Reconsidering Patient Centred Care: Between Autonomy and Abandonment [1 ed.] 9781800717442, 9781800717435, 9781800717459

Citation preview

Autonomy_ Abandonment

Reconsidering Patient Centred Care

This page intentionally left blank

Reconsidering Patient Centred Care: Between Autonomy and Abandonment ALISON PILNICK University of Nottingham, UK

United Kingdom – North America – Japan – India – Malaysia – China

Emerald Publishing Limited Howard House, Wagon Lane, Bingley BD16 1WA, UK First edition 2022 Copyright © 2022 Alison Pilnick. Published under exclusive licence by Emerald Publishing Limited. Reprints and permissions service Contact: [email protected] No part of this book may be reproduced, stored in a retrieval system, transmitted in any form or by any means electronic, mechanical, photocopying, recording or otherwise without either the prior written permission of the publisher or a licence permitting restricted copying issued in the UK by The Copyright Licensing Agency and in the USA by The Copyright Clearance Center. Any opinions expressed in the chapters are those of the authors. Whilst Emerald makes every effort to ensure the quality and accuracy of its content, Emerald makes no representation implied or otherwise, as to the chapters’ suitability and application and disclaims any warranties, express or implied, to their use. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library ISBN: 978-1-80071-744-2 (Print) ISBN: 978-1-80071-743-5 (Online) ISBN: 978-1-80071-745-9 (Epub)

Table of Contents

List of Abbreviations

ix

About the Author

xi

Acknowledgements

xiii

Introduction

1

Chapter 1 What Is Patient Centred Care?

7

Chapter 2 Analysing Patient Centred Care in Practice

27

Chapter 3 On Good Interactional Reasons for ‘Bad’ Healthcare Practice

61

Chapter 4 Rehabilitating Medical Expertise for the Twenty-first Century

85

Chapter 5 Looking to the Future: Moving Beyond Patient Centred Care? 111

Appendix: Conversation Analysis Transcription Symbols (as described in Jefferson, 2004)

127

References

129

Index

149

This page intentionally left blank

For my parents, Irene and Mike, who – amongst so many other things – took me to the library.

This page intentionally left blank

List of Abbreviations

CA ESRC HCP NHS NICE NIHR PCC PCM PLWD WMO WHO

Conversation analysis Economic and Social Research Council Healthcare professional National Health Service National Institute for Health and Care Excellence National Institute for Health Research Patient centred care Patient centred medicine People living with dementia World Medical Association World Health Organisation

This page intentionally left blank

About the Author

Alison Pilnick is Professor of Language, Medicine and Society in the School of Sociology and Social Policy. She has been researching communication in healthcare for over 25 years, with the aim of improving the experiences of patients and developing and informing training programmes for healthcare professionals. This research has been funded by bodies including ESRC, NIHR, UK Department of Health, The British Academy, the Swiss National Science Foundation and the General Research Fund of Hong Kong. All of Alison’s work is underpinned by a broader desire to use sociological expertise to improve healthcare policy and practice. She was elected a Fellow of the Academy of Social Sciences in 2015 in recognition of her work in this area.

This page intentionally left blank

Acknowledgements

Many people have contributed either directly or indirectly to the work that is contained here. The analysis for and writing of this book was funded by a British Academy Senior Research Fellowship (SRF 20/200100): ‘Between autonomy and abandonment: reconsidering patient centred care’. I am grateful both for the Academy’s financial support and the space this has afforded me to finally engage substantively with an issue that I have been skating around the periphery of for a number of years, in the gaps between other projects. The first seeds of this book were planted in discussions with Robert Dingwall, who was originally my PhD supervisor, and in some of the work we have done together in more recent years. Becoming a member of the Academy of Medical Science’s FLIER programme at the end of 2019 provided further impetus, and I thank the Academy and especially my fellow FLIERS for their enthusiastic, sometimes constructively challenging but always supportive engagement with this work. This book draws on data collected for a number of projects conducted over the last 25 years, and in collecting and analysing these data for other purposes I have had the benefit of working with a number of excellent colleagues and collaborators. These include Anssi Per¨akyl¨a, Becca O’Brien, Bregje de Kok, Christian Heath, David Greatbatch, Elizabeth Murphy, Jennifer Clegg, Jon Hindmarsh, Kathryn Almack, Eric Laurier, Olga Zayts, Rebecca Allwood, Robert Dingwall, Rowan Harwood, Sarah Goldberg, Sue Widdicombe, Suzanne Beeke, Tim Coleman and Virginia Teas Gill. I have also been fortunate enough to discuss and debate some of the data reproduced in this book with a number of talented PhD students and postdoctoral researchers conducting their own research into healthcare delivery across a range of local and international contexts, including: Emma Rowley, Gregory Hollin, Helena Webb, Joe Webb, Melanie Pearce, Rachael Drewery, Rachael Finn, Ruth Parry, Sarah Atkins, Veronika Schoeb. All of these colleagues have helped me to develop and clarify my ideas, but the responsibility for any errors or weaknesses in the analysis presented here is of course mine. My experiences of the COVID-19 pandemic are those which are shared with many people, including the emotional impact of personal loss and the practical impact of prolonged homeschooling. I am therefore grateful for the

xiv

Acknowledgements

understanding of colleagues at Emerald, initially Jen McCall and latterly Kimberley Chadwick, who have patiently adjusted deadlines and rejigged production schedules in response. I hope that my family are glad to see me finally emerge from the spare room with the finished manuscript, and thank them for their ongoing love and support.

Introduction This chapter will introduce the aims and scope of the book and set the scene for the following chapters. It will also outline the way the terms patient centred care and person centred care are used in key NHS policy documents and service standards. The distinction between these two terms and the varying ways in which they are used will also be considered briefly, with an explanation of the way this, and other terminology, will be used in this book. Finally, I will present a brief overview of the contents of each of the five chapters. The NHS constitution (Department of Health and Social Care, 2021), updated every 10 years, is the document which establishes the principles and values of the NHS in England. As part of this it sets out the rights to which patients, public and staff are entitled in relation to its operation and service delivery, as well as pledges which it is committed to achieve. The document is organised around seven key principles: one of these is that ‘The patient will be at the heart of everything we do’ (NHS, 2021, Principle 4). Whilst this principle is expanded on as follows: ‘NHS services must reflect, and should be coordinated around and tailored to, the needs and preferences of patients, their families and their carers’, the terms patient centred or person centred care are not used in this document. Despite this, other NHS policy documents present patient or person centred care as the means of achieving this principle in practice. The importance attached to the principle can be seen by the way it recurs in both NHS England policy for service delivery, and in Health Education England policy for workforce training and development. Health Education England states that ‘being person-centred is about focusing care on the needs of the individual. Ensuring that people’s preferences, needs and values guide clinical decisions, and providing care that is respectful of and responsive to them’ (see e.g. https://www.hee.nhs.uk/our-work/person-centredcare); as a result there is a stated need to develop a workforce with behaviours, skills and competencies that support and drive these person-centred approaches. Patient or person centred care is also described as the way to ensure ‘the needs of the patient are central to the development of new care models’, for example in integrated care (www.england.nhs.uk/integrated-care-pioneer/resources/patientcare). In order to turn this aim into a reality, NHS England has a dedicated person centred care team. As part of the drive towards a ‘truly person centred NHS’, this team produce guidance to staff across NHS England to ‘make sure that person centred care is core in all our work’ (www.england.nhs.uk/ourwork/ patient-participation).

Reconsidering Patient Centred Care, 1–5 Copyright © 2022 Alison Pilnick Published under exclusive licence by Emerald Publishing Limited doi:10.1108/978-1-80071-743-520221001

2

Reconsidering Patient Centred Care

The extent to which these principles underpin contemporary healthcare practice in England is clear, though their pervasiveness is certainly not limited to an English or a UK context (see for example the federally funded US Patient Centered Outcomes Research Institute: https://www.pcori.org/). However, as the wording of the documents quoted above show, they are commonly invoked at a level of abstract generality. What is also clear from the wording is that, in the abstract, they are largely regarded as an uncontentious force for good, acting to curb the traditional imbalance of power in the practitioner/patient relationship. In this sense the current landscape can be seen as the logical policy end point of efforts to reform the practice of medicine which began in the 1960s, aimed at addressing the occurrence of medical dominance over, or even oppression of patients (Pilnick & Dingwall, 2011). From this perspective, it is perhaps surprising that both a 2001 Cochrane systematic review of interventions aimed specifically at increasing patient centredness in consultations, and a 2012 update of this review (Dwamena et al., 2012; Lewin, Skea, Entwistle, Zwarenstein, & Dick, 2001) found a mixed picture. While such interventions are generally successful in modifying styles of communication and sometimes succeed in increasing rates of patient satisfaction, it is much less clear as to whether they result in positive health outcomes. These findings beg the question of how a policy that seems on the face of it morally incontrovertible can fail to work in practice. I will argue throughout this book that understanding this failure requires an understanding of the way in which healthcare interactions work in practice. Whilst healthcare policy initiatives generally originate in a top-down way as a result of moral or organisational imperatives, their actual implementation often depends on them being talked into being by healthcare professionals as they interact with their patients or clients. It follows that judging the success or otherwise of these initiatives ultimately depends on examining actual interaction between patients and professionals. While it is increasingly common to consider professional-patient interaction as an evaluative tool, a central contention of this book is that it is also the means by which we can understand not just what happens in healthcare delivery, but also why it happens. The detailed analysis of healthcare interactions involving a variety of professionals and from a wide range of contexts can help us understand why attempts to deliver patient centred care have not had the impact on healthcare that might have been expected. As I will argue, the logical consequence of this is that an understanding of interaction needs to be better embedded in healthcare policy making from the outset. There are important two notes on terminology to make here. The first is on the distinction between patient centred medicine (PCM), patient centred care (PCC) and person centred care. In general terms, PCM is often used in a broad sense, to denote a shift in the way the field of medicine is conceptualised. Patient centred care is then often (but not exclusively) used to describe individual consultations or actions within this field. The terms patient centred care and person centred care are sometimes used interchangeably in the literature, with usage of the latter term

Introduction

3

becoming much more common over the last 10 years. As has been the case for descriptions of NHS policy, in some contexts more recent iterations of service descriptors and frameworks have therefore substituted ‘person’ for ‘patient’ without any broader change. However, the term ‘person centred’ has different origins and is also sometimes used in a more ideological sense, signalling a wider aim of demedicalisation and a focus on the person beyond their role as a patient. This distinction and the ideology behind it will be discussed further in Chapter 1. In order to acknowledge the presence of the distinction as drawn by some commentators, I will use the terms separately here. In keeping with the majority of the literature that I will draw on, and as the title of the book suggests, I will generally use the term patient centred care. I will reserve the use of the term person centred for discussing research or commentary which seeks to highlight the proposed distinctions. The second note relates to the fact that almost all early research in this field focuses on doctors and patients, reflecting the way doctors, and in particular primary care physicians, were historically often seen as the primary site of healthcare delivery. To some extent this focus is still reflected in current research, and so when reporting on research that has used the terms doctor and patient, they will be reproduced here. However, that does not mean the wider arguments I will make are limited to this context, and the empirical examples I will draw on include a wide range of healthcare professionals practicing in a variety of settings. Additionally, as the remit of health services have developed and widened, ever growing numbers of allied health professionals are now involved in the delivery of healthcare, and for some of these professionals the recipients of their services are generally described as a clients, rather than patients. To some extent this may reflect the philosophy of a service (for example services that align themselves to some degree with a counselling paradigm, such as genetic counselling, tend to use the term client). For other services this can reflect the type of care that is delivered, for example in preventative healthcare services where a person is not currently (and may never become) ill. This book is intended to speak to both patients and clients, and to healthcare professionals beyond the doctors who feature in early research. My own data come from a wide range of healthcare settings, and for practical purposes, when I am referring to these data I will use whatever term the service where the data were collected chose to use. The remainder of this book is divided into five chapters: Chapter 1 will trace the history of patient centred care, and explore some of the varying ways in which it has been defined, including its relationship with the concept of Shared Decision-Making. It will move on to show how, despite the fact it is presented as the logical end point of reform to the doctor-patient relationship that has been ongoing since the 1960s, there is little clear evidence for its impact on healthcare. The promotion of PCC therefore rests on a moral position rather than on its outcomes. I will argue that understanding the failure of PCC to produce its expected impact requires a detailed understanding of the way in which healthcare interactions work in practice, which has hitherto been absent from healthcare policy.

4

Reconsidering Patient Centred Care

Chapter 2 will build on the conclusions of Chapter 1, highlighting that whilst many healthcare policy initiatives originate in a top-down way as a result of moral or organisational imperatives, their implementation often depends on being talked into being by healthcare professionals. I will introduce the analytic approach of Conversation Analysis that is used throughout this book, and show how analysing data from this perspective can begin to help us understand why it is that PCC has not had the positive impact on health outcomes that might have been expected. I will illustrate this argument with examples of the way choice, care and control are negotiated and enacted across a range of healthcare settings. In the second part of this chapter, I will examine the way in which checklist-based conceptions of PCC generally assume that ‘good’ and ‘bad’ communication can be identified a priori and acontextually. However, CA analyses show that there can be good organisational reasons for what may on the face of it look like ‘bad’ healthcare practice, and that practices which are assumed on the basis of everyday interaction to be inclusive, supportive and patient centred may not operate this way in a healthcare context. Chapter 3 will continue to develop the analysis presented in Chapter 2. However, while Chapter 2 illustrates a variety of ways which might lead to individual ‘failure’ to enact PCC, rooted in competing organisational imperatives, and ‘dysfunctions’ (Heritage, 2011) in the transfer of practices from ordinary conversation to healthcare interaction, in this chapter I will argue that there is a more fundamental reason why policies such as PCC are not enacted in practice in the way we might expect. Despite the fact that many healthcare policies have to be talked into being by healthcare staff, policies are not generally grounded in an understanding of how interactions actually work. Drawing again on real-life examples from my data corpus, I will show how some of the policies and practices used in the service of PCC result in interactional conflict because they require healthcare professionals to manage competing interactional norms. This means that, while these policies and practices may make organisational sense, or make moral sense from a particular perspective, they do not make good interactional sense. In other words, there can be good interactional reasons for what may look like ‘bad’ healthcare practice. I will argue that this problem will not be resolved simply through ‘better’ training in PCC but instead requires an acknowledgement and understanding of why and how these interactional conflicts arise in healthcare interaction. In part, this is linked to the way the conceptualisation and delivery of PCC can problematise professional expertise. Chapter 4 will consider the issue of medical expertise in more detail, outlining the changing ways it has been understood sociologically in order to understand how the healthcare professional/patient encounter has come to be viewed as a struggle for control. I will use data from healthcare interactions to show the ways in which medical expertise is collaboratively constructed, and how absolute prioritisation of the ‘epistemics of experience’ (Heritage, 2013) over this can ultimately lead to patient abandonment or to care which falls short of professional standards. A failure to understand this, because of a failure to understand how healthcare interaction works in practice, means that practices and policies

Introduction

5

introduced with the intention of empowering patients or improving their care can actually have the opposite effect. In the final chapter of this book (Chapter 5), I will consider what the consequences of my critique might be for a way forward. I will suggest that the pervasiveness of PCC needs to be understood in the context of its relationship with consumerism, and consider whether it is possible to envisage a reimagining of PCC or whether it is better abandoned altogether. In considering this, I will highlight the problematic way in which social science has been co-opted to serve a sociology in medicine rather than a sociology of medicine (Straus, 1957), and suggest that a critical sociological engagement allows us to propose alternative possibilities beyond more or ‘better’ training in PCC. Finally, I will return to a theme that runs through all the chapters of this book: that healthcare policies generally depend on interaction between individuals for their enactment; as with PCC, they have to be ‘talked into being’. This means that understanding why policies may not work as intended requires an understanding of human interaction. I will argue that this ultimately points to a need to consider to what extent an understanding of interaction can be used to inform policy making from the outset, rather than simply using it as a tool to measure the success or otherwise of its implementation. In concluding this introduction, it is important to note here that this book draws on research I have conducted over a 25-year period, funded by a variety of UK and international funders. I have provided more details on these projects in Chapter 2, as well as referencing some of the publications that have arisen from them in support of analysis throughout the book. None of these previous projects focused specifically on patient centred care, but in collecting and analysing these data for other purposes the context of PCC and the impact this has on healthcare interaction has always been present on the periphery. This book represents my attempt to place PCC at the centre of my work and to produce a substantive critical engagement with it.

