple and straightforward procedures have presented challenges to lasting implementation across all health care disciplines. For example, each year two million Americans acquire an infection while at a hospital and 90,000 die from this infection (Gawande 2007). Despite widespread efforts to promote sustained adherence to hand washing procedures among hospital clinicians—procedures that can substantially halt the spread of many of these hospital-acquired infections—clinician compliance with hand washing procedures in hospitals remains at a dismal 33% to 50% (Gawande 2007). Common reasons offered for not engaging in systematic hand washing among noncompliant hospital staff include inconvenience, insufficient time, skin irritation, noxious odors of cleansers, and even concerns that cleansers reduce fertility. Of course, the quality delivery of an indicated psychological treatment protocol is far more involved and complex than simply washing one’s hands, underscoring the tremendous and unique challenge for our field’s efforts toward sustained implementation. In addition to targeting individual trainee factors (Gallo & Barlow 2012), efforts to promote sustainability must also target organizational factors (see Beidas & Kendall 2010). The targeting of organizational factors is far less common across the majority of current dissemination and implementation efforts. Low levels of agency support, the absence of internal program champions, and fluctuating or insufficient agency resources interfere with organizational sustainability (Glisson et al. 2008). Workforce instability presents another challenge. Community agencies, particularly overburdened, poorly funded clinics, exhibit high staff turnover (Glisson et al. 2008). Given the relatively lengthy nature of meaningful dissemination and implementation efforts, by the time training is complete, a participating trainee may no longer work at the agency. Despite great resources invested in training this individual, the knowledge and skills do not transfer to the next individual holding the position. “Train-thetrainers” approaches, which focus on the training of local supervisors who presumably have greater job stability than individual clinicians, provide some protection against treatment fidelity drift across time. Finally, for these reasons, quality large-scale dissemination and implementation efforts are enormously costly. Some of the more successful programs have collectively cost hundreds of millions of dollars and still have not actualized the large and lasting impacts aspired to at the program’s outset. As such, rigorous evaluations at the patient, therapist, trainer, and organizational level,

464 David H. Barlow, et al.

which themselves are costly, have not always been at the forefront of dissemination and implementation agendas. Most commonly, research on dissemination programs has been restricted to investigations of the feasibility of implementation and changes in provider attitudes and knowledge, which may or may not correlate with changes in provider skill, and ultimately, patient outcomes. As McHugh & Barlow (2010) note, the majority of leading dissemination efforts have not included assessments related to patient outcomes, the number of patients receiving services, or clinician outcomes (e.g., clinician attrition, percent of clinicians who achieve competence, clinical skill level at pre- and posttreatment). Controlled comparisons in which key dissemination elements are systematically manipulated (e.g., heterogeneous stakeholder involvement, agency driven versus trainer driven, intensive versus spaced training, in-person versus web-based training, individual versus group supervision, tape feedback) will also be essential in order to optimize efforts. In the absence of such important evaluations, we may be ignoring that large components of our leading dissemination efforts are as misguided as were large components of our early models of psychotherapy. Concerns have been articulated elsewhere (J.S. Comer & D.H. Barlow, manuscript submitted) about putting all of our eggs into the dissemination and implementation basket in order to address our field’s sizable problems regarding psychological treatment availability and quality in practice settings. Importantly, low base-rate disorders and complex treatments present serious obstacles to dissemination and implementation efforts. The time and expenses associated with quality training and sustained implementation preclude largescale dissemination efforts for the treatment of a number of very serious mental health concerns that impose tremendous personal consequences but affect only a relatively small proportion of patients seeking care (e.g., chronic tic disorders, paraphilias, schizophrenia, selective mutism, Tourette’s disorder, trichotillomania). Moreover, dissemination science across a range of disciplines shows that innovations that are too complex do not get routinely incorporated into everyday practice (Rogers 2003), as innovation adopters show preference for user-friendly and “plug-and-play” methods. Within the context of broad dissemination and implementation, there appears to be little role for treatment methods that cannot be watered down for delivery in overburdened clinics by minimally trained professionals. Given the diversity of backgrounds represented across the mental health workforce (Ellis et al. 2009), some wellsupported psychological treatments may prove too complex for inadequately or insufficiently trained clinicians to implement. In the event that broad dissemination and implementation efforts actualize their enormous potential, we will have a high-quality workforce of generalist clinicians who are well equipped to skillfully address the majority of presenting mental health problems with relatively uncomplicated methods. The promotion of such a competent generalist mental health workforce will be increasingly essential in the years to come, as the Affordable Care Act envisions an increased role for the colocation of mental health care within integrated primary care settings. But a highly competent generalist workforce

Evidence-Based Psychological 465

alone will not ensure effective services for individuals with low base-rate conditions or individuals requiring mental health care for conditions so uncommon that they are not worth the considerable investment required for meaningful dissemination and implementation. Thus the authors of the current review have identified a vital need for available specialty care in the delivery of psychological treatments, transacting with the generalist workforce to collectively ensure the greatest quality and timely delivery of appropriate treatments to patients in need (J.S. Comer & D.H. Barlow, manuscript submitted). Given the limits of broad dissemination and implementation to local mental health workforces as well as the obstacles that traditionally interfere with the accessibility of expert specialty care (e.g., geographic workforce disparities), many believe technological innovations may be central to this effort (Kazdin & Blase 2011). Of course, technology-assisted delivery of psychological treatments still merits considerable empirical scrutiny. But in the event of continued support for technology-assisted treatments, specialty behavioral telehealth—in which expert specialty services are offered in real-time through the use of live videoconferencing either directly to clients in their homes or in other private settings with Internet access—may offer a transformative option for patients suffering with low base-rate conditions or conditions for which complex treatments show strongest support (J.S. Comer & D.H. Barlow, manuscript submitted). However, before a specialty behavioral telehealth delivery model can be systematically incorporated into mental health care, a number of logistic and professional issues will need to be resolved, including licensure and liability matters governing the practice of telehealth, credentialing standards, and reimbursement and appropriate Current Procedural Terminology coding for telehealth procedures. Final Remarks This update of evidence-based psychological treatments is best considered an interim report. In the current state of science and practice we have clear accomplishments we can point to, but we have much to be humble about. The barriers to greater availability of psychological treatments reviewed above represent challenges in the near to the intermediate term that must be overcome. Central to this mission will be identifying more clearly mechanisms of action in these treatments that will presumably allow the development of more efficient and effective interventions. As these goals are achieved and the public and policymakers alike become more aware of these developments, demand and availability should continue to increase, and we will hopefully be closer to our collective goal of relieving human suffering and enhancing human functioning. Disclosure Statement The authors are not aware of any affiliations, memberships, funding, or financial holdings that might be perceived as affecting the objectivity of this review.