This page intentionally left blank

Chapter 1

What Is Patient Centred Care? This chapter will begin by tracing the history of patient centred care (PCC), setting this against a backdrop of changing sociological conceptualisations of the doctor/patient relationship. It will highlight the fact that there is no universally accepted definition of PCC, and explore some of the differences in widely used definitions. I will move on to show how a lack of conceptual and definitional clarity have led to difficulties in measuring the success or otherwise of PCC, but that even allowing for this there is no clear picture of its positive impact. The available empirical evidence suggests that, in general terms, adoption of PCC can lead to improved patient satisfaction, but not to improved health outcomes. Despite this, most empirical research into PCC assumes the primary problem is individual or service level failure in its enaction, rather than questioning the principle itself. I will consider how this lack of critical interrogation of the concept is due to the fact that PCC is presented as a moral position, and in this sense can be seen as the logical policy end point of efforts to reform the practice of medicine which began in the 1960s, aimed at addressing the occurrence of medical dominance over, or even oppression of patients (Pilnick & Dingwall, 2011). In the latter part of the chapter I will examine the relationship between PCC and shared decision-making (SDM), and reflect on some of the concepts that underpin both terms, particularly in relation to the prominence given to patient autonomy and individual choice. Finally, I will consider the distinctions that are sometimes made between patient and person centred care, and the ideology that lies behind the choice to use either of these terms. This distinction will be set in the context of a wider critique of the way the ‘vocabularies of the therapeutic’ (Rose, 1999, p. 218) have increasingly found their way into the mainstream. I will conclude by arguing that in order to understand why PCC has not had the impact on health outcomes that might have been expected given its ‘moral rightness’, we need to examine the way healthcare interaction works in practice.

Tracing the History of Patient Centred Care One of the difficulties of any substantive engagement with patient centred care is that there is no universally accepted definition of the concept. What is generally agreed is that it was first introduced by the psychoanalyst Michael Balint in 1955–1956, as part of his work with UK General Practitioners. However, even at the outset, the concept was largely defined not by what it was but by what it was Reconsidering Patient Centred Care, 7–26 Copyright © 2022 Alison Pilnick Published under exclusive licence by Emerald Publishing Limited doi:10.1108/978-1-80071-743-520221002

8

Reconsidering Patient Centred Care

not, described and explicated as a contrast to the reductionist, biological approach of illness centred medicine. Underpinned by the philosophy of holism, Balint’s approach instead centred the idea that patients needed to be seen in their biopsychosocial entirety, so that doctors should consider the ‘pathology of the whole person’ (Balint, 1964/1957, p. 7). As both he and his wife, Enid Balint, noted, they were not always faithful to the specific term, occasionally using the term ‘patient-oriented medicine’ as an alternative (Balint, 1969; Balint et al., 1969). Whichever term was used, within this approach, and consistent with Balint’s psychoanalytic background, relational matters were paramount, since it was ‘Not only…the box of pills that mattered, but the way the doctor gave them to his patient’ (Balint, 1964/1957, p. 1). Balint urged the practitioners he worked with to consider the importance of listening, particularly when in doubt about how to act, reminding them that they were participants in ‘a peculiar, lopsided two person relationship’ (1964, p. 252). This lopsided nature, and the strength of the doctor’s position relative to the patient’s, was grounded in an asymmetry of medical knowledge, allowing the doctor the ability to explain and interpret, in contrast to the patient who came to see the doctor precisely because they could not understand or resolve the problem themselves. This asymmetrical view of the doctor–patient relationship can be traced back, sociologically speaking, to the work of Talcott Parsons in the 1950s. As a structural-functionalist, Parsons’ interest in sickness was grounded in the way it represented, for him, a form of deviance: a sick person was not a productive member of society because they could not fulfil their normal social roles. In order for societies to function, there needed be a recognised mechanism whereby people could be exempted from or returned to their normal roles. For Parsons, the management of illness is therefore a mechanism of social control for a capitalist society. In setting out his model of the sick role (Parsons, 1951), Parsons identifies three significant features of patients in this role: they are helpless; they do not possess the technical abilities to solve their own medical problems; and they are emotionally involved with their own illnesses, so that they cannot be relied upon to act rationally. Indeed, as Armstrong (2014) observes, for Parsons the very nature of patienthood was such that the sick were not competent to help themselves. Though the sick role model absolved patients of the responsibility for their illness through a recognition that they could not avoid or mitigate it themselves, it also presented them with a responsibility to submit to professional help. This help seeking necessarily placed them in a dependent position, with little space for independent action. As subsequent generations of sociologists have argued, the ultimate implication of Parsons’ model is the potential that a person becoming a patient also becomes vulnerable and docile, with ‘no responsibility for their predicament and minimal involvement in their own care’ (Armstrong, 2014, p. 164). In Parsons’ defence, some commentators have argued that over time, and as a result of his focus on the social contract in relation to health, an overly structural version of his work has become popularised (Rawls & Turowetz, 2021). It should also be noted that his conception of physicians’ professional power and autonomy encompassed a characterisation of the physician as at all times using their

What Is Patient Centred Care?

9

expertise to act in the patient’s best interests, an issue to which we will return in Chapter 4. Nevertheless, the idea of patient dependency and passivity became problematic with the changing nature of medical practice, and a changing understanding of the field of medicine itself. Advances in the development of screening programmes, a greater awareness of asymptomatic disease, and the wider conceptualisation of medicine as an endeavour that was not purely responsive to symptoms were contexts for which the typicalised Parsonian patient was ill-fitted. The 1950s model of the doctor who could make decisions with little need for explanation, and the patients who would act in accordance with them (see Fox, 1957, 2000). was no longer universally functional. In these scenarios, as Armstrong (2014) has argued, there needed to be a promotion of more autonomous action on the part of (would be) patients, which would in turn represent a rebalancing of the doctor–patient encounter. Szasz and Hollender (1956) were physicians who built on the work of Parsons to refine different models for doctor/patient interaction, grounded in the different contingencies of different types of illness. Their three models (activity-passivity; guidance-co-operation; and mutual participation) reflected a recognition of different medical contexts, and the fact that a one-size-fits-all approach could not be applied across them. Activity on the part of the doctor and passivity on the part of the patient was deemed necessary in some contexts, for example an emergency situation or an anaesthetised patient. In the guidance-co-operation model, the physician would ultimately tell the patient what to do and the patient would comply, but both parties were conceptualised as active and contributing to the relationship. In the mutual participation model, the relationship was conceptualised as that of a partnership, where ‘the physician helps the patient to help himself’ (sic) (Szasz & Hollender, 1956, p. 522). This model was envisaged as particularly applicable to chronic illness and psychoanalytic contexts, where the patient’s knowledge and experience could be used as a resource. This recognition of different relationships according to different medical contexts and different levels of patient need presented a more nuanced picture than the Parsonian model. However, with the exception of the activity-passivity model, it also transferred a level of responsibility to the patient. For the system to operate as the new services and interventions intended, a patient became someone who needed to have a level of attentiveness so as not to miss a symptom that might be a signifier of something serious, but who at the same time would not be a waster of valuable medical resources (Mechanic, 1962). Moving in to the 1970s, the shift of analytic focus from isolated roles of doctor and patient to the doctor-patient relationship continued. Most notably, the work of Patrick Byrne (a GP) and Barrie Long (a social scientist) shifted the focus of discussion away from the theoretical and into the empirical. The publication of their work ‘Doctors Talking to Patients’ in 1976 was the culmination of a research project for which 60 GPs had been invited to audio-record their surgeries, resulting in a corpus of over 2,500 consultations. Byrne and Long’s observations on the recordings included categorising doctor behaviours into whether they were deemed to be patient or doctor centred. As a practical project, GPs were then invited to score their behaviour against these two categories, in order to identify

10

Reconsidering Patient Centred Care

areas where their practice could be improved. The inherent assumption here, then, was that patient centred behaviours were to be aspired to and encouraged, while doctor centred ones could be the site of this improvement. Examples of doctor centred behaviours given by Byrne and Long (1976) included the use of closed questions, and the doctor directly giving information or opinions. The corresponding patient centred behaviours were identified as using broad questions or openings, and giving information by answering patient questions. In some contexts, Byrne and Long also identified negative behaviours: in the example given here, the negative behaviour was defined as evading patient questions. Byrne and Long’s (1976) work was a pivotal moment in the transformation of the concept of the patient from the Parsonian passivity of the 1950s to the active and empowered idealisation of the twenty-first century. What emerges from their view of the consultation is a vision of a healthcare provider guided by the knowledge and experience of the patient. However, it is also important to note that they did not view this guidance as without limits. In terms of determining a course of action such as a treatment, Byrne and Long conceptualised a scenario in which the doctor defined the limits, and requested the patient to make a decision within these limits: what might best be described as a kind of managed autonomy. I will return to the centrality of patient autonomy later in this chapter, in order to consider both the impact on patient centred medicine that follows from different ways of conceptualising this autonomy, and the prominence given to it. Contemporaneously with Byrne and Long, Stimson’s (1974) work on non-compliance – why patients did not take their medicines as doctors directed them to do – explored this autonomy from a different perspective. Disturbing the assumption that non-compliance was due to forgetfulness or a failure to understand instructions, Stimson showed how patients did not make decisions about medication in a social vacuum, but instead based on their own ideas and beliefs. In most cases, to not take a prescribed medication was a rational and deliberate act. This recognition of patient agency in following medical advice came to be reflected in the replacement of the term non-compliance with concordance, where a lack of concordance is seen as reflecting the failure of the consultation rather than a failure of the patient (Armstrong, 2014). Conceptualising concordance as opposed to compliance is another way of placing the emphasis on the professional-patient relationship, and of placing an equal emphasis on the perspective of the patient. As Armstrong (2011, 2014) documents, through the 1970s and into the 1980s, new aspects of patienthood continued to emerge, with a focus on patients’ individual identities and the reason for the encounter. This included Mishler’s (1984) influential work in which he drew on Habermas’ (1981) theory of communicative action, applying this to a corpus of medical encounters collected in the United States in the mid-1970s. Mishler conceptualised medicine as a conflict between the ‘natural attitude’ of patients vs the ‘scientific attitude’ of medicine, which ultimately resulted in the consultation as a contest between the voice of the lifeworld of the patient and the medical world of the doctor. From the perspective of the lifeworld, moral considerations defined the ends of the encounter, and so there was a desire for understanding. From the perspective of the medical world,

What Is Patient Centred Care?

11

technical considerations defined the end goal, and so there was an orientation to success. The consequence of Mishler’s analysis is that the doctor/patient encounter is viewed as a constant struggle, and through this struggle patients’ accounts and their meaningful experiences are suppressed; he goes so far as to suggest that this is not only ineffective medical care, but it is also inhumane. As my colleague and I have argued elsewhere (Pilnick & Dingwall, 2011), this ‘critical’ tradition recognises the social and organisational embeddedness of the clinician/patient encounter, and explains the lack of mutuality that is observed as a function of the clinician’s role in relation to capital (e.g. Mishler, 1984; Waitzkin 1991) or patriarchy (e.g. Fisher, 1995; Oakley, 1980; Todd, 1989). As a result, the ‘fundamental reconstruction of patients’ identity that began in the second half of the 20th Century’ (Armstrong, 2014, p. 163) took place within a set of wider political and ideological beliefs, with the result that medicine came to be seen as paternalistic and therefore problematic. And while the work of Mishler et al. was largely grounded in qualitative analysis, at the same time a broader empirical evidence base for these assertions was being developed through quantitative work in the ‘code and count’ tradition, applying analytic frameworks drawing on Bales’ (1951) Interaction Process Analysis. Though this originates in the work of Korsch and associates (e.g. Korsch & Negrete, 1972; Korsch, Gozzi, & Francis, 1968) it found more prominence in the Roter Interaction Analysis System (RIAS) (Roter, 1977; Roter & Larson, 2002), and studies using this system to code healthcare interaction still find that the majority of consultations examined centre on the clinician’s accomplishment of biomedical tasks (e.g. Innes, Skelton, & Greenfield, 2006). In response to the now accepted problem of paternalism, from the 1980s onwards patient centred medicine began to be promoted as an approach in its own right, rather than as a component or a product of broader approaches to medicine. The approach was developed by Joseph Levenstein and colleagues working in the family medicine dept at University of Western Ontario, Canada (Levenstein, McCracken, McWhinney, Stewart, & Brown, 1986). As they describe it, it has many foundational similarities with Mishler’s (1984) work; they present a picture of competing agendas, where the physician’s agenda is the explanation of the patient’s illness in terms of a taxonomy of disease, whereas the patient’s agenda is informed by their expectations, feelings and fears. However, while Mishler was content to demonstrate the mismatch and offer a critique of the wider structural factors which led to this, Levenstein and colleagues propose patient-centredness as a clinical method, whereby the two competing agendas can be addressed by the physician and any conflict between them dealt with by negotiation. The patient centred method is therefore contrasted with the disease centred method in which only the doctor’s agenda is addressed. Whilst the observation of this contrast could already be found in the sociological work previously discussed in this chapter, and while Byrne and Long’s (1976) work began to identify patient centred practices, Levenstein et al. (1986) argue that their aim is to operationalise the concept. Key to this operationalisation is the narrative competence of the clinician, which is seen as necessary to elucidate the patient’s agenda.

12

Reconsidering Patient Centred Care

From Levenstein and colleagues’ initial work, there has stemmed a range of attempts to further specify the characteristics of patient centred medicine. However, despite the existence of the term ‘patient centred’ since the 1950s, and despite a range of research which points to its absence in practice, there is little clear consensus regarding the meaning of the term. A range of definitions have been developed in the academic and clinical literature (e.g. McWhinney, 1989; Mead & Bower, 2000; Stewart et al., 1995). They use Byrne and Long’s (1976) description of patient centred medicine as an approach whereby the healthcare provider uses the patient’s knowledge and experience to guide the interaction within the consultation as a starting point. For McWhinney (1989, p. 25) the patient centred approach is one where ‘the physician tries to enter the patient’s world, to see the illness through the patient’s eyes’. Stewart et al. (1995) provide a more detailed description, identifying six related components: (1) exploring both the disease and the illness experience; (2) understanding the whole person; (3) finding common ground regarding management; (4) incorporating prevention and health promotion; (5) enhancing the doctor-patient relationship, and (6) ‘being realistic’ about personal limitations and issues such as the availability of time and resources. In a narrative review of the variety of ways in which patient centred care has been defined in the literature, Mead and Bower (2000) identified five dimensions, present to varying degrees across different definitions: (1) biopsychosocial perspective – the extension of the scope of medicine from the purely biological to the psychological and social levels; (2) the ‘patient-as-person’ – understanding the individual’s experience of illness; (3) sharing power and responsibility; (4) the therapeutic alliance – developing a professional–patient relationship based on care, sensitivity and empathy; (5) the ‘doctor-as-person’ – self-awareness and attention to emotional cues in the profession–patient relationship. Despite these varying definitions, and the continued proposal of further dimensions to the concept, it has become widely endorsed and embedded in official policy globally: for example, it is part of the first pillar of the WHO’s End TB strategy (WHO, 2021), and the World Medical Association’s (2020) Declaration on the Physician/Patient Relationship embeds the assumption that the end goal of this relationship is to deliver patient centred care. These policies also illustrate that the term ‘care’ is now frequently used rather than ‘medicine’ even when denoting a broader field rather than an individual interaction; this linguistic substitution both widens the scope of the term beyond specifically medical treatment, and also has applicability to a wider range of professionals who might be involved in delivering that treatment. However, and whichever term is used, different stakeholders attach importance to different aspects of PCC, and have varying views about whether its practice goes beyond the professional/patient interaction. For example, the UK-based charity the International Alliance of Patient Organisations includes in its conceptualisation of patient centred healthcare an expectation that patients should be involved in the formulation of healthcare policy (www.iapo.org.uk/what-we-do). Another UK charity, The Health Foundation, which has a significant role as a funder of UK health services research, emphasises the importance of a transfer of control from doctor to

What Is Patient Centred Care?