466 David H. Barlow, et al.

Literature Cited Addington J, Piskulic D, Marshall D. 2010. Psychological treatments for schizophrenia. Curr. Dir. Psychol. Sci. 19:260–63 Arkowitz H, Westra HA, Miller WR, Rollnick S. 2008. Motivational Interviewing in the Treatment of Psychological Problems. New York: Guilford Barlow DH. 1988. Anxiety and Its Disorders: The Nature and Treatment of Anxiety and Panic. New York: Guilford Barlow DH. 1991. Disorders of emotion. Psychol. Inq. 2:58–71 Barlow DH. 1996. Healthcare policy, psychotherapy research, and the future of psychotherapy. Am. Psychol. 51:1050–58 Barlow DH. 2000. Unraveling the mysteries of anxiety and its disorders from the perspective of emotion theory. Am. Psychol. 55:1247–63 Barlow DH. 2002. Anxiety and Its Disorders: The Nature and Treatment of Anxiety and Panic. New York: Guilford. 2nd ed. Barlow DH. 2004. Psychological treatments. Am. Psychol. 59:869–78 Barlow DH. 2006. Psychotherapy and psychological treatments: the future. Clin. Psychol. Sci. Pract. 13:216–20 Barlow DH. 2008. Clinical Handbook of Psychological Disorders: A Step-by-Step Treatment Manual. New York: Guilford. 4th ed. Barlow DH. 2010. Negative effects from psychological treatments: a perspective. Am. Psychol. 65:13–20 Barlow DH. 2011a. Aprolegomenon to clinical psychology: two 40-year odysseys. See Barlow 2011b, pp. 3–20 Barlow DH. 2011b. The Oxford Handbook of Clinical Psychology. New York: Oxford Univ. Press Barlow DH, Carl JR. 2011. The future of clinical psychology: promises, perspectives, and predictions. See Barlow 2011b, pp. 891–912 Barlow DH, Craske MG. 2007. Mastery of Your Anxiety and Panic: Workbook. New York: Oxford Univ. Press. 4th ed. Barlow DH, Farchione TJ, Fairholme CP, Ellard KK, Boisseau CL, et al. 2011. Unified Protocol for Transdiagnostic Treatment of Emotional Disorders: Therapist Guide. New York: Oxford Univ. Press Barlow DH, Gorman JM, Shear MK, Woods SW. 2000. Cognitive-behavioral therapy, imipramine, or their combination for panic disorder: a randomized controlled trial. JAMA 283:2529–36 Barlow DH, Nock MK. 2009. Why can’t we be more idiographic in our research? Perspect. Psychol. Sci. 4:19–21 Barlow DH, Nock MK, Hersen M. 2009. Single Case Experimental Designs: Strategies for Studying Behavior Change. Boston, MA: Pearson. 3rd ed. Barlow, DH, Sauer-Zavala, S, Carl, JR, Bullis, JR, & Ellard, KK. The nature, diagnosis, and treatment of neuroticism: Back to the future. Manuscript submitted for publication. Beidas RS, Edmunds J, Marcus S, Kendall PC. 2012. An RCT of training and consultation as implementation strategies for an empirically supported treatment. Psychiatr. Serv. 63:660–65 Beidas RS, Kendall PC. 2010. Training therapists in evidence-based practice: a critical review of studies from a systems-contextual perspective. Clin. Psychol. Sci. Pract. 17:1–30 Beidas RS, Mehta T, Atkins M, Solomon B, Merz J. 2012. Dissemination and implementation science: research models and methods. In The Oxford Handbook of Research Strategies for Clinical Psychology, ed. JS Comer, PC Kendall. New York: Oxford Univ. Press. In press

Evidence-Based Psychological 467 Bergin AE, Strupp HH. 1972. Changing Frontiers in the Science of Psychotherapy. Chicago, IL: Aldine Blackledge JT, Hayes SC. 2001. Emotion regulation in acceptance and commitment therapy. J. Clin. Psychol. 57:243–55 Bouton ME, Mineka S, Barlow DH. 2001. A modern learning theory perspective on the etiology of panic disorder. Psychol. Rev. 108:4–32 Brown TA. 2007. Temporal course and structural relationships among dimensions of temperament and DSMIV anxiety and mood disorder constructs. J. Abnorm. Psychol. 116:313–28 Brown TA, Antony MM, Barlow DH. 1995. Diagnostic comorbidity in panic disorder: effect on treatment outcome and course of comorbid diagnoses following treatment. J. Clin. Psychol. 63:408–18 Brown TA, Barlow DH. 2009. A proposal for a dimensional classification system based on the shared features of the DSM-IV anxiety and mood disorders: implications for assessment and treatment. Psychol. Assess. 21:256–71 Brown TA, Chorpita BF, Barlow DH. 1998. Structural relationships among dimensions of the DSM-IV anxiety and mood disorders and dimensions of negative affect, positive affect, and autonomic arousal. J. Abnorm. Psychol. 107:179–92 Bruns EJ, Hoagwood KE. 2008. State implementation of evidence-based practice for youths, part I: responses to the state of the evidence. J. Am. Acad. Child Adolesc. Psychiatry 47:369–73 Burgio KL, Locher JL, Goode PS, Hardin JM, McDowell BJ, et al. 1998. Behavioral versus drug treatment for urge urinary incontinence in older women: a randomized controlled trial. JAMA 280:1995–2000 Campbell-Sills L, Barlow DH. 2007. Incorporating emotion regulation into conceptualizations and treatments of anxiety and mood disorders. In Handbook of Emotion Regulation, ed. JJ Gross, pp. 542–59. New York: Guilford Campbell-Sills L, Barlow DH, Brown TA, Hofmann SG. 2006. Effects of suppression and acceptance on emotional responses of individuals with anxiety and mood disorders. Behav. Res. Ther. 44:1251–63 Campbell-Sills L, Ellard KK, Barlow DH. 2012. Emotion regulation in anxiety disorders. In Handbook of Emotion Regulation, ed. JJ Gross. New York: Guilford. 2nd ed. In press Chawla N, Ostafin B. 2007. Experiential avoidance as a functional dimensional approach to psychopathology: an empirical review. J. Clin. Psychol. 63:871–90 Chorpita BF, Miranda J, Bernstein A. 2011. Creating public health policy: the dissemination of evidence-based psychological interventions. See Barlow 2011b, pp. 210–24 Chorpita BF, Weisz JR. 2009. MATCH-ADTC: Modular Approach to Therapy for Children with Anxiety, Depression, Trauma, or Conduct Problems. Satellite Beach, FL: PracticeWise Clark DM. 2012. The English Improving Access to Psychological Treatments (IAPT) program: history and progress. See McHugh & Barlow 2012, pp. 61–77 Clark LA. 2005. Temperament as a unifying basis for personality and psychopathology. J. Abnorm. Psychol. 114:505–21 Clark LA, Watson D, Mineka S. 1994. Temperament, personality, and the mood and anxiety disorders. J. Abnorm. Psychol. 103:103–16 Comer JS, Barlow DH, The occasional case against broad dissemination and implementation: Retaining a role for specialty care in the delivery of psychological treatments. Manuscript submitted for publication. Comer JS, Mojtabai R, Olfson M. 2011. National trends in the antipsychotic treatment of psychiatric outpatients with anxiety disorders. Am. J. Psychiatry 168:1057–65