13

patient (https://www.health.org.uk/newsletter-feature/a-portal-for-resources-andlearning-on-person-centred-care). Other research funders, such as the US based, federally funded Patient Centred Outcomes Research Institute, which was constituted following provision in the Affordable Care Act (2010), exist specifically to fund research to establish the most effective PCC interventions. Their definition of patient centred outcomes research (https://www.pcori.org/researchresults/about-our-research/patient-centered-outcomes-research) puts an emphasis on outcomes that patients and other stakeholders identify as important, although given the breadth of potential stakeholders it is unclear what alternative outcome measures might be. It is perhaps not surprising then, that the twenty-first century has come to be identified as the century of the patient, as contrasted with the twentieth century which has been conceptualised as the century of the doctor and the medical industry (Buetow, 2016; Gulland, 2011). Given the lack of consensus into what, specifically, constitutes PCC, it is unsurprising that examining research into the delivery of PCC reveals a variety of approaches, where definitions are sometimes imposed a priori and sometimes developed as a result of empirical research, e.g. by asking patients what they think is important. In Stewart’s (2001) British Medical Journal editorial with the ambitious title ‘Towards a global definition of PCC’, she asserts that the best way of measuring PCC is an assessment made by the patient themselves. However, this raises significant methodological issues. The first is that any such assessment cannot be assumed to apply beyond that patient, which makes it difficult to establish any general standards or principles. Assessment is further complicated by a situation in which PCC is more commonly understood in terms of what it is not – disease centred, technology centred – then what it is. As the literature on healthcare complaints suggests (see Reader, Gillespie, & Roberts (2014) for a systematic review), patients may generally find it easier to articulate what is wrong with their care than what is right, and do not necessarily frame these complaints in ways that orient to improving the delivery of care. In practice, studies of PCC veer between attempting a direct assessment of patient centredness, and using a variety of scales and measures to assess proxy measures such as patient perceptions and health outcomes. Indeed, The Health Foundation (2014) identifies the existence of 160 commonly used measuring tools. To Stewart’s (2001) credit, she argues that qualitative research comes closest to being able to articulate its qualities, presumably (although she does not specifically say so) because of the difficulties of assessing it outside of its context, an issue to which we will return in Chapter 2. She concludes that patients want PCC which explores the main reason for the visit and their concerns, takes on board an integrated understanding of their world, and finds common ground in problems; she argues that delivery of PCC in this way should also support the ongoing doctor/patient relationship. Like the principle of PCC itself, this list of patient wants might appear on the surface to be unimpeachable. However, the list also assumes that these things are straightforwardly achieved, and so glosses over the interactional work required for this achievement. Finding common ground, for example, might be a goal to aspire to, but will potentially be more difficult to enact in practice, depending on the problem under discussion and the solutions that are available. Stewart’s (2001)

14

Reconsidering Patient Centred Care

framing also appears to assume a diagnosis/treatment-oriented interaction, which may work well for consultations in primary care but may be less applicable outside of this context. However, these considerations are often absent from discussions of PCC. From her extensive review of (largely US based) patient centred initiatives, Tanenbaum (2015, p. 272) concludes that the meaning of the term is at the same time contested and treated as obvious. In aiming to construct a typology of these initiatives, she notes the existence of internal contradictions which undermine the coherence of the term, including whether patient centred care is to be considered a process or an outcome. Buetow (2016) takes this argument further, suggesting that, notwithstanding the tensions that Tanenbaum identifies, the term PCC has now become so widely used as to become meaningless. What is certainly indisputable is that the large and growing number of practices and policies that describe themselves as patient centred are not always compatible, and that if this presents a challenge for practitioners to navigate it must also present one to patients. It follows that this difficulty in defining PCC carries through into evaluations of its effectiveness.

Evidence for the Effectiveness of PCC Given the dominance of the approach in healthcare policy, individual studies into the effectiveness of PCC present much more of a mixed picture than might be expected. There are studies, such as Stewart et al. (2000), where findings have shown a link between patient-centred practice and positive health outcomes, but there are also others, e.g. Kinmonth, Woodcock, Griffin, Spiegal, and Campbell (1998) where adoption of patient centred practice leads to clear negative effects. More recent work (Goldfarb, Bibas, Bartlett, Jones, & Khan, 2017 Ja´en et al., 2010; Lee & Lin, 2010) highlights that the assumed link between patient-centredness and improved patient outcomes remains elusive. Studies sometimes demonstrate improvement in patient satisfaction or patient experience as a result of PCC, but even this is not universal (e.g. Ja´en et al., 2010). The two most comprehensive attempts to assess the impact of patient centred care interventions are found in Lewin et al.’s (2001) Cochrane systematic review, and Dwamena et al.’s (2012) update of this. Lewin et al.’s (2001) original review found some evidence for improvement in patient satisfaction, but none for improvement in healthcare behaviour or health outcomes. They note that there is ‘considerable heterogeneity’ in terms of both the interventions that are badged as patient centred care and the health contexts they are used in, which, given what is known about the importance of understanding context in healthcare (Matthias, Salyers, & Frankel, 2013) makes it problematic to synthesise this body of work to produce general findings. In their 2012 updated review, Dwamena et al. also note that in the time between the two reviews, patient centred approaches have become increasingly advocated by consumers and clinicians and incorporated into training. The choice of terminology here, in the use of the term ‘consumer’ rather than ‘patient’, highlights the pervasiveness of consumerism in healthcare and illustrates the way

What Is Patient Centred Care?

15

in which the medical encounter has been reconceptualised around patient choice, a theme to which I will return briefly below and at more length in Chapter 5. Interestingly, for the purposes of this updated (2012) review, the presence of shared decision-making (SDM) alone is taken as an official indicator of PCC. Results from the updated review show positive effects on the consultation process, but mixed results on patient satisfaction, behaviour and health status. As with Lewin et al.’s (2001) original review, the greatest benefits are found from complex interventions, notably those that focus on patients and providers with condition-specific materials. However, even these show benefits in health behaviour and satisfaction, but mixed effects on health status. Dwamena et al. (2012, p. 24) also note that the greater success of complex interventions means it is a challenge to identify what the ‘active’ elements of PCC might actually be in these cases. Their findings lead them to ask the question as to whether ‘results [are] sufficient to justify the importance that PCC has taken on in training programmes in Europe, the UK and North America?’, concluding that the answer is yes if PCC is seen as worthy in its own right, as a moral position. They also emphasise what might be called the circularity of PCC: that research aimed at training staff in the delivery of PCC improves the delivery of PCC as measured by research. However, it fails to consistently improve any patient related variables such as behaviour, satisfaction or outcomes. Rather than demonstrating the effectiveness of PCC, then, these reviews instead illustrate how it has become a mark of quality of care in the absence of evidence for how it improves that quality of care. The issue of the apparently greater success of complex interventions found in both Lewin et al.’s (2001) review and Dwamena et al.’s (2012) update also highlight another issue with the evaluation of PCC. When PCC is reported as being effective on health outcomes, it is often components of PCC that are being reported. For example, quality of communication has, unsurprisingly, been found to affect health outcomes (Stewart, 1995), but this is not necessarily dependent on a PCC approach. The variety of ways in which PCC is operationalised in research risks a situation where the success of any component in any context is held up as evidence of the wider approach. What examination of these reviews highlights, then, is that what PCC interventions are most demonstrably effective at is training staff to act in accordance with those interventions. Given the lack of evidence for any impact on health outcomes, the findings also raise a wider, but very important question about what we consider the purpose of medicine in society to be. If it is assumed to be to improve health, then the evidence that PCC does this is at best patchy and condition specific, and at worst absent. It might be expected that these mixed findings would have led to broader questioning of the principle of PCC. Instead, as Pilnick and Dingwall (2011) have described, these findings are generally treated as the result of educational or motivational deficits among healthcare professionals, which will be remedied by more or better skills training in the practice of PCC. However, such efforts have been widespread in medical education since the late 1970s (see Pendleton, 1983), and PCC has been explicitly integrated into the now longstanding and influential Calgary-Cambridge consultation training model devised by Silverman, Kurtz, and Draper (1998) and Kurtz, Silverman, and

16

Reconsidering Patient Centred Care

Draper (2004). Despite this, as Pilnick and Dingwall (2011) have shown by drawing on published primary care consultation data, these consultations have proved remarkably resistant to change. However, this lack of change has also failed to precipitate any questioning of the strategy.

Shared Decision-Making and Patient Centred Care It is worth pausing here to consider the relationship between SDM and PCC. Since not all healthcare encounters involve a treatment decision, it may seem on the face of it that SDM has little to tell us about the wider operation of PCC. However, SDM is commonly considered a key component of PCC. Indeed, it is sometimes considered the defining component: in Dwamena et al.’s (2012) systematic review discussed above, the presence of SDM was on its own taken as a sufficient indicator of PCC. SDM began to appear as a concept in the 1970s and became more prominent from the 1980s onwards, to the point where a New England Journal of Medicine article (Barry & Edgman-Levitan, 2012) described it as the ‘pinnacle’ of patient centred care It is also embedded in UK National Institute for Health and Care Excellence guidance (NICE, 2021). However, just as there are problems with defining PCC itself, so there are problems in identifying consensus as to the overlap. Stevenson, Barry, Britten, Barber, and Bradley (2000) argue that SDM has a number of similarities with patient centred medicine, but that the former contrasts with the general approach of the latter by putting the focus squarely on the patient’s active involvement in the treatment decision. Perhaps the most well recognised definition of SDM is that put forward by Charles, Gafni, and Whelan (1997): a process where a patient (or his/her surrogate) and a clinician share information and take deliberate consensus building steps to reach agreement about a plan for testing/treatment. There are four components to the model Charles et al. (1997) propose: (1) both the doctor and the patient are involved; (2) both parties share information; (3) both parties take steps to build consensus about preferred treatment; (4) an agreement is reached on the treatment to be implemented. As Charles, Gafni, and Whelan (1999a, 1999b) subsequently expand, this sharing of information as originally envisaged in their definition is intended to go both ways, since ‘both patient and doctor have a legitimate investment in the treatment decision’. Framing the encounter as a struggle for control, as some critical analysts have done, therefore risks downplaying the doctor’s legitimate interest. However, as Kaldjian (2017) notes, more recent frameworks (e.g. Elwyn et al., 2012) tend to emphasise the empowerment of patients. In explicating this, Kaldjian (2017) identifies the potential conflict between the ends-based and purpose-oriented nature of healthcare, alongside its contemporary location within a landscape that connects treatments and tests with goals of care and values. This purpose-oriented nature means that there is a risk of biasing decisions in relation to interventions that reflect what is available or can be accessed quickly. The solution, he suggests, is to set ‘goals of care in the foreground so that the direction of decision-making is clear before too much focus is placed on interventional options’ (Kaldjian, 2017, p. 83). These goals might encompass cure but they might equally, for example, encompass comfort or

What Is Patient Centred Care?

17

improved function. He argues that a focus on goals is the means through which we can ensure that SDM becomes an ethical rather than scientific process. However, and as he acknowledges, goals are unlikely to be static, and nor are concepts of health and disease; this acknowledgment begins to lend SDM some of the amorphous qualities we have noted in relation to PCC above. In an attempt to address this issue, Makoul and Clayman (2006) carried out a systematic review of 418 articles on SDM, selected through a PubMed search, to identify concepts commonly used in definitions of SDM. They found that no single concept was listed in all these articles, and that 20 concepts appeared in more than 10% of the papers reviewed. The most commonly occurring concepts that they found were patient values/preferences, options and partnership. However, prominent within the literature was also a recognition of the need to make (or to explicitly defer) a decision based on the elicitation of patient values/ preferences, and the available healthcare options. The centrality afforded to making a decision based on patient values and preferences, enshrined in the NHS policy quoted in the Introduction to this book, has also been reflected in the development of instruments explicitly designed to address the quality of shared decisions. While a range of these exist, the two most prominent are perhaps those devised by Braddock et al. (1997) and Elwyn, Edwards, Wensing, and Grolt (2005). Braddock and colleagues’ model is grounded in the bioethics literature, and specifically in ethical models of informed consent. Drawing on this literature led them to identify six key elements which can be graded according to their presence or absence in the discussion leading to each decision. For the purposes of this model, an element is counted as either present or absent, i.e. it is considered to be present if it is discussed at all, however brief that discussion might be. However, as critics of the model have highlighted, the presence of individual actions such as providing an alternative are not necessarily factors which definitively indicate whether a decision has been shared. A patient who is given a clear set of alternatives may ultimately still be steered towards a practitioner’s preferred option (Brown, Butow, Butt, Moore, & Tattersall, 2004), and there is an existing body of detailed interactional work which shows how other factors may undermine a presentation of choice such as the location of that choice in the wider consultation (Pilnick, 2008) and the way in which choices themselves are presented (Toerien, Shaw, Duncan, & Reuber, 2011; Toerien, Shaw, & Reuber, 2013; Toerien, Reuber, Shaw, & Duncan, 2018). Elwyn et al.’s (2005) OPTION scale approaches SDM from a different perspective, providing a ‘skills framework’ intended by the authors to represent a set of competences as opposed to a more discrete list of items. This skills framework is operationalised through a 12-item scale, where each item is scored so that an overall numerical score can then be assigned to the consultation. The scoring scale allows distinctions to be made between how different options are presented to and explored with the patient, but also aims to take into account the level of decisionmaking that the patient requires. Other existing models (e.g. the MAPPIN’SDM ¨ (Kasper, Hoffmann, Heesen, Kopke, & Geiger, 2012) use different items, but similar scoring systems (typically grading an item from no effort to exemplary effort, totalling these scores and converting them into percentages, in order to

18

Reconsidering Patient Centred Care

produce a mean score for a consultation). However as Landmark, Ofstad, and Svennevig (2017) note, there is no meaningful comparison between many of the items scored across even these two scales. Despite the different ways of measuring SDM, the existence and widespread use of these different scales illustrates that there is a significant literature which attempts to assess and grade the actual mechanics of decision-making. This literature also demonstrates that the issue of decision-making is often conceptualised as having a straightforward relationship to treatment. However, this link is variable according to the healthcare setting and the purpose of the encounter. Notwithstanding this variability, tools for measuring SDM are not generally badged as context specific. The importance of context in SDM is highlighted by Matthias et al. (2013) who also make a distinction between internal context of the consultation, and external context or healthcare setting in which it takes place. Considering the former, they highlight how studies of SDM have frequently focused only on the point where a decision is made in a consultation. They also problematise the fact that scales designed to evaluate the nature and quality of shared decisions, such as the ones described above, tend to focus only on the sections of the consultation where decisions are actually made. However, they argue that no part of an interaction can be fully understood in isolation, and that what happens in the decision-making process itself cannot be extricated from the local context of what has happened before and what happens after. As I have shown in the context of antenatal screening for fetal anomaly (Pilnick, 2008), even a test which is clearly presented in that moment as a choice can have this presentation undermined by its location subsequent to other, routine tests which are not presented as decision-worthy. And as Matthias et al. (2013) point out, if a clinician who has been very directive at the outset of a consultation later tries to involve a patient in a decision, the patient may be at best confused or at worst reticent. Having pinpointed the problems with existing checklist-based approaches, Matthias et al.’s (2013) solution to this is to propose an alternative model for examining SDM – the ‘Four Habits Approach’. Like other approaches, this model focuses on eliciting the patient’s perspective and also points to the importance of demonstrating empathy. However, it also focuses on what is termed the ‘investment’ the practitioner makes in the beginning and end of the encounter, with the idea being that a consistent internal context must be provided to support participation. While not yet widely adopted, Matthias et al.’s (2013) model makes more explicit than others the problems of measurement that underlie SDM, particularly in relation to judging patient participation or engagement. Proxies such as how much a patient talks, or how many questions they ask, do not correlate in any simple or straightforward way with the extent to which a decision is shared. Large numbers of questions, for example, could illustrate a scenario in which a consensus cannot be reached; equally a patient who says little can do so because they have a good understanding of the issue under discussion and can quickly agree to participate in a future course of action. To be fair, Charles et al. (1997) do note this difficulty in their original explication of the concept, and are at pains

What Is Patient Centred Care?

19

to point out that, in their view, participation should not be considered a discrete step on a checklist, but an ongoing process throughout the consultation. Charles’ subsequent work with colleagues (Montori, Gafni, & Charles, 2006) does also incorporate some dimensions of context, but in Matthias et al.’s (2013) terms this is external rather than internal context. Montori et al. (2006) acknowledge the differences between the types of largely one-off treatment decisions that underpin the original Charles et al. (1997) model of SDM, and the type that patients with chronic conditions will need to make and revisit on an ongoing basis. They argue that chronic care requires a more active patient role in carrying out decisions, but it also provides a longer timescale over which to make them, or the possibility of trialling different options. By contrast, decisions in acute care are often urgent, may be irreversible, and may necessarily offer minimal opportunities for patient participation. Nevertheless, they suggest that the Charles et al. (1997) model can still be applied in the chronic care setting, but that there is a need to ‘emphasise the patient/clinician relationship as one of partners in making difficult treatment choices’ (Montori et al., 2006, p. 25). This positions SDM on a spectrum between informed decision-making and paternalistic decision-making. For Montori et al. (2006), the former carries the expectation that the patient will be the sole decision maker, embedding an assumption that they can fully understand the necessary technical medical information about available choices, and consider these in context. The latter operates through the doctor using professional knowledge and judgement to decide what is best for the patient. This continual refinement of SDM models in an attempt to better capture both the variety in and the nuance of practitioner/patient communication shows a similar trajectory to the way in which definitions of PCC have been developed and expanded. However, the same problem then applies to both concepts. In both cases, application of models or checklists repeatedly find practitioner attempts to deliver them as inadequate. For example, Stevenson et al. (2000) analysed 62 English GP consultations for the presence of the first two characteristics of SDM in Charles et al.’s (1997) model: (1) both the doctor and patient are involved and (2) both parties share information. They found little evidence that doctors and patients participate in the consultation in this way, and in the absence of these two components of the model there is no basis to implement the remaining two, which pertain to (3) building consensus about preferred treatment and (4) reaching agreement on the treatment to implement. It then becomes impossible to disentangle to what extent this is a problem of the practitioner(s) or a problem of the checklist. Nevertheless, it seems inevitable that at least part of the reason practitioners’ attempts are repeatedly found wanting is because of the difficulties of quantitatively measuring participation, preference elicitation or engagement in consultations in any meaningful way, as Landmark and colleagues’ (2017) comparison of scoring the same consultations for preference elicitation using different scales, and ending up with different results, illustrates. They call for a closer examination of language in its context to explain this variation, but this solution addresses the symptoms of the problem rather than the problem itself. Despite the failure to consistently find or embed SDM in healthcare encounters, it has been

20

Reconsidering Patient Centred Care

transformed, as Rapley (2008) notes, from its initial status as an analytic model of practice to as a normative model for practice. Just as in the case of PCC, this is in the face of a lack of evidence as to whether and how SDM actually works. Shay and Lafata (2015)’s systematic review of studies which measured SDM and evaluated its relationship with patient outcomes found that, when patients perceived it to be occurring, it tended to result in what they categorised as improved affective-cognitive outcomes. However, evidence was lacking for any impact on patient behavioural or health outcomes. Nevertheless, and just as in the case of PCC, healthcare practice is judged against this normative model and found wanting. One way to sidestep this problem is, as Kaldjian (2017) argues, to conceptualise SDM itself as a goal rather than an outcome. The process of arriving at a shared decision may be easy when there is no tension between a patient’s preference and what a doctor views as in the patient’s best interests. However, there is an asymmetry to the relationship in terms of access to medical information, and even the most fervent critics of the Parsonian model acknowledge that a patient is to some degree dependent on the support of healthcare professionals when ill. As Clarke (2017, p. 274) puts it, ‘Patients seek medical care precisely because of the knowledge, skill and experience of the practitioner that they lack: we cannot regard this as A Bad Thing’. Similarly, Kaldjian (2017) suggests that within SDM there is a place for ‘leaning in’ to a patient’s deliberations, or engaging in what he characterises as (respectful) persuasion. However, this begs the question of how long and hard a practitioner should engage in this kind of persuasion, and what happens if it is ultimately unfruitful. SDM therefore exposes a key tension between respecting patient autonomy and preserving practitioner integrity, which means that just like PCC, its presence or absence may be grounded in more complex issues than a checklist-based assessment of the quantity or quality of communication between the two parties.

The Rise of Autonomy As Buetow (2016) highlights, since the mid-1960s medical practice has been reshaped by the application of principle-based moral theories, and particularly a focus on the principle of patient autonomy and attendant discussions of patient rights. Armstrong (2014) notes that the term ‘patient autonomy’ was almost unheard of before the 1960s, and although it was an ethical principle, it served to create ‘a space in which agency could crystallise’ (2013, p. 170). The enactment of the principle in practice is more complicated than a simple assertion of rights, and is often taken to mean privileging patient autonomy regarding treatment and care. This focus on patient autonomy and agency in treatment and care from the 1960s onwards transforms the Parsonian conception of a patient from someone who is ‘looked after’ and compliant to someone who is expected to play an active part in their patienthood. However, the problem with principle-based theories is twofold. The first is that these are abstract criteria, so they cannot be applied as a straightforward checklist. To be of any use in practice, they must be set within the

What Is Patient Centred Care?

21

social context where the decision is to be made, and of the interests this involves. The second is that principles are not necessarily independent, but can conflict: as Kaldjian (2017) argues, beneficence-the principle of acting for patient benefit-is traditionally viewed as the main purpose of medicine and healthcare, which leads to a problematic mismatch. This mismatch, which pits autonomy in the form of patient preference against beneficence in the form of a clinician’s assessment of best interest, is at the heart of the difficulties encountered in implementing PCC. From a more sociological perspective, the acknowledgement of the conflict of these two principles is that it potentially leads to another framing of medicine as a struggle for control, where each and every encounter is a potential battleground (see Pilnick 1998; Sharrock 1979). This conflict framing feeds through into expositions of PCC that conceptualise it as a means by which patients can regain control where it has been improperly impeded (US Institute of Medicine 2001; Health Foundation 2013), rather than seeing healthcare, and the achievement of improved health outcomes, as part of a co-operative enterprise.

Autonomy and Individual Choice From a sociological perspective, bioethics can be considered as much a social movement as an application of philosophy (Bosk, 2000). As Dingwall (2002) has observed, a bioethical approach reflects the liberal individualism embedded in US (and increasingly UK) political culture, and as such contains unacknowledged values and interests in the same way as any other set of social claims. Honneth (2014, p. 16) unpicks these claims of liberal individualism in his wider critique of neoliberalism, observing how ‘the idea that the value of human subjects lies in their capacity for self-determination [is] an idea which has only gradually attained such a dominant profile’, but that as a result concepts of justice and freedom have become inextricably intertwined, with the result that individual autonomy is placed at the centre of all other ethical relativities. The consequence of this, Honneth (2014) argues, is that any demand for justice in a modern society depends for its legitimacy on making some kind of reference to the autonomy of the individual. Approaches to healthcare that foreground the principle of patient autonomy or control generally conceptualise this as operationalised through the enactment of individual choice. Mol (2008) notes that it is ‘the pivotal liberal principle that people are allowed to make their own choices so long as they do not harm others’ (2013, p. x). However, in her exposition of the ‘logic of choice’, which she contrasts with the ‘logic of care’, Mol also identifies the limitations of the former logic. In neoliberal terms, citizens are defined by the ability to control their bodies. However, since bodies with disease can be impossible to control, the implication is that citizenship is only possible for the healthy. As a result, she argues that the logic of choice is concerned with individuals who wish to be free, but an alternative logic of care is concerned with individuals who would die if they were left alone. From this perspective, centring free choice in healthcare is deeply problematic.

22

Reconsidering Patient Centred Care

In addition to the context-specific limitations Mol (2008) identifies, there are also other practical problems that result from a focus on medicine purely as a moral enterprise. Parsons’ (1951) conception was of a doctor who would always act dispassionately and in the patient’s best interest in their actions. The idea of patient autonomy as a necessary defence against potential coercion by a clinician was therefore outside of the scope of his model. He also did not pay much attention to individual patient choice, beyond laying out the general obligations the sick role placed on patients to comply with medical advice in order to get well and assuming that the possibility of social sanctions would be sufficient for people to comply with this obligation. But as Kaldjian (2017: 48) notes, a significant problem with framing medicine as a moral enterprise is that PCC ‘gives patients moral licence to make whatever choices they legally want…even when [they] lack moral integrity’. These choices may have implications for an individual patient that clinicians find problematic, because of a clinician’s view that it compromises their own welfare. However, as in the case of the treatment of communicable or notifiable diseases, patient choices may also have implications for others, yet the idea of ‘relational autonomy’ (Keller, 1997) and its attempts to move beyond the isolated individualism that a focus on choice embeds, is rarely considered in the literature on patient centred care. Buetow (2016) further contributes to this argument by pointing out that if we conceptualise practitioners as moral agents, then in these terms care becomes the most important component of the service they deliver. As a result, health outcomes become secondary. This observation is helpful in explaining the mixed results of evaluations of PCC interventions that have been described above. A focus on care may well improve patient satisfaction, but it will not necessarily have a positive impact on health outcomes. This is partly because care is a process without clear boundaries: as Mol (2008) argues, the distinction that is often made between care (activities that are done to make daily life more bearable) and cure (intervention in the course of a disease) can be simplistic and unhelpful, as they inevitably overlap. And, as has often been noted by sociologists of health and illness in the twenty-first century, as the prevalence of chronic disease increases, cure becomes in some contexts an unattainable goal. In the case of type 1 diabetes, which is the case study Mol (2008) uses, treatment of and life with diabetes are inseparable, because without treatment a type 1 diabetic will die. This inseparability also shows the problem of considering ‘medicalisation’ only from a critical perspective, or as something to be overcome.

Autonomy and Professionalism The rise of patient autonomy also brings with it an additional conflict. As Armstrong (2014) has observed, there is a fundamental paradox in that the autonomy that sociologists have identified as a key characteristic of professional status (Freidson, 1970) becomes, viewed through the lens of critical analysts, something that is to be transferred to the patient. It is certainly true that since the 1970s there has been what is sometimes referred to as ‘role convergence’

What Is Patient Centred Care?

23

(Prior, 2003) as patients have become more educated and informed about healthcare matters. It is also undeniable that the widespread and ready availability of medical information via the internet has had a significant impact here (see Stevenson et al., 2019, 2021). However, operationalising care with a focus on autonomy also embeds a further assumption: that autonomy is both desirable and achievable for all patients across a wide range of healthcare settings. Yet as the US sociologist Charles Bosk has argued (1992), the dark side of patient autonomy is potential patient abandonment, where patients can be left to make decisions that they feel at best uncomfortable with and at worst unqualified for. Though these arguments may seem abstract, they also feed through into empirical findings. Studies which have aimed to assess patient preferences for patient centred practices find that these preferences are not consistent (e.g. Little et al., 2001). Their study conducted in a UK primary care setting concluded that while there was a general preference for a PCC approach, this varied according to how ill patients were and according to their desire for a prescription. And studies which have found the assumed link between patient centredness and improved patient outcomes to be elusive (e.g. Lee & Lin, 2010) have prompted further consideration of the primacy afforded to autonomy as a principle. This has led to further debates over what ‘counts’ as patient centred care, with Lee and Lin (2010, p. 817) concluding that PCC ‘must focus less intensely on …respecting autonomy and pay greater attention to a broader set of ethical considerations’. Through the remaining chapters of this book, I will illustrate some of the ways in which an enactment of PCC which is concerned to centre patient autonomy and choice can prove problematic, including how it can lead to the abandonment feared by Bosk (1992). However, while Lee and Lin suggest the solution to this is to be found in a broader ethical grounding, I will contend that those concerned to reform healthcare practice need instead to pay greater attention to a broader set of interactional considerations.

Patient Centred Care Versus Person Centred Care: A Note on Terminology Even the casual reader making a brief foray into the literature on PCC will quickly identify that terminology is not consistent. The Health Foundation (2014) notes that the terms person centred, patient centred, individualised, personalised and many others have been used to denote apparently similar things. However, as I have noted in the Introduction, the term patient centred medicine (PCM) is generally used to identify a general approach, while the term patient centred care (PCC) is sometimes used to denote the individual application of PCM and sometimes to recognise that the delivery of healthcare increasingly extends beyond the strictly medical and foreground the importance of broader notions of care. In more recent years, the term ‘person centred care’ has become more frequently used. For example, the term is now used by the World Health Organisation (WHO, 2007) in order to set the individual receiving healthcare

24

Reconsidering Patient Centred Care

(the patient) in the context of the life they live as a whole person as they manage their condition in the context of their life, family and community. As I have already observed, in some contexts it appears that the terms ‘patient centred’ and ‘person centred’ are used interchangeably; in others the term person centred care appears to denote something quite distinct. One obvious problem here, already discussed in this chapter in relation to SDM, is that variable definitions of patient centred care make it incredibly difficult to disentangle from other concepts with which features are shared. As Buetow (2016, p. 8) has argued, if person centred care is so conceptualised simply as a means of emphasising the personhood of the patient relative to the term patient centred, then there is not much difference between the two, or as he puts it, ‘the new emperor has no clothes that clearly differentiate it’. Historically, however, it is important to recognise that the term ‘person centred’ derives from the writings of the psychotherapist Carl Rogers, and was initially used to describe a psychotherapeutic approach. ‘Person-centred therapy’ was so called because its focus was on the client’s subjective view of the world. As such it represented a move away from the traditional ‘therapist as expert’ model and instead towards a non-directive approach which aimed to empower the client (Rogers, 1951, 1961). Rogers rooted his approach in the idea that every person has the capacity and desire for personal growth and change. However, as Rose (1999, p. 218) has noted, ‘the vocabularies of the therapeutic are increasingly deployed in every practice addressed to human problems’, and the straightforward application of principles founded in what he calls the psy disciplines to areas beyond these disciplines are not without problem. He makes a wider critique of the way in which the mainstreaming of psychological language has allowed government to act upon personal choices of individuals in order to align them with prevailing political values such as consumerism. Indeed the foregrounding of the wishes and choices of the individual, so that consumer choice is equated with individual choice, is as I have already noted, a hallmark of neoliberal thought (Harvey, 2005; Hall, 2011; Honneth, 2014). Additionally, in terms of the specific application to medicine which we are concerned with here, whilst there is a widespread acknowledgement that an individual has privileged access to their own inner thoughts and feeling states, otherwise known as epistemic primacy, it is also generally accepted that there is an asymmetry of healthcare knowledge between patient and healthcare professional – what Heritage (2012) describes as an epistemic gradient. The transfer of the client/patient’s expert status from the field of psychotherapy to the field of medicine therefore cannot be straightforwardly assumed: an issue to which I will return in some depth in Chapter 4. An additional problem with the transfer of concepts directly from psychotherapy to medicine is that in psychotherapy a person is not necessarily ‘ill’ in the medical sense of the word, and so the description of a patient does not apply here. In this sense there is no ideological choice to be made in this setting through the use of the label ‘person’. However, when a choice is made between using the two in a clinical setting, as Barnett (2018) notes, the term person centred is now sometimes used as a shorthand for a descriptor of power balance in an encounter. Barnett proposes that choosing to use person centred denotes participants as

What Is Patient Centred Care?

25

equals, whereas patient centred denotes an expert/non expert relationship. But, this categorisation also relies on a simplistic problematisation of expertise: Barnett goes on to argue that the use of the term ‘patient’ encourages a more paternalistic attitude to the person who is being treated. While Barnett (2018) argues that there is a clear distinction between the two terms, Buetow (2016) notes that sometimes patient centred care is defined as a subtype of person centred care, as in the WHO’s somewhat vague assertion that patient centred care is ‘an important part’ of person centred care. Despite the fact that the WHO (2007) describe the primacy of person centred care as a ‘major shift in thinking’, any explanation of precisely why this is such a shift is not forthcoming, beyond the emphasis on wider context already described above. However, there are also proponents of patient centred care who would argue that this is already encompassed in their models or definitions: for example, Levenstein et al.’s (1986) focus on ‘understanding the patient’ highlights the importance of the wider context of their life. From a different perspective, H˚akansson Eklund et al. (2019) propose that the main goal of patient centredness is to achieve a functional life for the patient, whereas the main goal of person centredness is to achieve a meaningful life for them. Of course, no one would deny the importance of a meaningful life, but this conceptualisation sets up an unhelpful dichotomy: this is not necessarily an either/ or choice, because for many people a functional life will be a pre-requisite of it being meaningful. It is therefore unsurprising that H˚akansson Eklund et al. (2019) also highlight the way in which some components of both approaches are likely to overlap (for example, the importance of communication and respect); however, these principles are so generalised that they are likely to be components of the delivery of any kind of adequate healthcare. One possible means of distinguishing meaningfully between the two concepts can be found in the work of Kitwood (1997), who argues that person centred care is more of a culture than a process. His influential work in the context of dementia care describes a focus on personhood as the recognition that the person has absolute value and should never be treated as a means to an end. Philosophically, this position is rooted in a Kantian formulation of humanity (Kant, 1785/1996). However, Kitwood also goes on to argue that modern western culture has over-emphasised the significance of both autonomy and rational capability, and that personhood should be linked more strongly to feelings, emotions, and the ability to live in relationships. This means that Kitwood’s conception of person centred care has a very different starting point from more consumerist conceptions of patient centred care that focus on individual choice. Kitwood (1997) also argues that true person centred care should encompass the personhood of all people involved, and not just the patient, which raises interesting issues in terms of care delivery when perspectives between patients and carers or practitioners may be misaligned or in conflict. Extending this line of thought, Buetow’s (2016) critique of patient centred care is centrally concerned with scrutinising the principle of primacy of patient welfare, which he suggests is generally regarded as uncontestable since patients by definition have greater immediate health needs than clinicians. His central argument is ultimately that PCC fails to sufficiently

26

Reconsidering Patient Centred Care

acknowledge that the clinician is a person too. For Buetow (2016) then, patient centred approaches assume the primacy of patient welfare, and person centred approaches assume the moral equality of both persons in the relationship. Another way of conceptualising this, he argues, is that patient centred care’s focus on the roles people occupy means that participants are reduced to an embedded social function as patient or clinician. Buetow’s conception of patient centred care, then, suggests healthcare practice has not moved on significantly from Parsons’ (1951) role driven conception. We will return to this potential role misalignment between healthcare professionals and their patients or clients in Chapter 2, and to the extent to which the Parsonian model still has relevance for modern healthcare in Chapters 4 and 5. Ultimately, perhaps the most helpful way to think about the two terms in a practical sense is that, as H˚akansson Eklund et al. (2019) concede, they may be appropriate in different contexts. Personhood as a philosophical concept, and the role of the patient in healthcare encounters, are not straightforwardly interchangeable. Accepting this difference implies that the use of the terms interchangeably, or one as a subset of another is not only problematic, but also potentially sets healthcare professionals up to fail, by further obfuscating the standards by which their practice will be judged. Accordingly, for the rest of this book the term ‘patient centred’ will be used, unless quoting from or referring to sources that themselves use the term ‘person centred’.