468 David H. Barlow, et al. Craske MG, Barlow DH. 2008. Panic disorder and agoraphobia. In Clinical Handbook of Psychological Disorders: A Step-by-Step Treatment Manual, ed. DH Barlow, pp. 1–64. New York: Guilford. 4th ed. Craske MG, Kircanski K, Zelikowsky M, Mystkowski J, Chowdhury N, Baker A. 2008. Optimizing inhibitory learning during exposure therapy. Behav. Res. Ther. 46:5–27 Crits-Christoph P, Siqueland L, Chittams J, Barber JP, Beck AT, et al. 1998. Training in cognitive, supportive expressive, and drug counseling therapies for cocaine dependence. J. Consult. Clin. Psychol. 66:484–92 Cuijpers P, Driessen E, Hollon SD, van Oppen P, Barth J, Andersson G. 2012. The efficacy of non-directive supportive therapy for adult depression: a meta-analysis. Clin. Psychol. Rev. 32:280–91 Cuijpers P, Hollon SD, van Straten A, Bockting C, Berking M, & Andersson G. Does cognitive behaviour therapy have an enduring effect that is superior to keeping patients on continuation pharmacotherapy? Manuscript submitted for publication. Daleiden EL, Chorpita BF, Donkervoet C, Arensdorf AM, Brogan M. 2006. Getting better at getting them better: health outcomes and evidence-based practice within a system of care. J. Am. Acad. Child Adolesc. Psychiatry 45:749–56 Donta ST, Clauw DJ, Engel CC, Guarino P, Peduzzi P, et al. 2003. Cognitive behavioral therapy and aerobic exercise for Gulf War veterans’ illness: a randomized controlled trial. JAMA 289:1396–404 Driessen E, Cuijpers P, Hollon SD, Decker JJ. 2010. Does pretreatment severity moderate the efficacy of psychological treatment of adult outpatient depression? A meta-analysis. J. Consult. Clin. Psychol. 78:668–80 Eack SM, Hogarty GE, Cho RY, Prasad KM, Greenwald DP, et al. 2010. Neuroprotective effects of cognitive enhancement therapy against gray matter loss in early schizophrenia: results from a 2-year randomized controlled trial. Arch. Gen. Psychiatry 67:674–82 Ellard KK, Fairholme CP, Boisseau CL, Farchione TJ, Barlow DH. 2010. Unified Protocol for the Transdiagnostic Treatment of Emotional Disorders: protocol development and initial outcome data. Cogn. Behav. Pract. 17:88–101 Ellis AR, Konrad TR, Thomas KC, Morrissey JP. 2009. County-level estimates of mental health professional supply in the United States. Psychiatr. Serv. 60:1315–22 Ennett ST, Tobler NS, Ringwalt CL, Flewelling RL. 1994. How effective is drug abuse resistance education? A meta-analysis of Project DARE outcome evaluations. Am. J. Public Health 84:1394–401 Eysenck HJ. 1952. The effects of psychotherapy: an evaluation. J. Consult. Psychol. 16:319–27 Eysenck HJ. 1965. The effects of psychotherapy. Int. J. Psychiatry 1:97–178 Fairburn CG, Cooper Z, Doll HA, O’Connor ME, Bohn K, et al. 2009. Transdiagnostic cognitive-behavioral therapy for patients with eating disorders: a two-site trial with 60-week follow-up. Am. J. Psychiatry 166:311–19 Farchione TJ, Fairholme CF, Ellard KK, Boisseau CL, Thompson-Hollands J, et al. 2012. The Unified Protocol for Transdiagnostic Treatment of Emotional Disorders: a randomized controlled trial. Behav. Ther. 43:666–78 Gallo KP, Barlow DH. 2012. Factors involved in clinician adoption and nonadoption of evidence-based interventions in mental health. Clin. Psychol. Sci. Pract. 19:93–106 Gawande A. 2007. Better: A Surgeon’s Notes on Performance. New York: Metropolitan Books Gleacher AA, Nadeem E, Moy AJ, Whited AL, Albano AM, et al. 2011. Statewide CBT training for clinicians and supervisors treating youth: the New York State Evidence Based Treatment Dissemination Center. J. Emot. Behav. Disord. 19:182–92 Glisson C, Schoenwald SK, Kelleher K, Landsverk J, Hoagwood KE, et al. 2008. Therapist turnover and new program sustainability in mental health clinics as a function

Evidence-Based Psychological 469 of organizational culture, climate, and service structure. Adm. Policy Ment. Health 35:124–33 Goode PS, Burgio KL, Locher JL, Roth DL, Umlauf MG, et al. 2003. Effects of behavioral training with or without pelvic floor electrical stimulation on stress incontinence in women: a randomized controlled trial. JAMA 290:345–52 Grant PM, Huh GA, Perivoliotis D, Stolar NM, Beck AT. 2012. Randomized trial to evaluate the efficacy of cognitive therapy for low-functioning patients with schizophrenia. Arch. Gen. Psychiatry 69:121–27 Griffith JW, Zinbarg RE, Craske MG, Mineka S, Rose RD, et al. 2010. Neuroticism as a common dimension in the internalizing disorders. Psychol. Med. 40:1125–36 Gross JJ. 1998. Antecedent- and response-focused emotion regulation: divergent consequences for experience, expression, and physiology. J. Personal. Soc. Psychol. 74:224–37 Guastella AJ, Richardson R, Lovibond PF, Rapee RM, Gaston JE, et al. 2008. A randomized controlled trial of D-cycloserine enhancement of exposure therapy for social anxiety disorder. Biol. Psychiatry 63:544–49 Harmer CJ, Goodwin GM, Cowen PJ. 2009. Why do antidepressants take so long to work? A cognitive neuropsychological model of antidepressant drug action. Br. J. Psychiatry 195:102–8 Harvey AG, Murray G, Chandler RA, Soehner A. 2010. Sleep disturbance as transdiagnostic: consideration of neurobiological mechanisms. Clin. Psychol. Rev. 3:225–35 Heinrichs M, von Dawans B, Domes G. 2009. Oxytocin, vasopressin, and human social behavior. Front. Neuroendocrinol. 30:548–57 Hersen M, Barlow DH. 1976. Single Case Experimental Designs: Strategies for Studying Behavior Change. New York: Pergamon Hofmann SG. 2007. Enhancing exposure-based therapy from a translational research perspective. Behav. Res. Ther. 45:1987–2001 Hofmann SG, Smits JA, Asnaani A, Gutner CA, Otto MW. 2011. Cognitive enhancers for anxiety disorders. Behav. Res. Ther. 99:275–84 Hollon SD. 2011. Cognitive and behavioral therapy in the treatment and prevention of depression. Depress. Anxiety 28:263–66 Horowitz M, Wilner N, Alvarez W. 1979. Impact of Event Scale: a measure of subjective stress. Psychosom. Med. 41:209–18 Huppert JD, Bufka LF, Barlow DH, Gorman JM, Woods SW, Shear MK. 2001. Therapists, therapist variables, and cognitive-behavioral therapy outcome in a multi-center trial for panic disorder. J. Consult. Clin. Psychol. 69:747–55 Huppert JD, Franklin ME, Foa EB, Simpson HB, Barlow DH. 2007. The relative contribution of therapists and technique: a reassessment using randomized trials of CBT for anxiety disorders. Presented at Annu. Meet. Soc. Psychother. Res., 38th, Madison, WI Huppert JD, Kivity Y, Barlow DH, Gorman JM, Shear MK, & Woods SW. Therapist effects and the outcome–alliance correlation in cognitive behavioral therapy for panic disorder with agoraphobia. Manuscript submitted for publication. Huppert JD, Schultz LT, Foa EB, Barlow DH, Davidson JR, et al. 2004. Differential response to placebo among patients with social phobia, panic disorder, and obsessive-compulsive disorder. Am. J. Psychiatry 161:1485–87 Kazdin AE. 2011. Evidence-based treatments research: advances, limitations, and next steps. Am. Psychol. 66:685–98 Kazdin AE, Blase SL. 2011. Rebooting psychotherapy research and practice to reduce the burden of mental illness. Perspect. Psychol. Sci. 6:21–37 Keller MB, McCullough JP, Klein DN, Arnow B, Dunner DL, et al. 2000. A comparison of nefazodone, the cognitive behavioral-analysis system of psychotherapy, and their combination for the treatment of chronic depression. N. Engl. J. Med. 342:1462–70