The Pervasiveness of Patient Centred Care I have shown in this chapter that, notwithstanding the difficulties in defining or measuring PCC, and despite the lack of any consistent evidence for its impact on health outcomes, the concept has become firmly embedded in healthcare policy and practice. As Buetow (2016, p. 22) has observed, ‘the burden of proving the rightness of…patient centred healthcare itself has fallen on the commentators who favour something else’. He goes on to suggest that we must consider whether the model of PCC is important for its own sake, regardless of impact, or whether it only important as a means for improving quality of patient care. As I have illustrated, the latter justification is at best weak, based on the available evidence. The case for the continued importance of PCC, then, rests on moral grounds. The healthcare professional-patient interaction is generally acknowledged to be more than just a service encounter, and if medicine is conceptualised as a moral enterprise rather than a scientific one, then the adoption of patient centred care has appeared self-evidently morally right. However, how a moral principle can be transformed into an interactional one is opaque from the perspective of policy directives, and problematic from the perspective of studying actual interaction. It is to examining actual healthcare interaction that I will now turn.

Chapter 2

Analysing Patient Centred Care in Practice This chapter will build on the conclusions of Chapter 1, whereby in order to understand why the implementation of PCC has not provided the positive results that might be expected, we need to begin with an examination of healthcare interaction in practice. As I have already observed, whilst many healthcare policy initiatives originate in a top-down way as a result of moral or organisational imperatives, their implementation often depends on being talked into being by healthcare professionals. It follows that judging the achievability, success or otherwise of these directives ultimately depends on examining actual interaction between patients and professionals. The chapter will begin with some brief background to the ‘problem’ of communication in healthcare, before providing an introduction to the method of Conversation Analysis (CA), the analytic approach which will be used through the rest of this book to unpick the details of healthcare interaction. I will then move to consider a problem that has been evident to sociologists for decades: that there may be ‘good’ organisational reasons for what looks like ‘bad’ (in this case, not ostensibly patient-centred) healthcare practice. In particular, the notion of patient or client choice, and how this can be exercised, is inextricably bound with organisational contingencies. Using examples from my data alongside published examples from the literature I will demonstrate the difficulties of the assumption that autonomous choice and control are central component of PCC. Finally, I will turn to show how the specific purpose of healthcare interactions means that practices which function unproblematically in everyday interaction can become problematic when transposed to a healthcare context. Checklist-based approaches to assessing PCC or SDM embed the assumption we know what ‘good’ and ‘bad’ interaction looks like at the outset, and that a one-size-fits-all policy can be universally applied. However, practices which are assumed and intended to be inclusive or supportive can in fact have the opposite effect.

The ‘Problem’ of Communication in Healthcare As the previous chapter has shown, and as Barnes (2019) observes, there is a long line of both clinicians and academics who have argued that a medical consultation is both a social and relational situation. It follows from this that communication is central to the negotiation of not only treatments and outcomes but also relationships between healthcare practitioners and their clients. The importance of Reconsidering Patient Centred Care, 27–59 Copyright © 2022 Alison Pilnick Published under exclusive licence by Emerald Publishing Limited doi:10.1108/978-1-80071-743-520221003

28

Reconsidering Patient Centred Care

this communication is underscored by the fact that communication is consistently at the heart of complaints about medical services (Barnes, 2019). What Barnes demonstrates in relation to older data still holds true: for both the first and second quarter of 2020–2021, complaints to the NHS about communication made up approximately 18% of new complaints in relation to hospital and community health services, and for both quarters formed the largest proportion by all subject areas (NHS Digital, 2021). This is not a new or recent problem: Lewin, Skea, Entwistle, Zwarenstein, and Dick (2001) note in their systematic review of patient centred interventions that communication problems are common in professional/ patient interaction and patients are often dissatisfied with communication, going on to state that ‘some communication problems are attributed to the fact healthcare providers focus on diseases and their management, rather than people, their lives and their health issues’ (Lewin et al., 2001, p. 3). From this perspective, then, the solution is straightforward, and lies in a greater uptake of PCC and its exhortation to focus on an individual’s needs and wants. However, it is worth noting here that there is a gap of 20 years between Lewin et al.’s findings and the most recent NHS complaints figures, which suggest that despite the increased importance afforded to PCC in healthcare, and the efforts invested in training healthcare staff in this approach, there has not been a corresponding increase in general patient satisfaction with healthcare communication. Given the prevalence of dissatisfied patients, then, this state of affairs supports Barnes (2019) assertion that the importance of communication is often ignored in the wider evaluation of health services and particularly in the making of policy. She cites Waitzkin and Stoeckle’s (1967, p. 263) observation that when it comes to evaluating macro level change within health services, attention is rarely focused on ‘the micro-level of the doctor-patient interaction’. It is my contention here that this has led to a mismatch between the aspects of communication that patients orient to as problematic, and the policy initiatives that have been designed to fix them. The starting point for my analysis, then, is that the communication problems reported by patients are real, but that the problem lies in the way healthcare services have tried to resolve them. As described in Chapter 1, evaluations of PCC often conclude by finding fault with healthcare practitioners (HCPs), so that the problem becomes conceptualised as one of inadequate training. The result is a circularity in which more training to promote patient centred care is seen as the only solution, rather than any questioning of PCC itself. I will use my own data and the wider CA literature to engage critically with this conceptualisation, illustrating some of the inevitable reasons why HCPs may fail, or may ultimately be unable, to enact PCC as it is commonly understood.

Conversation Analysis as a Method for Studying Healthcare Interactions Conversation analysis is a method that has been widely used to study healthcare interactions since the late 1970s (see e.g. Heritage & Maynard, 2006a; Pilnick, Hindmarsh, & Gill, 2009). It is part of the wider research programme of

Analysing Patient Centred Care in Practice

29

ethnomethodology, which was developed as a theoretical and specialist approach within sociology in the 1960s by Harold Garfinkel (Garfinkel, 1967; Heritage, 1984). Ethnomethodology is simply defined as an approach which sets out to cover the methods and competencies that we, as members of social groups, employ in constructing our sense of social reality. Garfinkel developed the approach as part of a wider reaction against the then prevailing theoretical systems in social science, and particularly Parsonian structural functionalism. Key to this wider critique was the view that this approach underemphasised the role of human agency in society, by concentrating on the analysis of social structures. Garfinkel’s rejection of Parsons was not wholesale, however; as Rawls and Turowetz (2021) describe, he set out to provide an empirical foundation for aspects of Parsons’ position that had been criticised for their abstraction. In aiming to understand how people make sense together and how interactions work, and in documenting these interactional practices, he focused on the details that ‘seemed unnecessary to those conventional sociologists … who imagine that social practices can be abstractly specified in generic terms’ (Rawls & Turowetz, 2021, p. 156). In this sense, ethnomethodology can be seen as part of a wider move towards an interactionist sociology, though occupying a distinctive position within this move. Conversation analysis as a specific approach also originated in the 1960s, as the result of the work of Harvey Sacks, and his collaborators Gail Jefferson and Emmanuel Schegloff (see Sacks, 1992). CA is grounded in the action orientation of talk, and the way that participants use talk-in-interaction as a tool to get things done. The approach uses rigorous study of recordings of naturally-occurring interactions to enable the identification and examination of communication practices. The detailed level of analysis enables consideration of practices which would otherwise have remained unconscious (in the sense of inarticulable) to the participants. In contrast to many other qualitative approaches to the study of talk, CA analyses what people actually do when communicating, rather than what they think or say they do, and so can be used to identify practices and to reveal what is interactionally more or less effective in a particular setting (Sidnell & Stivers, 2013). The guiding focus for analysis is participants’ own orientations to the interaction as it unfolds, and the way these are displayed in talkin-interaction. This analysis is guided by three principles: that interaction is structurally and sequentially organised, so for example, however a question is framed, the production of an answer is treated as necessary by the participants; that contributions to interaction are contextually oriented, so a single contribution to an ongoing set of actions cannot be adequately understood outside of its context; and that no detail of an interaction can be dismissed a priori as unimportant, so features such as pauses, hesitations and overlapping talk should not be treated as ‘noise’ in a transcript but instead as an analytic resource. From a CA perspective these principles underpin the limitations of other approaches to the study of talk, notably those which extract and code isolated utterances, or rely on polished transcripts. Early CA work was largely focused on audio recordings of everyday conversations between friends and peers (e.g. Sacks, Schegloff, and Jefferson’s (1974) foundational work on turn taking in interaction), reflecting the

30

Reconsidering Patient Centred Care

availability of recording technology as well as a desire to understand how ‘ordinary’ interaction worked before applying the analysis to other settings. However, with the widespread availability of video-recording technology since the 1980s (e.g. Heath, 1981, 1986), video data has been increasingly used where this is possible and practicable. In whatever setting a CA analysis is conducted, there are a set of standard procedures. Recordings are transcribed using a standard CA transcription system (Jefferson, 2004 (see Appendix)), designed to preserve as much detail as possible about how things are said as well as what is said.1 Recordings are then subjected to repeated close examination, considering recordings and transcriptions together. Analysis is then generally conducted in three stages, as Sidnell (2013) describes: observation of the dataset; identification of the phenomenon of interest and collection of all examples from across the dataset; then use of both single encounters and comparison across multiple examples, to describe the practice. Preliminary findings from the analysis are generally tested at group data sessions attended by project team members and other researchers using this approach. At these data sessions the analysis is developed and refined, to establish a robust and shared understanding (ten Have, 2007). This approach enables empirically grounded, concrete specifications of what is done in interaction and how this is achieved (Halkowski & Gill, 2010), and these findings can also be shown to others and verified by reference to the data (Sacks, 1984). The main aim of a CA analysis is to describe the procedures, methods and expectations that people use in producing their own behaviour and dealing with and interpreting the behaviour of others. The basic resource for analysis is the understandings and interpretations that people themselves display of one another’s conduct. By tracking the ways in which they do this, CA researchers aim to explicate the ‘rules’ and norms that people rely on in interacting with one another. From this perspective, then, we can begin to understand how enacting policy based on organisational or moral norms while failing to consider how this might impact on, or be impacted by, interactional norms, is problematic – a theme to which I will keep returning throughout this book. Since the 1980s, the approach has become widely applied to healthcare settings and its findings accepted in healthcare research. As Pilnick et al. (2009) describe, CA studies have illuminated some of the fundamental organisational features and interactional processes in a broad array of medical encounters. There are a large number of published reviews of the application of CA to healthcare (Barnes, 2019; Beach, 2013; Drew, Chatwin, & Collins, 2001; Gill & Roberts, 2013; Heritage & Maynard, 2006b; Pilnick, 2022; Robinson & Heritage, 2014) and to avoid duplication, I will not cover this ground in detail here. However, it is worth noting that this research began in the late 1970s with investigations of real time interactions between physicians and patients, and with a recognition that, given 1

For all the data extracts used in this book, the transcripts are reproduced as they were originally transcribed (or where examples from published work are used, as they were originally published). The varying level of detail of CA transcription reflects the aims, resources, analytic focus and data permissions of the original projects.

Analysing Patient Centred Care in Practice

31

the extent to which medical practice consists of this kind of interaction, CA could be a uniquely useful tool for understanding how to improve that practice (Frankel, 1983; Frankel & Beckman, 1982). The findings that CA research generates can provide useful tools for reflection on existing practice in the form of a greater awareness of and sensitivity to the way particular actions might impact on patients and trajectories of interaction. This in turn facilitates the development of recommendations to improve communication. And since these recommendations are grounded in the details of actual interaction, rather than being derived from external models or principles, it is known from the outset that they are practical and workable. As West argues, ‘… it is only through systematic empirical study of the minutiae of doctor-patient interaction that we can learn what constitutes the alleged communication ‘‘gap’’ between doctors and patients, and how it might be transformed’ (West, 1983, p. 103). Whilst the significant contribution of CA to understanding medical interaction should not be understated, it is also important to recognise that this is situated in a wider sociological contribution. Through the systematic and detailed study of what Goffman (1983) termed ‘The Interaction Order’, CA research has also illuminated the study of social order. Part of Garfinkel’s impetus for the development of ethnomethodology was his critique of Parsons’ motivational approach to the problem of order, which was dependent on internalised norms. A CA approach reveals how, turn by turn in conversation, participants produce the social organisation of different types of medical encounters – with their attendant tasks and projects, asymmetries of authority and expertise, and particular interactional dilemmas (Pilnick et al., 2009). The generation of social order is illustrated through the ways participants organise their work routines and how they engage in and display sense making practices in real time. Collins, Drew, Watt, and Entwistle (2005) argue that CA is an ideal method for research in the field of healthcare, because of its rejection of a priori models. As they highlight, CA studies are not based on preconceptions of what patient participation ought to look like. As a result, rather than seeking to identify the presence or absence of existing key concepts or elements in the way that much of the literature reviewed in the previous chapter has done, CA analysts focus instead on how practices such as decision-making are enacted, and how the interactional configuration of this shapes, encourages or reduces opportunities for patient involvement (Pilnick & Zayts, 2016). CA work on medical interactions began with the study of primary care, which is the most common form of medical interaction that many of us experience. Its (usually) dyadic nature and the fact it takes place within a predictable, largely static, location also means it is readily amenable to recording and analysis. Primary care consultations, at least as far as acute visits are concerned, generally have two well defined and shared goals: to diagnose the patient’s medical problem and to agree on a course of treatment. However, this glosses over the multiple phases that doctor and patient must work through in order to achieve this. As Heritage and Maynard (2006a, pp. 14–15) have outlined, to achieve these goals, the doctor and patient must (1) come together and establish a relationship (opening), (2) the patient expresses the reason for the visit (presenting complaint),

32

Reconsidering Patient Centred Care

(3) the doctor examines the patient (examination), (4) the doctor produces an evaluation of the patient’s condition (diagnosis), (5) the doctor proposes treatment for the condition, in consultation with the patient (treatment), and (6) the doctor and patient terminate the visit (closing). But achieving something approximating this structure is not a simple matter, and issues and dilemmas emerge for both parties throughout these phases. For example, patients have to work out how to introduce and frame their concerns (Robinson & Heritage, 2005), how to direct the doctor’s attention towards some diagnostic possibilities but away from others (e.g. Gill & Maynard, 2006; Gill, Pomerantz, & Denvir, 2010), and how and when to introduce knowledge about their illness that they may have gained from external sources such as the internet (Stevenson et al., 2019, 2021). From the point of view of doctors, potential issues include eliciting and addressing patient concerns (Heritage, Robinson, Elliott, Beckett, & Wilkes, 2007; Robinson, 2001) and preparing patients for diagnoses which may be difficult either because they suggest there is no medical problem (Heritage & Stivers, 1999) or because they contain unwelcome diagnostic news (Maynard, 2003; Maynard & Frankel, 2006). It is only once these issues and dilemmas have been navigated that agreement can be reached between both parties in relation to the two shared goals of securing patient agreement regarding diagnoses and treatment recommendations (Per¨akyl¨a, 2006; Roberts, 1999; Stivers, 2006; Stivers et al., 2018). Characterisations of healthcare encounters in which patients are subservient to the power of doctors have been common both in mainstream sociology and amongst critical interactional analysts such as Waitzkin (Waitzkin, 1991; Waitzkin & Waterman, 1974). In this context, perhaps the single most important contribution of CA research into healthcare has been to reveal and unpack the fundamentally collaborative and contingent nature of medical encounters (Heritage & Maynard, 2006a; Maynard & Heritage, 2005). It is a fundamental feature of all interaction that participants are ongoingly attentive to the talk and visible conduct of their co-participants. As Pilnick et al. (2009) note, people rely on each other to make sense of emergent conduct by virtue of what has happened immediately before: in other words, in the light of the sequential context. This means that the position of an utterance can give it a particular sense (e.g. as resistance to a course of action, or as a complaint) without this having to be spelled out explicitly. Since interaction is collaborative, interactions are typically accomplished through sequences, so that even where one participant initiates a sequence (e.g. presents a proposal), it can only be completed when the recipient produces an appropriate second part (e.g. an agreement or a disagreement with the proposal). Whatever the second participant produces in response to the sequence initiation will be understood as responsive to it, unless it is marked otherwise. Importantly, as Pilnick et al. (2009) also describe, the production of action in any interaction (or what might be categorised as the ‘what is being done here’) is a contingent matter, since a recipient will display through their talk an understanding of the first speaker’s utterance, but the original speaker might or might not accept or correct this particular displayed understanding (Schegloff & Sacks, 1973).