470 David H. Barlow, et al. Kernberg O. 1973. Summary and conclusions of psychotherapy and psychoanalysis. Final report of the Menninger Foundation’s Psychotherapy Research Project. Critical evaluations. Author’s reply. Int. J. Psychiatry 11:95–103 Labuschagne I, Phan KL, Wood A, Angstadt M, Chua P, et al. 2012. Medial frontal hyperactivity to sad faces in generalized social anxiety disorder and modulation by oxytocin. Int. J. Neuropsychopharmacol. 15:883–96 Lambert A. 2008. New Treatment Options for Circadian Rhythm Sleep Disorders. Bethesda, MD: Clinical Compass. http://www.neurosciencecme.com/email/2008/100708_ckc.htm Landes SJ, Linehan MM. 2012. Dissemination and implementation of dialectical behavior therapy: an intensive training model. See McHugh & Barlow 2012, pp. 187–208 Leykin Y, DeRubeis RJ. 2009. Allegiance in psychotherapy outcome: separating association from bias. Clin. Psychol. Sci. Pract. 16:54–65 Litz BT, Salters-Pedneault K, Steenkamp MM, Hermos JA, Bryant RA, et al. 2012. A randomized placebo-controlled trial of D-cycloserine and exposure therapy for posttraumatic stress disorder. J. Psychiatr. Res. 71:962–68 Lochman JE, Boxmeyer C, Powell N, Qu L, Wells K, Windle M. 2009. Dissemination of the Coping Power program: importance of intensity of counselor training. J. Consult. Clin. Psychol. 77:397–409 Lyon AR, Frazier SL, Mehta T, Atkins MS, Weisbach J. 2011. Easier said than done: intervention sustainability in an urban after-school program. Adm. Policy Ment. Health 38:504–17 Marcus SC, Olfson M. 2010. National trends in the treatment for depression from 1998 to 2007. Arch. Gen. Psychiatry 67:1265–73 Mayou RA, Ehlers A, Hobbs M. 2000. Psychological debriefing for road traffic accident victims. Three-year follow-up of a randomised controlled trial. Br. J. Psychiatry 176:589–92 McCullough JP. 2002. What questions are we trying to answer with our psychotherapy research? Clin. Psychol. Sci. Pract. 9:447–52 McHugh RK, Barlow DH. 2010. The dissemination and implementation of evidence-based psychological treatments. A review of current efforts. Am. Psychol. 65:73–84 McHugh RK, Barlow DH, eds. 2012. Dissemination and Implementation of Evidence-based Psychological Interventions. New York: Oxford Univ. Press McHugh RK, Whitton SW, Peckham AD, Welge JA, Otto MW. 2012. Patient preference for psychological vs. pharmacological treatment of psychiatric disorders: a meta-analytic review. J. Clin. Psychol. In press McLaughlin KA, Nolen-Hoeksema S. 2011. Rumination as a transdiagnostic factor in depression and anxiety. Behav. Res. Ther. 49:186–93 Mitchell MD, Gehrman P, Perlis M, Umscheid CA. 2012. Comparative effectiveness of cognitive behavioral therapy for insomnia: a systematic review. BMC Fam. Pract. 13:40 Moher D, Schulz KF, Altman D. 2001. The CONSORT statement: revised recommendation for improving the quality of reports of parallel-group randomized trials. JAMA 285:1987–91 Monfils MH, Cowansage KK, Klann E, LeDoux JE. 2009. Extinction-reconsolidation boundaries: key to persistent attenuation of fear memories. Science 324:951–55 Morgenthaler T, Kramer M, Alessi C, Friedman L, Boehlecke B, et al. 2006. Practice parameters for the psychological and behavioral treatment of insomnia: an update. An American Academy of Sleep Medicine report. Sleep 29:1415–19 Morin CM, Colecchi C, Stone J, Sood R, Brink D. 1999. Behavioral and pharmacological therapies for late-life insomnia: a randomized controlled trial. JAMA 281:991–99 Morin CM, Vallieres A, Guay B, Ivers H, Savard J, et al. 2009. Cognitive behavioral therapy, singly and combined with medication, for persistent insomnia: a randomized controlled trial. JAMA 301:2005–15

Evidence-Based Psychological 471 Nathan PE, Gorman JM. 2007. Treatments That Work. New York: Oxford Univ. Press. 3rd ed. Newman MG, Przeworski A, Fisher AJ, Borkovec TD. 2010. Diagnostic comorbidity in adults with generalized anxiety disorder: impact of comorbidity on psychotherapy outcome and impact of psychotherapy on comorbid diagnoses. Behav. Ther. 41:59–72 Nock MK. 2007. Conceptual and design essentials for evaluating mechanisms of change. Alcohol. Clin. Exp. Res. 31:4–12s Norton PJ. 2008. An open trial of a transdiagnostic cognitive-behavioral group therapy for anxiety disorder. Behav. Ther. 39:242–50 Olfson M, Marcus SC. 2009. National patterns in antidepressant medication treatment. Arch. Gen. Psychiatry 66:848–56 Olfson M, Marcus SC, Corey-Lisle P, Tuomari AV, Hines P, L’Italien GJ. 2006. Hyperlipidemia following treatment with antipsychotic medications. Am. J. Psychiatry 163:1821–25 Otto MW, Tolin DF, Simon NM, Pearlson GD, Basden S, et al. 2010. Efficacy of D-cycloserine for enhancing response to cognitive-behavior therapy for panic disorder. Biol. Psychiatry 67:365–70 Paul GL. 1967. Strategy of outcome research in psychotherapy. J. Consult. Psychol. 31:109–18 Pfammatter M, Junghan UM, Brenner HD. 2006. Efficacy of psychological therapy in schizophrenia: conclusions from meta-analyses. Schizophr. Bull. 32(Suppl. 1):S64–80 Piacentini J, Woods DW, Scahill L, Wilhelm S, Peterson AL, et al. 2010. Behavior therapy for children with Tourette disorder: a randomized controlled trial. JAMA 303:1929–37 Powers E, Witmer H. 1951. An Experiment in the Prevention of Delinquency: The CambridgeSomerville Youth Study. New York: Columbia Univ. Press Ressler KJ, Rothbaum BO, Tannenbaum L, Anderson P, Graap K, et al. 2004. Cognitive enhancers as adjuncts to psychotherapy: use of D-cycloserine in phobic individuals to facilitate extinction of fear. Arch. Gen. Psychiatry 61:1136–44 Rogers EM. 2003. Diffusion of Innovations. New York: Free Press. 5th ed. Rosellini AJ, Brown TA. 2011. The NEO Five-Factor Inventory: latent structure and relationships with dimensions of anxiety and depressive disorders in a large clinical sample. Assessment 18:27–38 Rosenberg RP. 2006. Sleep maintenance insomnia: strengths and weaknesses of current pharmacologic therapies. Ann. Clin. Psychiatry 18:49–56 Roy-Byrne P, Craske MG, Sullivan G, Rose RD, Edlund MJ, et al. 2010. Delivery of evidence-based treatment for multiple anxiety disorders in primary care: a randomized controlled trial. JAMA 303:1921–28 Ruzek JI, Karlin BE, Zeiss A. 2012. Implementation of evidence-based psychological treatments in the Veterans Health Administration. See McHugh & Barlow 2012, pp. 78–98 Schnurr PP, Friedman MJ, Engel CC, Foa EB, Shea MT, et al. 2007. Cognitive behavioral therapy for posttraumatic stress disorder in women: a randomized controlled trial. JAMA 297:820–30 Schoenwald SK. 2012. The transport and diffusion of multisystemic therapy. See McHugh & Barlow 2012, pp. 227–46 Schoenwald SK, McHugh RK, Barlow DH. 2012. The science of dissemination and implementation. See McHugh & Barlow 2012, pp. 16–42 Shear MK, Houck P, Greeno C, Masters S. 2001. Emotion-focused psychotherapy for patients with panic disorder. Am. J. Psychiatry 158:1993–98 Sholomskas DE, Syracuse-Siewert G, Rounsaville BJ, Ball SA, Nuro KF, Carroll KM. 2005. We don’t train in vain: a dissemination trial of three strategies of training clinicians in cognitive-behavioral therapy. J. Consult. Clin. Psychol. 73:106–15 Sivertsen B, Omvik S, Pallesen S, Bjorvatn B, Havik OE, et al. 2006. Cognitive behavioral therapy vs zopiclone for treatment of chronic primary insomnia in older adults: a randomized controlled trial. JAMA 295:2851–58