Analysing Patient Centred Care in Practice

33

In more recent years, conversation analysts have moved beyond the ‘typical’ dyadic doctor-patient encounter to consider interactions between a wider range of healthcare professionals and their patients or clients, and also to consider the importance of interactions between healthcare professionals themselves. Studies have explored settings including but not limited to AIDS and HIV counselling (e.g. Per¨akyl¨a, 1995; Silverman, 1997), surgery and anaesthesia (e.g. Hindmarsh & Pilnick, 2002; Mondada, 2011, 2014; Sanchez Svensson, Luff, & Heath, 2009), palliative care and end of life settings (e.g. Ford, Hepburn, & Parry, 2019; Pino & Parry, 2019) psychotherapy (e.g. Per¨akyl¨a, 2019; Per¨akyl¨a, Antaki, Vehvil¨ainen, & Leudar, 2008), psychiatry (e.g. Bolden, Angell, & Hepburn, 2019; McCabe, 2021), telephone helplines (e.g. Butler, Potter, Danby, Emmison, & Hepburn, 2010; Greatbatch et al., 2005; Hepburn, Wilkinson & Butler, 2014; Leydon, Ekberg, & Drew, 2013), emergency calls (e.g. Kevoe-Feldman & Pomerantz, 2018; Whalen, Zimmerman, & Whalen, 1988) physiotherapy (e.g. Parry, 2009; Schoeb, Staffoni, Parry, & Pilnick, 2014), dementia care (e.g. Allwood et al., 2017; Lindholm, 2015; Williams, Webb, Dowling, & Gall, 2019) and reproductive medicine (e.g. Lehtinen, 2005; Nishisaka, 2007; Pilnick, 2008). As Pilnick et al. (2009) highlight, there is also a growing range of studies relevant to the sociology of healthcare that stand outside formal healthcare settings, for example in the context of Alcoholics Anonymous (Arminen, 1998) and the family (Beach, 1996; Jenkins, 2015). Across this broad range of settings, CA work has been used not only to identify features of more or less successful communication practices, but also to develop formal communication skills training, for example in acute psychiatry (McCabe et al., 2016) palliative care (Parry et al. www.realtalktraining.co.uk), primary care (Heritage et al., 2007) and dementia care in the acute hospital (O’Brien et al., 2018).

The Data for This Book My own dataset contains data collected for a variety of funded projects, in settings including primary care, acute and chronic secondary care, pharmacy, reproductive medicine and genetic counselling, learning disability services, anaesthetic rooms and operating theatres. These projects were funded by a variety of sources, including ESRC, UK Department of Health/NIHR, Foundation for the Sociology of Health and Illness, the Big Lottery Fund and the General Research Fund of Hong Kong. None of these previous projects have focused specifically on patient centred care, instead examining issues such as risk communication, advice giving practices, the presentation and delivery of test results, the impact of second language use on healthcare consultations, and the constitution of collaborative teamwork. However, all of these data have been collected in services and settings where patient centred care has been espoused as an underpinning principle. In analysing these data in the service of other projects I have often observed instances where the delivery of PCC became problematic, for a variety of reasons. I therefore returned to these data to conduct a more systematic analysis of these instances. The fact that these consultations were audio or

34

Reconsidering Patient Centred Care

(more usually) video recorded, alongside the existence of detailed written transcripts, has provided direct and repeated access to the practices the participants use to accomplish actions and activities during healthcare encounters. As I have previously noted, this access allows for the systematic study of healthcare interaction and the detailed specification of recurrent interactional processes. My analysis is also grounded in the large, and rapidly growing, wider literature where CA has been used to explore health care interaction. This has enabled me to integrate my own analytic observations with existing knowledge in the field and allowed consideration of examples from different contexts outside my own dataset where appropriate.

‘Good’ Organisational Reasons for ‘Bad’ Healthcare Practice Within the large and rapidly growing number of CA studies in a wide range of healthcare contexts, there are some which examine how ‘ideal’ versions of practice may be unattainable or indeed undesirable in local practice contexts. In broader sociological terms, documenting the ways in which organisational imperatives and contingencies may impact on healthcare practice is by no means a new endeavour, and in Garfinkel’s (1967) ground breaking book ‘Studies in ethnomethodology’, the chapter ‘“Good” organisational reasons for “bad” clinic records’ (written with Egon Bittner) focuses on the ways that professional work in this setting depends upon local practices and competencies which may be in conflict with official clinic policy, and suggests that clinic records should be treated as the record of a therapeutic contract between clinic and patient rather than as an actuarial record which presupposes a standard reading. While this study pertains specifically to the use of documents, the same principle can be used to understand healthcare interactions. For example, in some settings, speed and efficiency are the primary criteria that a professional performance will be judged on, and sometimes these organisational imperatives mean that ideals have to be compromised (Antaki & Webb, 2019; Garfinkel, 1967). There is an existing body of CA work that illustrates how interactional practices can be context sensitive; for example Antaki and colleagues’ work on support for those with intellectual disability across a range of settings shows how support workers in a group home context do not enact the same practices as those running a therapeutic gardening group, and that the practices they do use are tied to the specifics of the setting (Antaki, 2013; Antaki & Webb, 2019; Finlay, Antaki, & Walton, 2008). Examining isolated interaction in these settings through the lens of patient centred care can lead to a judgement that professionals are failing to orient to the interactional needs of clients. However, analysing interaction across settings and placing these analyses in their local contexts shows that in contrast to the gardening group, support workers may sometimes be orienting to a greater degree to the need to get a specific job done (e.g. beginning to prepare a meal, or getting someone home in time for dinner). In these circumstances, timeliness and efficiency are criteria that a professional performance will be judged on, as much as the interaction that takes place, and it may be necessary to

Analysing Patient Centred Care in Practice

35

make a judgment between short term and longer-term care objectives. In this setting, as in many other healthcare settings beyond the primary care physician/ patient dyad, there can be direct conflict between the autonomy and choice of the client, and ‘getting the job done’.

The Problem of Choice in Practice In Chapter 1, I noted Mol’s (2008, p. x) observation that it is ‘the pivotal liberal principle that people are allowed to make their own choices so long as they do not harm others’, arguing that this principle can be clearly seen in conceptions of patient centred care and shared decision-making that foreground ‘returning control’ to patients through choice and autonomy. However, as Mol (2008) also notes, deciding to do something is rarely enough to actually achieve it. In the context of primary care treatment decisions, the issue of patient choice may seem on the face of it relatively unproblematic: given the availability of more than one possible option which have different advantages and disadvantages, it seems generally reasonable to allow the patient to be the final arbiter of which will best fit with their perspective and priorities. There is a significant body of CA work on ways in which this kind of choice can best be offered to minimise directiveness on the part of the practitioner and to avoid presenting some choices as preferred (e.g. Stivers et al., 2018; Toerien, 2021; Toerien, Shaw, & Reuber, 2013). But of course, even these choices are constrained by largely invisible organisational restrictions such as what has been approved locally or nationally for the treatment of a particular disorder, what referrals are available for a primary care physician to make, etc. Outside of primary care treatment decisions, these constraints often become more readily apparent. This is likely to be context dependent: there may still be situations of equipoise, where available choices are judged to be medically equivalent (de Haes, 2006) but there may also be situations of urgency, safety or simply uncertainty. Perhaps the easiest way to begin to highlight this conflict is to look at data from settings where the competence or capacity of the patient or client is at issue. In these settings, the issue of the tension between offering choice vs ‘getting the job done’ becomes very visible, in a way that illustrates the shortcomings of making judgements about the delivery of PCC purely based on the scored or checklisted presence or absence of particular kinds of features of encounters. Accordingly, we will start by considering two examples of data from these settings, before turning to look at other healthcare contexts to illustrate how these issues pervade healthcare encounters more generally. In the context of cognitive impairment and/or disabilities, Antaki and Webb (2019) note that by receiving support from another person, you necessarily give up some agency and control, so that doing an activity, or indeed making a choice, becomes a shared task that must be interactionally navigated. So, while a person may choose to do an activity, the person supporting them may have greater knowledge and understanding about how to accomplish it. In their study of a

36

Reconsidering Patient Centred Care

range of support settings for people with intellectual disability, Antaki and Webb (2019) argue that support workers use their knowledge and their ability to decide on appropriate actions in order to prioritise a larger objective in the future over a locally delivered objective in the present.2 The extract below is an illustration of this, taken from their (2019) data. Before this extract, Bonnie has indicated she wants to cook a stew for her evening meal, using a recipe she has already decided upon. She is shopping with Padma, her personal assistant, for the ingredients. They arrive at the meat aisle of the supermarket. Antaki and Webb (2019): Extract 4 7 Bonnie:

Oh mince

8 Padma:

Err not so much mince, I think you’re better

9

off with (.) chunks of meat but you could have

10

different t- like if you prefer have like (0.3) beef

11

it just needs to be like (1.3) chunks.

12 13 Padma:

(1.5) It says por- that’s got (.) bone in so

14

you probably don’t want that one

15 Bonnie:

I don’t know what [( )]

In their analysis of this extract, Antaki and Webb (2019) note the delicacy in the way Padma disagrees with Bonnie’s proposal to select mince; rather than produce a direct contradiction, this is mitigated with the use of ‘not so much’ followed by the offering of an alternative proposal to select chunks of meat. However, and despite this delicacy, this short extract illustrates the inherent tension. Though Bonnie has (as Antaki and Webb point out), nominal control of the larger choice of what to cook and eat, Padma’s knowledge, and the authority that the expression of this knowledge gives her, allows her to guide and potentially to override Bonnie’s choices. Contextually, in this specific setting, they argue that a person with a cognitive impairment who requires personal support has low status in terms of assumed knowledge, and dubious authority in terms of autonomously deciding how to act. While these specific circumstances may not apply in other healthcare contexts, the general principle can be transferred; this interaction is an example of how in many wider healthcare contexts there is a more explicit need for a rapid professional judgment of what might be in the best interests of another person than might be found in doctor/patient interaction in primary care. Antaki and Webb’s (2019) work, then, shows some of the ways in which support workers engaged in ostensibly joint projects use their wider knowledge to over-ride, ignore or counter service users’ proposals. However, and importantly,

2

Antaki and Webb describe this process in terms of epistemic and deontic authority, and I will return to these terms in Chapter 4.

Analysing Patient Centred Care in Practice

37

Antaki and Webb are not critical of these support workers; they argue that what their data illustrate are the fact that across support settings there is a higher preference order for completing the larger objective, and that this is how these workers manage the dilemma that they face. As they state, it is ‘not to say that the supporters are being overbearing, callous or insensitive: … they have a more overarching objective in mind and may be working to an end which, although more distant, is in the client’s interests’ (Antaki & Webb, 2019, p. 2). And as in other examples in their data, the higher order preference may involve fundamental issues such as safety, or ensuring that someone is fed. What their work shows, then, is that although decisions whether or not to accede to a person’s wishes are taken on a moment-to-moment basis, these decisions have to be set in the wider context of institutionally mandated roles and objectives. The result is that staff in this setting are not wholly free to act. As Antaki and Webb (2019) highlight, this is linked to the fact that in intellectual disability settings there is a recognised dilemma of care vs control. Whilst framing a dilemma in these terms may be an overstatement for most branches of medicine (with the notable exception of mental health), the general principle of care vs constraint on what kind of care is allowable, achievable, affordable, and likely to be successful is ever present in healthcare, and permeates all kinds of healthcare encounters. And this also highlights a wider tension inherent in PCC, in relation to the prioritisation of a larger objective in the future over a locally delivered objective in the present. While the latter is likely to be what gives immediate patient/client satisfaction, the former may be more important in terms of the eventual outcome. Understanding this dilemma sheds light on why studies of PCC interventions may show that they improve patient satisfaction but do not ultimately improve health outcomes. The presumed moral infallibility of PCC also assumes a resolution of the dilemma that is satisfactory to both parties. However, assuming that in some contexts there may be no way of reconciling these components again raises fundamental questions about what we consider to be the purpose of medicine in modern society, to which we will return in Chapter 5. It is perhaps easy to dismiss the example above either as atypical for healthcare, or alternatively as a scenario in which the fact the client cannot have unfettered choice is inconsequential. However, it is also worth examining the corollary to this situation: what happens where (unrealistic or impossible) individual choice is privileged over a longer-term objective. The extract below comes from a dataset collected as part of a Big Lottery funded project to examine transitions from child to adult services for young people with intellectual disabilities (see Clegg, Murphy, Almack, & Harvey, 2008; Pilnick, Clegg, Murphy, & Almack, 2010, 2011). One component of this project was examining the formal Transition Planning Meetings that were held between young people, their families, and education, health and social care workers in order to formulate and implement a plan for the transition. Such meetings typically take place in a young person’s current educational setting, and the young person will usually attend for most or all of the meeting before returning to their timetabled activities. Meetings are intended to cover all aspects of a person’s life, including education, employment and health, and as such professionals from all these areas who are involved

38

Reconsidering Patient Centred Care

with the young person are invited to attend and speak. The meetings are typically lengthy and can vary considerably in content dependent on the support that the young person requires. However, if the young person has verbal capacity they generally begin by asking them what their aspirations are, in an explicit attempt to place them at the centre of the meeting. Alec, the young person in this extract, who has been attending a special educational setting, has stated right from the outset that he has a clear career ambition: that he would like to join the police force.3 What is immediately noticeable in the two extracts from the meeting that are reproduced below is that, although Alec has stated at the very outset that he would like to join the police, the other participants avoid directly suggesting that this may be beyond his competency. Discussion about alternative career or educational options therefore takes place in a context where the idea of joining the police has not been rejected. Present at the meeting are Alec, his mother, a teacher from the setting Alec currently attends, a transition coordinator (employed by the local authority), a specialist nurse and a careers worker. Alec’s intellectual disability is officially classified as moderate/severe (SPTeach2c 5 teacher from the special educational setting, TC 5 transition coordinator).

Meeting 2A 1219 SPTeach2c:

So do you think you’d like to try to do some

1220

wo::rk (.) making some sandwiches in a ´?5 bigger cafe

1221 1222 Alec: 1223 1224 SPTeach2c: 1225

5No (0.4) What do you think that you’re going to do when you finish school?

1226

(0.2)

1227 Alec:

Police

1228

(0.8)

1229 TC-6:

Police (0.3) well it’s very difficult isn’t

1230

it to go ↓straight into the police5

1231 Alec: 1232

5Yeah (0.2)

1233 TC-6:

you’d probably have to do something to get

1234

↑ready for that

1235

(0.5)

1236 Alec:

Yeah((unclear))I’m going to have a card here

3 This extract has previously been discussed in Pilnick et al. (2010) as part of an analysis into the relationship between professionals’ question design and young persons’ answers in this setting.

Analysing Patient Centred Care in Practice 1237

(1.0)

1238 Mother:

Identity card

1239

(0.6)

1240 Alec:

Yeah ((unclear))

1241 SPTeach2c: 1242

You’ve got an identity card at college haven’t you?

1243

(0.3)

1244 Alec

No (.) I mean a different one (.) it would

1245

be for the police

1246

(0.2)

39

1247 Mother:

°For the police°

1248

(0.4)

1249 SPTeach2c: 1250

Okay so police is coming out really strongly and I haven’t got a picture of the police

As this extract begins, the teacher uses the word ‘so’ to connect a new proposal to the prior discussion, though as Bolden (2006) has shown, ‘so’ is generally used in this way not only to connect but also to indicate that the connection is to previous unfinished business which remains a concern (Bolden, 2006). This new proposal concerns making sandwiches in a caf´e, and it is known to all the participants that Alec has some previous experience of this through a work placement in a charity-run organisation. This activity, then, is known to be within Alec’s current competencies. However, this proposal is immediately and explicitly rejected by Alec in line 1222. The teacher’s follow up question, with its emphasis on ‘do’, orients to the need for the meeting to produce a clear and purposeful plan for Alec, and one which incorporates further education, training, work or placement with a day service (Pilnick et al., 2010). However, Alec’s response is to minimally and clearly state his desire to work for the police again in line 1227. Following a short silence, it is the transition coordinator who responds to him. Her response, beginning at line 1229, refers obliquely to the fact that this is unlikely. However, her framing of this as ‘it’s very difficult’ presents the unlikelihood as a general problem, rather than a specific problem for Alec. She also declines to reject Alec’s expressed desire in any definitive way, instead suggesting it could follow from an intermediate stage where he will ‘probably’ have to do something else first to ‘get ready’ for his desired objective. Following the exchange about identity cards, the teacher concludes this section of the discussion by underlining that Alec’s desire has been heard and noted. As the meeting proceeds the teacher is producing a pictorial representation of it for Alec to follow, but is unable to record this aspect pictorially due to the absence of an appropriate picture; this is a state of affairs which reinforces the unrealistic nature of the aspiration relative to the range of possibilities which are depicted on the available cards. The indirect way this response to Alec is constructed illustrates its dispreferred nature (Pomerantz, 1984), and as Pilnick et al. (2010) have observed, it shares some characteristics with features of bad news delivery (Maynard, 2003) in that it

40

Reconsidering Patient Centred Care

is delayed by opinion markers and then hinted at rather than produced explicitly. This dispreferred, ‘bad news’ type delivery is understandable in the context where a professional is unable to support a young person’s expressed choice, and is no doubt related to the wider context of UK government policy emphasising competency and choice in this area (Department of Health, 2001, 2009). However, the interactional priority given to managing this delicacy means that in this part of the meeting, joining the police as a possible future course of action is not clearly rejected but instead presented as something which is delayed rather than immediate. There is some further discussion about specific competencies Alec is known to have (the ability to be trusted with keys) and about available college placements, and then the conversation continues as below, as the teacher moves to bring the section of the meeting with Alec in attendance to a close.

2A (Contd) 1348 SPTeach2c:

So (.) do you think that the best thing that

1349

you can do (.) is carry on going with mum

1350

(.) and and carry on ↑having a look at

1351

different places?

1352

(0.2)

1353 Alec:

I want to go in the police

1354

(0.3)

1355 Mother: 1356

Well (.) I don’t think that you can do that yet Alec so you will perhaps

1357

have to think about some different things5

1358 SPTeach2c:

5Yes you’ve got to do some different things

1359

first (0.4) then if you’re still

1360

interested in the police (.) °I’m sure that

1361

there are ways of finding that you can have

1362

some (.) contact with the police°

1363 ((teacher continues))

(0.5)

The teacher’s opening utterance in this section again uses the word ‘so’, again tying to previous unfinished business (Bolden, 2006), and the proposal which follows is that Alec should continue to look at possible colleges, sites of employment etc. with his parent, before making a final choice. However, this proposal is mitigated by the use of ‘I think’, which avoids claiming any rights over Alec’s ultimate decision. In response to this, Alec once again states –clearly and unequivocally – his desire to join the police (line 1353). On this occasion it is his mother who responds, but her utterance, like that of the teacher, is tentative, hedged with ‘I don’t think’ and the conditional ‘you will perhaps have to think’. As in the previous extract from this meeting, there is also a temporal aspect to her

Analysing Patient Centred Care in Practice

41

response to the proposal: her use of ‘yet’ has the effect of rejecting it only for the present time, rather than for ever. This temporal aspect of the rejection is subsequently echoed in the teacher’s last utterance, so although it initially provides for a more definitive rejection- ‘you’ve got to do some different things first’ (my emphasis) (1358–59), it is still not rejected out of hand. Shortly after the extract shown here, Alec returns to his classroom. However, the organisational imperative of the meeting is that a future plan must be decided for him, as he cannot continue in his current educational setting. Discussion continues as to an appropriate way forward, and in his absence it is decided that a residential college placement that he has been offered will be the best available option. This is despite the fact that Alec has not expressed a desire, or even any willingness, to consider this during the meeting. The end result then, is that while Alec’s choice has been privileged in the short term, this has occupied the time allowed for the meeting to such an extent that a decision is ultimately made in his absence and without his input or approval. While an analysis of the interaction that takes place here would likely show many of the characteristics that are commonly asserted as important in achieving PCC (such as establishing Alec’s perspective, listening, considering options), it is hard to see any way to frame this eventual outcome as client-centred. These extracts from Alec’s Transition Review Meeting are particularly interesting when placed alongside the extract from Antaki and Webb’s (2019) data above. In the example involving Padma and her personal assistant (and elsewhere in Antaki and Webb’s data), it might be argued that there is a lack of sensitivity to a client’s expressed wishes or choices, though as we have seen there are organisational imperatives which help us to understand this. In this example involving Alec, continued sensitivity to his expressed wish or choice means the opposite is true. Alec’s choice is outside of his current or likely competency, so there is no realistic prospect that a person with his level of intellectual disability can become a police officer. Despite this, his desire to join the police is treated with the utmost delicacy. Participants to the meeting avoid rejecting it overtly, with the result that, over an extended period, its rejection is so delicate that it may be difficult for Alec – or indeed for anyone – to recognise these as rejections at all. It is easy to understand why this might be done: rejecting Alec’s proposal carries with it the implication that Alec does not have the ability to recognise a realistic future for himself, and by extension cannot discern or act in his own best interests. An overt rejection of his plans will therefore also make his lack of ability explicit. In a service which operates with an ethos of client centred care, and a dominant discourse of self-determination and choice (Department of Health, 2001, 2009), this is exceptionally difficult terrain for professionals to navigate. However, the practical consequences of the interactional maintenance of Alec’s choice are that he continues to pursue the proposal, and that a considerable amount of interactional work is devoted to managing this and closing it down, until the meeting eventually runs out of time. Alec is therefore absent- and voiceless-when final decisions are made about his future. This extract is an acute example of the problems that occur when organisational imperatives and policy directives collide, so that professionals here are placed in an impossible situation in terms of the delivery of care. The service is

42

Reconsidering Patient Centred Care

designed to be client centred, and all the data collected from this setting show the efforts of staff to place clients at the centre of meetings that involve them. Client centred practitioners carry responsibility for enabling their clients to participate, and to express choices and desires. However, staff are also faced with managing a situation where some choices and desires, however clearly and consistently articulated, cannot feasibly be enacted, and where they are nevertheless ultimately required to provide some kind of feasible transition. This scenario, then, is perhaps an extreme example of the limits of individual choice when set against organisational imperatives, but it shares aspects with Antaki and Webb’s (2019) examples, and with Finlay et al.’s (2008) CA work showing how staff in residential care settings are constantly balancing being a good keyworker (for example providing choice) and a good employee (for example getting a person fed by a reasonable time). An additional aspect to this, of course, which is true of much healthcare practice beyond the primary care doctor/patient dyad, is that professionals often have an immediate audience of other professionals to their conduct. How staff manage these conflicting demands is available for others to see, meaning that it is not only client or patient competency that is at stake here, and making it even more difficult to step outside of the moral shelter that a displayed orientation to features of client centredness provides. The two examples above have been used to illustrate how the conflict between organisational or institutional goals and what is deemed to be good practice can be played out in service delivery, and why there might be good organisational reasons for what looks like a failure to enact PCC in a given interactional context. The example of Alec has also shown how an interactional focus on client centred practices in the short term could actually be argued to undermine the provision of client centred care and support in the longer term: when Alec fails to make a feasible or appropriate choice himself, the choice must ultimately be made for him. Alec may well have been satisfied with the interaction he was party to, but it is far less clear he will be satisfied with the eventual outcome of the meeting. This trade-off between short and long term PCC is, as I have already noted, one that is rarely recognised in the PCC literature and cannot be captured by the use of checklists or measurement scales. Of course, it might be argued that the kinds of issues I have discussed above, where there is a direct conflict between PCC and other institutional goals, could potentially be addressed by reviewing or reframing these local goals. However, the extent to which this is achievable across health services, broadly defined, is debatable. In all healthcare contexts there are limits to what is available to a patient or client to choose, determined ultimately by resources but also influenced by wider policy. In the examples of Bonnie and Alec above, it could be argued that they are attempting to make choices with individual consequences (though for a consideration of the impact of apparently individual choices in the context of relational autonomy, see Pilnick et al., 2011). However, sometimes these conflicts have much wider healthcare implications, for example in terms of public health. It is to considering the ways tensions between choice and public health manifest in healthcare interactions that I will now turn.

Analysing Patient Centred Care in Practice

43

Personal vs. Public Health Antimicrobial resistance is one of the major problems of the twenty-first century healthcare; indeed the World Health Organisation has declared it one of the top 10 global public health threats facing humanity (WHO, 2021) Antimicrobial resistance (who.int). It has implications not only for the direct treatment of infections, but also for the conduct of medical procedures such as surgery and chemotherapy. And in regions where access to clean water and sanitation is an issue, the spread of antibiotic resistant microbes has catastrophic potential. As a result, there has been increased scrutiny of the antibiotic prescribing decisions primary care physicians make, so that they are asked to balance prescribing appropriately with exercising proper antimicrobial stewardship. Stivers’ (2002, 2006, 2011; Stivers & Timmermans, 2021) CA work on antibiotic prescribing for paediatric patients examines how decisions are made in this context. Stivers (2006) uses CA to show how doctors routinely pursue responses to treatment recommendations when these are not forthcoming. Where patients produce only minimal receipts or silence in response to these recommendations, doctors do more work to justify their proposals, and to elicit more extended responses. This is an interactional demonstration of an understanding that, for a decision to be shared, a patient needs to play an active role. Stevanovic and Per¨akyl¨a (2012) argue that this is because treatment decisions concern a patient’s future actions, and so they are required to demonstrate that they assume some responsibility in this regard. The decision to prescribe – or not prescribe – antibiotics therefore presents a context where both active acceptance is required in relation to a treatment decision but where also, on occasion, patients (or in a paediatric setting, their parents) attempt to explicitly advocate for antibiotic prescription, or to challenge the doctor’s reluctance to supply it. The example below illustrates just such a challenge (DOC 5 doctor, MOM 5 mother of patient).

Stivers, 2002: 1114 1. DOC: . 2.

hh So: uh: m a- at this time I don’t wanta commit ‘er to: antibiotics

3. DOC:

Like two weeks, or three weeks, or whatever:?

4. DOC:

.h I thi:nk I’ll go ahead and treat her for the eye:s?,

5.

an’ I wanta give her some decongestant

6. DOC:

Uh:m, .hh- Ya know i- She doesn’t look like uh:m (.)

7.

Why don’t we go ahead and try the decongestant first.

((26 lines of discussion of treatment not shown)) 34. MOM:

,I mean I usually don’t- I- I usually wait to bring

44

Reconsidering Patient Centred Care

35.

her in at least until [()

36. DOC: 37. DOC:

[You wait unti- Yeah:, .hhh Uh: [m

38. MOM: 39. DOC: 40. 41. MOM:

[Cuz it’s such a big deal to come here [ () [Yea:h,5 h I mean: if you wa :nt ya know- I mean she looks.5 5Can I at least have thuh prescription an’ I’ll decide

42:

whether or not to fill it in a couple day:s,

43: DOC:

For the antibiotics?

44. MOM:

Ye[ah

45. DOC:

[Uh::m_ I really don’t like to do tha:t,

46.

because: I mean .hh She doesn’t look like she has

47. 48.

sinusitis:. Ya know?, Uhm, if you really wanta be su : re we can go ahead and take X rays to make

49.

su: re if it’s really opacity; , cause unnecessary

50.

treatment for sinusitis: she can get resistant to

51.

uh lot of those antibiotics? a lot of those bugs. I mean.

52. DOC:

.hh An :d it’s- it’s not really good for her:.

53.

(1.0)

54. DOC:

So : : we try to minimi:ze ya know- treatment until

55.

it’s really necessary

In this extract, the parent explicitly presses the doctor to prescribe antibiotics for her child’s illness (building a case over lines 34–35 and 38, that she is not someone who troubles the doctor unnecessarily, and that it is difficult for her to visit the doctor’s surgery if a prescribing decision is delayed, and finally making a direct request in 41–42). In this case, the physician responds by proposing further investigations to determine whether antibiotics are necessary (lines 47–49) and providing an explanation for why immediate antibiotic treatment is not appropriate. Parental choice is therefore rejected in the service of a wider public health objective. It is worth noting here that in other extracts in her data, Stivers does show an apparent effect of parents’ interventions in eliciting prescriptions despite the absence of appropriate clinical indicators though she acknowledges that this is rare. Such cases show that there may indeed be something of a ‘struggle’ evident in some doctor/patient encounters, and an attempt by patients (or their advocates in the consultation) to ‘take control’. However, they also demonstrate that this is a rather different kind of struggle to that conceptualised by Waitzkin and other interactional analysts who adopt an inherently critical approach to medical interaction, being locally oriented and generally short-lived. In terms of control, Stivers’ data also show that patients tend not to persist to the point where the

Analysing Patient Centred Care in Practice

45

doctor’s authority is explicitly contested. The parent in the extract above has been told that there is no current clinical indication for antibiotics, but her response is not to directly contradict this. Instead, she asks to preserve the possibility of antibiotics if the condition worsens or fails to resolve. The way this request is framed, whilst foregrounding her parental authority as a competent judge of her child’s illness progression, also preserves an orientation to the doctor’s medical authority. This orientation to the authority of the doctor can also be seen in the rare examples in Stivers’s data where parents apparently overcome their physician’s initial reluctance to prescribe. In these cases, the outcome is still framed as the doctor’s decision (for example by revising the diagnosis to one where antibiotics would be an appropriate treatment). More recently, Stivers and Timmermans (2021) have turned their attention specifically to scenarios where patients do produce resistance to treatment recommendations, where antibiotic treatment is not proposed. Whilst most patients in their dataset accept physician recommendations, they highlight the importance of the practices of expectation management and persuasion in interactions with those patients who do not. The fact that these practices are present at all is a further demonstration that the operationalisation of patient choice in healthcare is not straightforward. Seeing control as located within one or other party ignores the fact that antibiotic prescribing, like other treatment decisions, is negotiated between doctor and patient or patient’s representative. Indeed, as Stivers and Timmermans (2021, p. 114007) conclude, ‘we might wish that prescriptions were not the outcomes of negotiations between patients and physicians, but failure to recognise this reality has cost us valuable time in preserving antibiotics as an effective treatment for infection’. This conclusion further underlines the importance of an understanding of healthcare interaction for the formation of healthcare policy. Reviewing these findings and setting them in the context of other research on disagreement over treatment preferences, Stivers and McCabe (2021) note that PCC advocates envisioned an active patient who would voice preferences, but who would remain rational, and thus make choices that remained in line with standard of care practices. However, examination of actual consultations shows that patients do sometimes advocate for treatments that are not appropriate or are outwith the current standard of care. Though they do not frame it in this way, the research they draw on highlights the way the dilemma of care vs control identified earlier in this chapter can in fact manifest across a wide range of healthcare settings, from the everyday primary care clinic to the more specialist setting. Antibiotic prescribing is of course not the only setting where there can be conflict between one person’s choice and public health responsibilities. As de Kok et al. (2018) note, these debates are very much alive in HIV treatment settings, although the issue here is less one of patient advocacy for an inappropriate treatment and more one of reluctance to embark on a long term medication regime which may be difficult to incorporate into daily life and produce unwanted side effects. However, the adoption of a patient centred approach which centres

46

Reconsidering Patient Centred Care

the patient’s feelings and preferences in relation to treatment decisions, and which may ultimately lead to treatment rejection, also has significant wider implications for public health. Mol (2008) has argued that the ‘logic of choice’ underpinning modern healthcare delivery is grounded in a western-centric neoliberal ideology in which autonomy is celebrated. As a result, patient centredness and shared decision-making are often assumed to be desirable precisely because they return control and hence result in empowerment. However, as de Kok et al.’s (2018) review shows, sociological and anthropological studies raise questions about the empowering effects of shared decision-making, by illustrating how new forms of treatment like antiretroviral therapy can change identities, social relationships, and perceived rights and responsibilities. New roles, rights and responsibilities are produced, both for clients and providers. Nguyen’s (2005) concept of therapeutic citizenship is helpful here in illuminating how a shared illness identity can be created associated with patient rights (e.g. treatment access) and responsibilities (e.g. adherence) that correlates directly to the Parsons (1951) model of the sick role. De Kok and colleagues argue that by fulfilling a ‘duty’ to adhere, people living with HIV can maintain a moral identity (e.g. being a good and responsible client, relative, or partner); such a duty sits uncomfortably with any notion of individualised choice. Indeed, Mol writes (in a pre-COVID-19 context, though the relevance to pandemic management is clear), that ‘microbes and liberalism do not go well together’ (2008, p. 79); this is of course the premise on which nineteenth-century public health was based, though this premise also incorporated a recognition that good care aimed at collectives must also address the conditions in which the collectives live, including water, sanitation, etc. This in turn speaks to Nguyen’s (2005) argument that specific forms of therapeutic citizenship will emerge from, and need to be examined in, specific historical, economic, socio-cultural and political contexts. Of course, the basic principle of public health is that, generally speaking, it improves by keeping the healthy, rather than by caring for those with a disease. The ‘logic of choice’ therefore creates a tension in relation to organisational imperatives: as long as individual health is at stake then people should make their own decisions, but in public health terms they may need to learn how to make ‘better’ ones (Mol, 2008). Mol’s framing of this tension assumes a clear cut distinction between the personal and the public which may not always be present in healthcare, but in so doing it also highlights once again the context-dependent complexity that must be managed in the interaction between practitioner and patient, which will not be resolved simply by a focus on the patient’s perspective or their choice.