472 David H. Barlow, et al. Spring B, Neville K. 2011. Evidence-based practice in clinical psychology. See Barlow 2011b, pp. 128–49 Stangier U, Schramm E, Heidenreich T, Clark DM. 2011. Cognitive therapy versus interpersonal psychotherapy in social anxiety disorder. Arch. Gen. Psychiatry 68:692–700 Strupp HH. 1964. The outcome problem in psychotherapy: a rejoinder. Psychother. Res. Pract. 1:101 Tarrier N. 2008. Schizophrenia and other psychotic disorders. In Clinical Handbook of Psychological Disorders: A Step-by-Step Treatment Manual, ed. DH Barlow, pp. 463–91. New York: Guilford. 4th ed. Teri L, Gibbons LE, McCurry SM, Logsdon RG, Buchner DM, et al. 2003. Exercise plus behavioral management in patients with Alzheimer disease: a randomized controlled trial. JAMA 290:2015–22 Walker DL, Ressler KJ, Lu KT, Davis M. 2002. Facilitation of conditioned fear extinction by systemic administration or intra-amygdala infusions of D-cycloserine as assessed with fear-potentiated startle in rats. J. Neurosci. 22:2343–51 Wampold BE. 2001. The Great Psychotherapy Debate: Models, Methods, and Findings. Hillsdale, NJ: Erlbaum Wang PS, Berglund P, Olfson M, Pincus HA, Wells KB, Kessler RC. 2005b. Failure and delay in initial treatment contact after first onset of mental disorders in the National Comorbidity Survey Replication. Arch. Gen. Psychiatry 62:603–13 Wang PS, Lane M, Olfson M, Pincus HA, Wells KB, Kessler RC. 2005a. Twelve-month use of mental health services in the United States: results from the National Comorbidity Survey Replication. Arch. Gen. Psychiatry 62:629–40 Weisz JR, Chorpita BF, Palinkas LA, Schoenwald SK, Miranda J, et al. 2012. Testing standard and modular designs for psychotherapy treating depression, anxiety, and conduct problems in youth: a randomized effectiveness trial. Arch. Gen. Psychiatry 69:274–82 Weisz JR, Gray JS. 2008. Evidenced-based psychotherapy for children and adolescents: data from the present and a model for the future. Child Adolesc. Ment. Health 13:54–65 Weisz JR, Kazdin AE. 2010. Evidence-Based Psychotherapies for Children and Adolescents. New York: Guilford. 2nd ed. Wilamowska ZA, Thompson-Hollands J, Fairholme CP, Ellard KK, Farchione TJ, Barlow DH. 2010. Conceptual background, development, and preliminary data from the Unified Protocol for Transdiagnostic Treatment of Emotional Disorders. Depress. Anxiety 27:882–90 Wilhelm S, Buhlmann U, Tolin DF, Meunier SA, Pearlson GD, et al. 2008. Augmentation of behavior therapy with D-cycloserine for obsessive-compulsive disorder. Am. J. Psychiatry 165:335–41 Witt DM, Winslow JT, Insel TR. 1992. Enhanced social interactions in rats following chronic, centrally infused oxytocin. Pharmacol. Biochem. Behav. 43:855–61 Xue YX, Luo YX, Wu P, Shi HS, Xue LF, et al. 2012. A memory retrieval-extinction procedure to prevent drug craving and relapse. Science 336:241–45

Index

ABCT see Association for Behavioral and Cognitive Therapy action tendencies 205, 217; anxiety 210; depression 210; emotion-related 207, 345; fear 207, 341; fight-or-flight behavior 270, 341; negative emotions 226; panic 207, 270, 271 ADIS see Anxiety Disorders Interview Schedule Affordable Care Act 462 Agency for Health Care Policy and Research (AHCPR) 391, 392, 396 Agency for Healthcare Research and Quality 455 agoraphobia 356, 374; cognitive-behavioral therapy 284; conditioning theories 268, 269; genetic factors 292; low extraversion 360, 367; mild 127; panic disorder with 263, 265, 285, 287, 292, 295, 325, 335, 394, 451; sex differences 293; situational avoidance 364 Agras, S. 395 AHCPR see Agency for Health Care Policy and Research Ainsworth, M. D. S. 241, 243 Akiskal, H. S. 216 alarm theory 272, 287; future 273; reaction 205; see also false alarms; learned alarms; true alarms Allen, L. B. 441 Alloy, L. B. 213, 214, 227, 228, 236 alternative classification systems 333 – 9; higher-order models 334 – 7; nosological emphasis on shared dimensions 337 – 9 Alzheimer’s disease 444 American Academy of Sleep Medicine 444 American Psychological Association (APA) 417, 434 American Psychological Association, Task Force on Psychological Intervention

Guidelines: 399, 402; Template for Developing Guidelines: Interventions for Mental Disorders and Psychosocial Aspects of Physical Disorders 397 Ametaj, A. A.: “Evidence-Based Psychological Treatments: An Update and a Way Forward” 439 – 70 amygdala 271, 274, 296, 357, 364, 372, 449 anger and stress disorder 216 – 17 Anxiety Control Questionnaire (ACQ) 234 anxiety disorders 202 – 5; causes of anxiety and 212 – 13; commonalities and dimensions 356 – 7; conceptual model and implications 251 – 3; delivery of treatment and dissemination of results 65; and depression 226 – 7; development 225 – 62; versus panic 207 – 8; see also generalized anxiety disorder; stressors and resilience Anxiety Disorders Interview Schedule (ADIS) 324, 333; Lifetime Version (ADIS-IV-L) 324, 325 anxiety reduction 281; see also fear and anxiety reduction anxiety sensitivity theory 266, 267 – 8, 362 – 3 arousal: autonomic 271, 360 Association for Behavioral and Cognitive Therapy (ABCT) 434 attributional style 235 – 7; child 238, 239 Attributional Style Questionnaire 211 aversiveness 281 avoidance behavior 281, 339, 341, 367; agoraphobic 265 Bandura, A.: self-efficacy 84, 85 – 6, 249 Barlow, D. H. 272 Barlow, D. H., works of 289, 302; “The Development of Anxiety: The Role of Control in the Early Environment”

474 Index 225 – 62; “Direct-to-Consumer Marketing of Evidence-Based Psychological Interventions: Introduction” 431 – 8; “The Dissemination and Implementation of Evidence-Based Psychological Treatments: A Review of Current Efforts 407 – 30; “Evidence-Based Psychological Treatments: An Update and a Way Forward” 439 – 70; “Health Care Policy, Psychotherapy Research, and the Future of Psychotherapy” 391 – 406; “A Modern Learning Theory Perspective on the Etiology of Panic Disorder” 263 – 322; “The Nature, Diagnosis, and Treatment of Neuroticism: Back to the Future” 353 – 88; “A Proposal for a Dimensional Classification System Based on the Shared Features of the SDM-IV Anxiety and Mood Disorders: Implications for Assessment and Treatment” 323 – 52; Unified Protocol for Transdiagnostic Treatment of Emotional Disorders 442, 452, 454 Basoglu, M. 285 Bechara, A. 274 Beck, A. T. 266 Beck, J. G. 202, 210 Beck Depression Inventory 210 – 11 Beck Institute for Cognitive Behavior Therapy 459 behavioral inhibition system (BIS) 226, 228, 230, 248, 251 – 3, 358 benzodiazepine 204; high-potency 208 – 9 Bergin, A. E. 441 Berwick, 435 Birbaumer, N. 301 BIS see behavioral inhibition system Blackwell, B. 297 Blehar, M. 241, 243 Bouton, M. E.: “A Modern Learning Theory Perspective on the Etiology of Panic Disorder” 263 – 322 Breier, A. 208 Broderson, L. 248 Brown, T. A.: “A Proposal for a Dimensional Classification System Based on the Shared Features of the SDM-IV Anxiety and Mood Disorders: Implications for Assessment and Treatment” 323 – 52 Bullis, J. R.: “Evidence-Based Psychological Treatments: An Update and a Way Forward” 439 – 70; “The Nature, Diagnosis, and Treatment of Neuroticism: Back to the Future” 353 – 88

Burhans, K. K. 235 Burke, P. 234 Bush, C. 297 CALM see Coordinated Anxiety Learning and Management Cambridge Somerville Youth Study 440 Campbell, S. B. 242 Cannon, W. B. 205 Cappell, H. 279 Carey, G. 211 Carl, J. R.: “The Nature, Diagnosis, and Treatment of Neuroticism: Back to the Future” 353 – 88 Carton, J. S. 240 catastrophic misinterpretation 263, 266 – 7, 269, 282, 300, 301, 302, 303 categorical classification 331 – 2 certain bodily sensations: conditioned response is often determined by the nature of 276 – 8; effects are modulated by other stimuli 282 – 5; “modulate” other responses 281 – 2l; trigger constellations of behavior 275 – 6 Chambless, D. L. 268 – 9, 272 Champoux, M. 231 Charney, D. S. 208 Chesney, M. A. 216 children and negative emotions 234 – 9; attributional style 235 – 7; developmental considerations 235; locus of control 234; structural relations 237 – 9 Children’s Activity Inventory 237 – 8 Children’s Attributional Style Questionnaire 236 Chorpita, B. F.: “The Development of Anxiety: The Role of Control in the Early Environment” 225 – 62 Clark, D. M. 202, 266, 267, 275, 301, 303 Clark, L. A. 211, 227, 359, 368 – 9 Clarke, J. 288 classical conditioning in panic disorder 273 – 300; certain bodily sensations “modulate” other responses 281 – 2l; certain bodily sensations to trigger constellations of behavior 275 – 6; conditioned response is often determined by the nature of certain bodily sensations 276 – 8; interoceptive conditioning 278 – 81 Clements, C. M. 214 clinical utility 399 – 401 Cloninger, C. R. 202 cognitive-behavioral therapy (CBT) 366, 395, 401, 412, 416; agoraphobia 284

Index 475 cognitive enhancement procedures 445 cognitive theories 202, 247, 273, 282, 300, 301, 303; panic disorders 266 – 9 Cole, D. A. 237, 238, 253 Comer, J. S.: “Evidence-Based Psychological Treatments: An Update and a Way Forward” 439 – 70 comorbidity 228, 328 – 31, 333, 343, 355, 356, 366, 367, 369, 399, 400, 449, 452; distorted disorder 346; diagnostic 323, 328, 332, 334, 451; patterns 211, 343, 375 conditioned stimuli (CS) 268 – 9, 274 – 86, 288, 290, 291, 293, 301, 302, 304, 305, 307n2, 441 – 2 conditioning theories 266, 272, 273, 274, 275, 283, 441; agoraphobia 268 – 9 Consolidated Standards of Reporting Trials Statement 455 contextual stimuli 284 control: development of control-related cognitions 239 – 41; lack of 213, 216, 227, 229, 230, 231, 233, 236, 245, 251, 296, 354; see also Anxiety Control Questionnaire; uncontrollability Cook, E. W. 277 Coordinated Anxiety Learning and Management (CALM) 459 – 60 Costa, P. 297 Cox, B. J. 205 Craske, M. G. 289, 299 Critical Incident Stress Debriefing 458 CS see conditioned stimuli Davis, M. 240, 271 DCS see D-cycloserine D-cycloserine 449 – 50 Decola, J. 289 depression 208 – 12, 270; amygdala 357; attributional style 235 – 7, 238, 239; basal cortisol level 248, 250, 254n1; causes of anxiety and 212 – 13; childhood 243, 246, 247, 252, 253; cognitive therapy 395, 419; comorbidity 328; control 225, 226 – 7, 230, 234, 240, 242, 295, 300; dimensional classification system 335, 338, 339, 341, 342, 343; double 213 – 16; emotion suppression 365; major 240, 245, 268, 295, 331; mixed anxiety- 228, 339; neuroticism 245, 354, 356, 357, 359, 360, 361, 363, 367, 370, 371, 374; situational avoidance 364; trait 245; treatment 364, 394, 396, 400, 419; unipolar 342, 400; versus anxiety 271, 294; see also Hamilton Rating Scale for Depression de Silva, P. 362

Diagnostic and Statistical Manual of Mental Disorders (DSM) 302; II 6; III 104, 169, 170, 171, 172, 175, 177, 178, 183, 206, 207, 268, 301, 324, 366; III-R 206, 246, 268, 324, 325, 328; IV 227, 273, 323 – 52; IV limitations 332 – 3; IV-TR 356, 357, 359, 360, 365, 370, 371; V 324, 356, 360, 366, 368, 451 diagnostic reliability 324 – 8, 331, 332, 355, 366 Dienstbier, R. A. 250, 251 dimensional classification system based on the shared features of DSM-IV anxiety and mood disorders 323 – 52; alternative classification systems 333 – 9; categorical classification 331 – 2; comorbidity 328 – 31; diagnostic reliability 324 – 8; DSM-IV limitations 332 – 3; future 345 – 7; higher-order models 334 – 7; limitations to current DSM system 324 – 33; new proposal 339 – 45; nosological emphasis on shared dimensions 337 – 9; transdiagnostic treatment protocols 343 – 5 dodo bird thesis 454, 456 – 7 Dorward, J. 205 double anxiety 213 – 16 double depression 213 – 16 Drugan, R. C. 231 drug studies 403 DSM see Diagnostic and Statistical Manual of Mental Disorders Dumas, J. E. 246 Dweck, C. S. 235, 239 Dworkin, B. R. 280 Ehlers, A. 297 – 8, 301 Eifert, G. H. 287 Ellard, K. K.: “The Nature, Diagnosis, and Treatment of Neuroticism: Back to the Future” 353 – 88 emotional disorders 201 – 224; commonalities and dimensions 356 – 7; dimensional diagnosis 365 – 8; latent temperamental structure 357 – 61; nature 361 – 5; see also anxiety disorders; depression; fear; fear and anxiety reduction; negative affect; neuroticism; panic disorders; stress and anger disorder emotion-focused treatment 456 empirically supported treatments 396 – 9 evidence-based psychological interventions 431 – 8, 447; application of marketing principles 433 – 5

476 Index evidence-based psychological treatments 407 – 70; advances in the understanding of psychopathology 441 – 3; barriers 448 – 63; comorbidity, temperament, and a transdiagnostic perspective 451 – 3; current efforts 410 – 22; dissemination and implementation 409 – 10, 458 – 63; future directions 426 – 7; lack of public health 447 – 8; mechanisms of action and idiographic approaches 453 – 8; national programs 410 – 16; past 439 – 41; present 441 – 8; state programs 416 – 19; treatment augmentation 449 – 50; treatment developers 420 – 2 experiential avoidance 363, 364, 442 exposure-based treatments 441, 442, 450 extinction 269, 275, 278, 281, 283 – 4, 285, 305, 442, 449, 450, 454 extraversion, low 360, 367 Eysenck, H. J. 354, 358 – 9, 439 – 40, 460 Eysenck Personality Inventory 245

arousal 271, 360; comorbidity 356; depression 271; dexamethasone suppression 208; dimensional classification system 325, 331, 335, 338; efficacy 394; emotion suppression 365; experiential avoidance 363; genetics 293; imipramine 208; intolerance of uncertainty and distress 361; mood disorders 323; parenting style 246; positive and negative affect 451 genetic factors: agoraphobia 292 Girgus, J. S. 236 Glover, D. 289 glucocorticoid 232, 233, 251 Goddard M. J. 280 – 1 Goldstein, A. J. 268 – 9, 272, 394 Grant, P.M. 445 Gray, J. A. 208, 226, 227, 228, 230, 248, 251, 253, 271, 335, 358, 359 Greeley, J. 279 Gunnar, M. 231, 248 Guskin, K. 240

Fairburn, C. G. 394, 453 false alarms 214, 217, 272, 273, 288, 292 familial influences on negative emotions 239 – 48; anxiety development 243 – 8; attachment theory 241 – 3; development of control-related cognitions 239 – 41; neuroendocrine responding 248 Fanselow, M. S. 271 fear 205, 207, 212, 214, 215, 216, 217, 218, 226, 229, 232, 241, 251, 265, 271, 274, 276, 284, 286, 290, 294, 295, 296, 300, 301, 304, 305, 306; anxiety 202; bodily sensations 298; children 253, 299; extreme 270; loss of control 298; marketplace 265; non 266; panicing while driving 282; relevant 277, 279, 288, 293; unconditioned 272; see also agoraphobia fear and anxiety reduction: action tendencies 207, 341 fear of fear 269 Fedoravicius, A. 297 Feirig, C. 243 fight-or-flight behavior 205, 207, 216, 217, 226, 270 – 1, 272, 288, 293, 341, 358, 359 Finesinger, J. E. 287 Flykt, A. 274 Forsyth, J. P. 277, 287 Four P’s 434

habituation 233, 295 Hadzi-Pavlovic, D. 246 Hallam, R. S. 202 Hamilton Anxiety Rating Scale 209 – 10 Hamilton Rating Scale for Depression 209 – 10 Harmer, C. J. 452, 454 Harrison, B. 205 Hauch, J. 205 Hayward, C. 268, 300 health care policy, psychotherapy research, and the future of psychotherapy 391 – 406; efficacy of psychological interventions 393 – 6; objections to empirically supported treatments and the role of psychotherapy research 396 – 9; research on clinical utility 399 – 401 Heller, W. 297 helplessness 203, 208, 213, 215, 225, 227, 228, 230, 231, 235, 236, 240, 250, 294, 295, 296 Heninger, G. R. 208 Hertsgaard, L. 248 heteroreflex 280 Hodes, R. L. 277 homoreflex 280 Hugdahl, K. 287

generalized anxiety disorder 214, 215 – 16, 227, 362, 364, 367, 370, 371, 374; amygdala response 357; autonomic

idiographic approaches 449, 457 – 8 imipramine 208, 303 Impact of Events Scale 458

Index 477 insomnia 395, 444 – 5 Intensive Community Training model 459 International OCD Foundation (IOCDF) 434, 435 interoceptive conditioning 269, 272, 275, 278 – 81, 282, 288, 441, 442 IOCDF see International OCD Foundation Izard, C. E. 202, 249 Jackson, R. J. 268 Jacobsson, L. 244 Johnson, S. L. 337 Journal of the American Medical Association 443 Kagan, J. 248, 359, 368 Keller, M. B. 214, 215 Kelly, J. P. 190 Kelly, K. A. 214 Kenardy, J. 286 Kendler, K. S. 292, 299 Kessler, R. C. 292 Killen, J. D. 268 Klein, D. N. 216, 218 Konorski, J. 282 Kraemer, H. C. 268 lack of control 213, 216, 227, 229, 230, 231, 233, 236, 245, 251, 296, 354 LaFreniere, P. J. 246 Lang, P. J. 202, 203, 277 Larson, M. 248 Lavori, P. W. 215 Lê, D. A. 279 learned alarms 272 – 3 LeDoux, J. E. 274 Lee, M. A. 85, 217 Leggett, E. L. 235 Lerew, D. R. 268 Lester, L. S. 271 Lewis, M. 242 – 3 Liebowitz, M. R. 287 Lindemann, E. 287 Lindstrom, H. 244 Linehan, M. M. 421, 422 Lundqvist, D. 274 Maier, S. F. 231 Mandler, G. 213 Manicavasagar, V. 246 Margraf, J. 301 Marks, I. M. 285, 441 Masten, A. S. 249 Mayou, R. A. 458 McCauley, E. 234

McGuffog, C. 243 McHugh, R. K. 462; “Direct-toConsumer Marketing of EvidenceBased Psychological Interventions: Introduction” 431 – 8; “The Dissemination and Implementation of Evidence-Based Psychological Treatments: A Review of Current Efforts 407 – 30 McNally, R. J. 267, 274 McNaughton, N. 228, 271 mechanisms of change 372, 452, 454 Menninger Clinic 440 Metalsky, G. I. 214 Metzler, C. W. 433 Mineka, S. 212, 213, 214, 231, 233, 288 – 9; “A Modern Learning Theory Perspective on the Etiology of Panic Disorder” 263 – 322 Minor, T. R. 229, 231 Mitchell, J. E. 234, 395 modern learning theory perspective 270 – 3 Modular Approach to Therapy for Children with Anxiety, Depression, or Conduct Problems (MATCH) 453 morphine 276, 278 – 9, 442 Moss, S. 234 Mowrer, O. H. 264 Mumme, D. 240 Nachmias, M. 248 National Alliance for Mental Illness 393 National Child Traumatic Stress Network 411, 415 National Health Service 393, 402, 411, 412, 434, 445, 446, 455; “Improving Access to Psychological Therapies” 447 National Institute for Health and Clinical Excellence 411, 455 National Institute of Education 396 National Institute of Occupational Safety and Health 396 National Institutes of Health (NIH) 391, 396, 457; conference 392 negative affect 203 – 4, 211, 216, 218, 227, 231, 237, 246, 253, 268, 270, 292, 299, 300, 334, 335, 358, 361, 363, 365, 451, 452 negative emotions: action tendencies 226; anxiety and depression 226 – 7; control and anxiety 229 – 30; control in animal models 230 – 3; control in children 234 – 9; familial influences 239 – 48;influences on anxiety 228; model 226 – 9; related theoretical models 227 – 8;

478 Index see also anxiety disorders; children and negative emotions; depression; familial influences on negative emotions; fear; fear and anxiety reduction; panic disorders; stress and anger disorder neuroticism 292, 353 – 88; commonalities and dimensions among anxiety and related disorders 356 – 7; depression 245, 354, 356, 357, 359, 360, 361, 363, 367, 370, 371, 374; dimensional diagnosis 365 – 8; future research 375; latent temperamental structure of emotional disorders 357 – 61; malleability 368 – 72; nature of emotional disorders 361 – 5; prevention and treatment 372 – 5 New England Journal of Medicine 443 NIH see National Institutes of Health NIMH see National Institute of Mental Health N-methyl-D-aspartate receptor 449 nocturnal panic attacks 266, 303 Nolen-Hoeksema, S. 236, 237, 240 nomothetic approaches 449, 453, 457, 458 nonfearful panic 266 Norton, G. R. 205, 206 nosology 323, 333, 337, 345, 346, 355, 356, 365 – 6 Nowicki, S. 234, 240, 254 Nowicki-Strickland Locus of Control Scale (NSLOC) 234, 254n3 NSLOC see Nowicki-Strickland Locus of Control Scale obsessive-compulsive disorders (OCD) 360, 362, 364, 369, 374, 394, 451; D-cycloserine 450; dimensional classification system 325, 328, 331, 335, 341; ego-dystonic intrusive cognitions 341; high levels of emotion suppression 365; placebo response 456; social marketing 433, 435; see also International OCD Foundation OCD see obsessive-compulsive disorders Öhman, A. 274, 288 organizational factors 423, 461 Öst, L. G. 68, 287 Overmier, J. B. 231 overprotection 244, 246 panic attacks: potency 288 – 91; spontaneous 200, 202, 205, 206; unexpected 200, 206, 265, 268, 269, 272, 294, 295 panic disorder 201 – 2: with agoraphobia 263, 265, 285, 287, 292, 295, 325, 328,

331, 335, 344, 394, 451; D-cycloserine 449; emotion-focused treatment 456; model 205 – 7; versus anxiety 207 – 8, 270 – 2 panic disorder, modern learning theory perspective on the etiology of 263 – 322; anxiety sensitivity theory 267 – 8; classical conditioning 273 – 300; clinical evidence 285 – 8; cognitive theories 266 – 7; conditioning theories 268 – 9l; current theories 266 – 70; modern learning theory perspective 270 – 3; nonspecific biological (genetic) factors 292 – 4; nonspecific psychological factors 294 – 7; potency of panic attacks 288 – 91; relationships between learning theory perspective and other accounts 300 – 4; specific psychological factors 297 – 9; true alarms, false alarms, and learned alarms 272 – 3; vulnerabilities for development 292 – 300 Parker, G. 244 – 5, 246, 247 Patenall, V. R. A. 279 Paul, G. L. 37, 440 – 1, 460 Pavlov, I. 3, 20, 44, 79, 264, 275, 280, 307n2; conditioning 228 PDA see panic disorder: with agoraphobia pentobarbital 278 – 9, 442 Perris, C. 244 Pfammatter, M. 445 Piacentini, J. 444 Post, R. M. 217 posttraumatic stress disorder (PTSD) 324, 325, 328, 331, 332, 334, 341, 343, 356, 357, 363, 364, 365, 367, 394, 413, 414, 450, 458 Poulos, C. X. 279 psychological interventions 399, 407, 440, 448, 452, 455, 456; American Psychological Association, Task Force on Psychological Intervention Guidelines 397, 399, 402; depression in Alzheimer’s disease 444; difference between psychotherapy 446; direct-to-consumer marketing of evidence-based 431 – 8, 447; efficacy 393 – 6, 401, 402, 432; evidencebased 431 – 8, 447; manualized 396; neuroticism 375; responsiveness 369, 372; temperamental vulnerabilities 375 psychotherapy research 396 – 9 Public Health model 459 randomized clinical trials (RCTs) 393, 396, 397, 457 Rapee, R. M. 234, 289, 299, 302, 373

Index 479 Ray, J. C. 233, 254n2 Razran, G. 269, 278 RCTs see randomized clinical trials Reiss, D. 246 – 7, 267, 269 Rescorla, R. A. 283 Reznick, J. S. 248, 359 Rhodes, L. 205 Robinson, J. L. 359 Rogers, C. 440 Rosen, L. A. 290, 296 Roth, A. 395 Roth, M. 209 Roth, W. T. 301 Rotter, J. B. 213, 234, 244, 252 Ryan, B. K. 231 Ryan, S. M. 299 Sackeim, H. A. 236 safety behaviors 281, 283, 305 – 6, 364 Sanders, M. R. 433 Kirby, J. M. 433 Sanderson, W. C. 289, 302 Santucci, L. C.: “Direct-to-Consumer Marketing of Evidence-Based Psychological Interventions: Introduction” 431 – 8 Sapolsky, R. M. 232, 233, 254n2 Sauer-Zavala, S.: “The Nature, Diagnosis, and Treatment of Neuroticism: Back to the Future” 353 – 88 schizophrenia 394, 399, 440, 444, 445, 446, 454, 455, 462; see also National Alliance for Research in Schizophrenia and Depressive Disorders (NARSAD); Schedule for Affective Disorders and Schizophrenia (SADS) Schmidt, N. B. 268 Schneiderman, N. 277 Schulkin, J. 290 SD see systematic desensitization self-efficacy 142, 229, 249, 339, 367 Seligman, M. E. P. 231, 235, 236, 250, 295, 395, 399 Sengiin, S. 285 Serketich, W. J. 246 sex differences: agoraphobia 293 Shapiro, R. W. 214 Shear, M. K. 456 Siegel, S. 278 – 9 Silove, D. 246 Skinner, B. F. 240 Snidman, N. 248 social marketing 433 – 4, 436; Four P’s (product, price, place, and promotion) 434 – 5 Solomon, R. L. 264

spontaneous attacks 200, 202, 205, 206 Sroufe, L. A. 242 – 3, 245 Stangier, U. 455 State-Trait Anxiety Inventory 209 “steeling effect” 249 Strange Situation 241 stress and anger disorder 216 – 17 stressors and resilience 249 – 51; physiological arousal 250 – 1 Stroebel, C. F. 231 Strupp, H. H. 366, 440, 441 Substance Abuse and Mental Health Services Administration 415, 434 sustainability 408, 409, 413, 415, 416, 418, 421, 422, 423, 426, 460, 461 systematic desensitization (SD) 441 Szymanski, J. 433, 435 talismans 281, 340 TAU see treatment-as-usual Taylor, C. B. 268, 301 Teasdale, J. D. 214, 235 – 6 Tellegen, A. 211 Teri, L. 444 Tourette’s disorder 444, 454, 455, 462 trait anxiety 245, 246, 268, 292, 293, 294, 355, 358 treatment-as-usual (TAU) 445, 460 Triple P Positive Parenting Program 434, 435 true alarms 272, 273, 288 Tupling, H. 244 Turkat, I. D. 297 Turner, J. E. 238, 253 two-process theory 281 Type A behavior 216, 217 UCS see unconditioned stimulus unconditioned stimulus (UCS) 269, 441, 442 uncontrollability 203, 212, 213, 215, 217, 226, 231, 233, 235, 239, 251, 253, 338, 339, 360, 367 unexpected panic attacks 200, 206, 268, 269, 272, 294, 295; recurrent 265 unified protocol for transdiagnostic treatment 374 unipolar mood disorders 335, 343, 344, 361 US Department of Human Services 455 US Food and Drug Administration 444 Van Dercar, D. H. 277 Veterans Affairs Health Care System 446, 447, 455, 459 Veterans Health Administration: Mental Health Strategic Plan 413, 447, 459

480 Index vigilance 271, 273, 276; chronic 266; excessive 339; hyper 203, 342 von Knorring, L. 244 Wagner, A. R. 282, 283 Wall, S. 241, 243 Waters, E. 241, 243 Watson, D. 211, 227, 359 Watson, J. B. 264 Weiss, J. M. 231 Whitehead, W. W. 297

Williams, J. B. W. 295 Wilson, G. T. 395, 401 Winget, C. 297 Wolfson, A. 240 Wolpe, J. 441; conditioning theories 268; systematic desensitization 441 Xue, Y. X. 450 Zung Self-Rating Depression Scale 209