Ordinary Talk vs. Healthcare Talk: The Problem of One-SizeFits-All Assumptions About Context Thus far, I have considered conflicts between wider organisational imperatives and individual choice, illustrating firstly that individual choice is often an illusory ideal, and secondly showing how and why this foregrounding of choice can be

Analysing Patient Centred Care in Practice

47

problematic in a healthcare context. However, even if local and organisational goal alignment can be achieved, this addresses only one category of interactional happening which works to problematise (the delivery of) PCC. Close examination of healthcare talk shows that there are more subtle ways in which institutional context can influence interaction, so that practices we assume are patient centred based on our ordinary interaction turn out not to be so. In the previous part of this chapter I have shown how the fundamental healthcare dilemma of care vs constraint can be seen clearly in wider medical settings, but can also pervade the primary care doctor/patient encounter. In the second part of this chapter I will begin to explore the problem of assuming PCC practices are universal, and can easily be identified a priori. In considering this issue, I draw first on the work of Heritage (2011), before turning to a range of examples from my own data. Based on his extensive CA work across healthcare settings, Heritage (2011) has argued that there are some common problems which arise because of the transfer of everyday interactional practices, which function perfectly well in ordinary conversation, into the healthcare setting. He calls these ‘interactional dysfunctions’ and argues that they can be addressed by training healthcare professionals through highlighting how ‘ordinary’ practices might need to be altered in healthcare contexts. Heritage’s starting point is that the presence of conversation in human society long precedes the development of institutions such as medicine. As a result, he argues that ‘the practices of human conversational interaction…are deeply embedded in our history and ontology as a species’ (2011, p. 39). However, these practices were not developed for the purpose of healthcare conversations but for fundamental issues in the management of human communication and social relations: issues such as speaking with, hearing and understanding others; managing social solidarity and affiliation; and regulating information management. He goes on to identify three examples of practices that are well established in ordinary conversation that can cause problems when translocated into healthcare. One of these is the strong expectation in ordinary conversation that we do not tell people what they already know, and he demonstrates how this can cause problems in healthcare settings where a patient is not clear what might already be known about them through their contact with other healthcare professionals or processes that have brought them to their current encounter. A second example is what Whalen et al. (1988) term ‘activity contamination’, which is predicated on the understanding in interaction that we ordinarily deal with the matter currently at hand before turning to another. Heritage draws on the work of Stivers (2002) to illustrate how, in consultations with the parents/carers of paediatric patients for upper respiratory tract infections, physicians who use examination or test results to pre-emptively explain why antibiotics are unsuitable in this particular case can instigate an adversarial relationship, because this casts parents as having wanted antibiotics all along and then rejects their hypothesised desire as inappropriate. In these instances, the activity of making a treatment recommendation is ‘contaminated’ with public health education. His third example relates to question design. Polar questions (those requiring yes/no answers) are a common component of everyday interaction, but it is very difficult to frame such questions in ways which

48

Reconsidering Patient Centred Care

do not prefer a particular response (Heritage, 2011, 2010; Sacks, 1987). As Heritage (2011) notes, by formulating a polar question we unavoidably communicate expectations towards a likely response. For example, as part of a patient history taking, a question “Are your bowels normal?” is grammatically designed to invite a ‘yes’ response, whereas ‘any ulcers?’ is designed to invite a no. Specifically, as Heritage (2011) describes, the word ‘any’ has a negative polarity. This is illustrated by the fact that ‘I don’t have any pain’ is a common utterance which would be received by a recipient as concordant, whereas ‘I have any pain’ would not be. The relevance of this issue for medical practice that Heritage describes is that it is a commonly acknowledged issue in primary care that patients may attend with more than one problem, and that some patients report leaving consultations with unaddressed concerns (Heritage et al., 2007). The acknowledged significance of the ‘doorknob phenomenon’ and the ‘While I’m here doctor…’ type utterance (Faden & Gorton, 2018) illustrate how patients may find it hard to raise a second or subsequent concern within a consultation. One possible solution, which might on the face of it be considered patient centred, would be to ask all patients ‘What other concerns do you have?’. However, as Heritage notes, patients without additional concerns may then struggle to understand what he calls ‘the aggressive pursuit’ of additional grounds for a visit. He argues that this could then be delegitimising, leading patients to assume that their original presenting concern was not sufficiently worthy of medical attention to be the sole reason for the visit, and demonstrating the same difficulty in balancing short term/long term patient benefit that we have seen in other examples above. An alternative approach which is commonly employed by physicians is to ask ‘Do you have any more questions/ issues?’ at the end of a consultation, since this does not presuppose the presence of any issues or indeed set up an expectation that there will be. However, since the word ‘any’ has negative polarity in linguistic terms, as we have seen, a question framed in this way is set up to presume a negative answer. Heritage et al’s (2007) work shows how the use of the word ‘some’ in place of ‘any’ solicits more responses from patients, so that this is a more successful practice in terms of soliciting patient concerns. He uses this finding to argue that communication training for healthcare professionals needs to be adjusted to highlight to practitioners how ‘ordinary’ practices might need to be altered in healthcare settings, a point to which I will return both later in this chapter and in Chapter 3. Heritage does not suggest his list of dysfunctions is exhaustive, and in my own data there are numerous examples of different kinds of dysfunctions, where interactional practices that are generally promoted as positive or which work well in ordinary interaction prove problematic in a health care context. I will examine some of these below.

The Use of Broad Questions The use of broad or open-ended questions is commonly identified as a component of good healthcare practice, allowing patients to formulate their own concerns in

Analysing Patient Centred Care in Practice

49

their own terms, and being seen as a way to increase the amount of information gained from patients (Takemura et al., 2005). Indeed Byrne and Long (1976) presented the use of broad questions as one of the original illustrations of their idea of patient centredness. However, as Heritage (2009, 2011) has shown, all question formulations are not treated as equivalent by their recipients, and small differences in linguistic formulation can have significant impact on responses. In addition, in a primary care context asking what ‘other’ concerns a person might have makes sense, because they have already presented one concern as the reason for the visit. In other settings, where care delivery is structured differently, this open-endedness does not necessarily provide the intended patient centred opportunity to contribute. Instead, it can provoke some consternation over a failure to formulate a response that is appropriate to the setting. The extract below is taken from the NIHR-funded VOICE project (Harwood et al., 2018), which examined communication between staff and people living with dementia who were admitted to generalist healthcare of the elderly wards in the acute hospital setting, ultimately developing a communication skills training programme to improve this (O’Brien et al., 2018). In order to develop the training intervention, video recordings of everyday interactions between patients and health care professionals were analysed using CA (Allwood et al., 2017; O’Brien, Beeke, Pilnick, Goldberg, & Harwood, 2020; O’Brien et al., 2018; Pilnick, O’Brien, Beeke, Goldberg, & Harwood, 2021). In the extract below, an open-ended question is asked as part of a closing sequence, and the design of the question explicitly orients to the projected endpoint of the interaction: ‘anything you want to ask me before I go?’. The patient’s response to this orients to her difficulty in producing an appropriate answer, which was typical when these kinds of questions were asked in our dataset (HCP 5 healthcare professional, PAT 5 patient).4

131_224 158

HCP:

159

all right then, (0.6) any↑thing you want to ↑ask me before I go?

160 161

PAT:

162

HCP:

°no° can you su↑ggest anything (0.4) that I’ve missed ou::t? no: I ↑don’t think so, (0.4) ↑we’re (0.4) we’re quite

163

happy with how things are goi::ng, (0.4) here, (0.4) a:nd

164

(0.4) and everything seems okay with you for no::w,

4

This extract is also considered in Allwood et al. (2017) as part of a typology of different approaches to the closure of encounters in this setting.

50

Reconsidering Patient Centred Care

In this example, the healthcare professional (HCP) concludes a previous section of interaction with the phrase ‘all right then’, in what can be identified as an initial move to closure. Following a short pause, she produces the question ‘anything you want to ask me before I go?’. The patient initially appears to orient to the negative polarity and the pre-closing nature of the question with a ‘no’, but then immediately produces a question herself which makes clear that she does not feel she has the expertise to judge whether this is an appropriate end point for the encounter. As Allwood et al. (2017) highlight, the HCP’s question does not contain any topic clues or boundaries that would scaffold the patient’s response, and so the patient returns the agenda to the HCP. In response, rather than introduce any new topics, over lines 162–164 the HCP instead produces a summary assessment of the patient’s progress, grounding this assessment not only in their own judgment but also the collective judgement of the wider healthcare team. This phenomenon, where a patient pursued this kind of open question and demonstrated a lack of understanding about what might reasonably be asked here, recurred in our data. I have noted above Heritage’s work which demonstrates that the negative polarity of ‘any’ makes a negative response preferable when questions are framed in this way. Robinson, Tate, and Heritage (2016) have added to this work by showing that additional items are more likely to be forthcoming from patients if (1) they are asked about additional ‘concerns’ rather than ‘questions’ and (2) this solicitation of additional concerns is carried out earlier in the consultation, rather than at the end. However, these findings are from primary care consultations. In the context of an acute inpatient ward, which is where the extract above was recorded, interactions may commonly be at the initiation of the professional, and may not be responsive to a specific problem, making these suggestions difficult to adopt. It is also worth noting that the formulation ‘any more’ can be particularly problematic in this kind of setting, as it implies that some service or commodity has already been delivered or received; this might not be apparent from a patient perspective.5 Heritage (2010) argues that question design has a profound impact on patient response, and that this generally transcends contextual factors such as the characteristics of doctor and patient, specific medical practice, etc. However, as the example above shows, this may not always be the case. While the large body of work (not just from a CA perspective) in primary care can do much to inform healthcare practice more widely, there can still be problems with the transference of primary care doctor–patient interactional norms to wider health care practice. When a patient chooses to make an initial appointment in primary care, they do so with a prior agenda, and they know what this agenda is: a health concern that

5

As Alice objects to the Mad Hatter at the tea party in Wonderland, it is not possible to have more when you haven’t had any.

Analysing Patient Centred Care in Practice

51

they want addressed. In a context where a patient has initiated a meeting to raise a concern it might be reasonable to assume, or at least to consider whether they have another. But for routine follow up appointments, appointments that are organised at the behest of a professional, and consultations or interactions encounters elsewhere in the healthcare system, this is at best problematic, and at worst (as in some examples in the VOICE dataset) potentially distressing. This is because these open formulations have the potential to transform the non-problem of lacking any additional concerns into a problem of lacking the knowledge to produce one. The blanket recommendation of the use of broad or open-ended questions, which has pervaded conceptualisations of patient centredness since the work of Byrne and Long (1976) is therefore not quite as straightforward as is usually assumed. Other commonly checklisted or measured examples of good practice can be similarly problematic, as I will illustrate below.

Patient or Client-Led Agenda Setting In everyday interactions, if a meeting between two or more people is at one person’s behest, it is generally accepted that they will supply the reason for the meeting and direct the remit of what is to be covered in that time. As a result, most primary care interactions begin with a question from the doctor which seeks to ascertain why the patient is there. However, CA research shows that patients usually quickly pass the interactional initiative back to the doctor in order to benefit from their expertise, as we will consider in more detail in Chapter 4. In other types of healthcare setting which operate on PCC principles, it is also common to give the patient or client the interactional floor at the outset in order for them to set an agenda. However, the interactional practicalities of establishing a client-led agenda can be problematic, as in the extract below. This extract comes from data I collected for University of Nottingham/Wellcome Trust funded work examining the practice of genetic counselling, and particularly the interactional operationalisation of the principle of non-directiveness in this setting (Pilnick, 2002a, 2002b). The field of genetic medicine operates under the shadow of the spectre of eugenics (see Kevles, 1995; Shakespeare, 1998) and the foregrounding of individual choice through the practice of non-directiveness is therefore a cornerstone of practice in this area, though there has been considerable debate over whether the ideal of non-directiveness is ever achievable interactionally (see Kessler, 1992, 1997). This specific extract is taken from a consultation with a young man with a family history of Huntington’s disease, an inherited neurodegenerative disorder with a typical onset in middle age and for which there is no cure. He has made initial enquiries about being tested to establish if he has the genetic mutation characteristic of Huntington’s disease. Prior to this consultation, he has already had consultations with a psychiatrist (Dr Smith), and a genetic counsellor (Sarah), and these are referred to in the opening phase of the consultation. Those present in this consultation are the clinical geneticist (G) and the client (P).

52

Reconsidering Patient Centred Care

E31 1 97 1: G: ((following greeting sequence))Grand(.) E:::r(.)Sarah’s (.) told me 2: the gist of (.) (a) couple of things you discussed 3:

and of course Dr Smith (.) said in

4:

the letter that (0.2) e:::rm (.) y-your late father

5:

had Huntington’s disease and [that

6: P: 7:

[mmm (0.3)

8: G: you’d thought things through and I think he 9: 10:

(0.2) prompted things a little bit when he (.) he asked you some leading (hhh) questions5 5yeah (.) [and

11: P: 12: G:

[and you thought things through and

13:

you wanted just to find out a little bit more

14:

and (.) look ahead t-e:rr an’ an’ and consider

15:

the (0.2) pros and cons of sort of (.) e:r (.)

16:

what the next step might be

17: 18:

(0.3) Can (you) just ask? (.) Have you got a (.) extra

19:

questions that (.) if (.) you want to add to (.)

20:

the obvious list (.) that5

21: P:5U::m (.) not (.) not (that I think of the moment) 22:

(0.3)

In this extract, then, as is common in secondary care settings, the doctor opens the ‘business phase’ of the consultation with a summary of medical events to date, demonstrating what is known by him about the client. A proposal for a general agenda is subsequently presented; this is framed very much in terms of the client’s wishes with the repeated use of ‘you’ (e.g. ‘you wanted just to find out a little bit more’). Following a pause in line 17, where the client neither confirms nor rejects this proposed agenda, the geneticist begins in line 18 to invite the client to add to this agenda. In framing this question, he avoids the use of ‘any’ and the problems related to its polarity noted above, but nonetheless, and despite the explicit invitation at line 21, a negative response is received from the client. This is an interesting example of attempting to solicit a client-led agenda, for two interactional reasons. The first is that the geneticist frames any contributions the client might make here as additions ‘to the obvious list’. In order to add productively to this list, the client would first require some knowledge of what is already contained in this list. Whilst we know that this is not the first occasion on which the client has discussed the possibility of genetic testing for Huntington’s disease, it is the first occasion that he has discussed it with the clinical geneticist, and it will not necessarily be apparent to him whether or how issues previously discussed with other health care professionals might or do relate to the current consultation. The

Analysing Patient Centred Care in Practice

53

second is that describing a list as obvious creates a scenario where face (Goffman, 1955) and competence are potentially under threat: once something is constituted as ‘obvious’, there are considerable interactional difficulties in any admission of ignorance or request for clarification of this list. The client’s response here sidesteps this issue of what is ‘obvious’ to him by producing a response which asserts that, although he has nothing to add to the list now, he leaves open the possibility that he may do so at a later point. It could perhaps be argued that this is an atypical example because of the way this specific request is constructed and the prior knowledge that this assumes. However, attempts to solicit a client led agenda meet no more success in my dataset when different formulations are used, as the example below illustrates. This is a second example taken from the Transitions dataset, this time involving a young person called Stephen. The meeting is being chaired by a Local Authority social services team manager, with his mother, father, two support workers, a day service manager and a teacher in attendance. In the meeting from which the extract reproduced here is taken, there is an opening statement (not reproduced here), which is ostensibly addressed to Stephen but which informs all participants that Stephen will be placed at the centre of the review. General introductions then follow before the business of the review is returned to. However, as we will see in the extract below, it does not follow that this formal or explicit verbal positioning of the young person at the centre of the process leads to practical success in eliciting any kind of agenda from them (SSTM1 5 social services team manager, TC-1 5 transitions coordinator):

Meeting 1A 105

SSTM1:

Okay Stephen(0.2) um::m so this is your review

106

like I explained earlier (.) so are there any (.)

107

any ↑particular issues that you would like to

108

discuss as part of your review?

109 110

(0.2) Stephen: E::r

111

(4.0)

112 113

Can’t think of anything (0.2)

114

SSTM1:

115 116 117

No (1.0) oka::y (2.0) mum and dad (.) if are there any ↑general things that you’d like to add? (2.2)

Father:

Not really er the e::r (.) I wasn’t quite sure

118

what (.) ↓what was going to happen at this

119

particular meeting and in fact it it was as late

120 121

as yesterday that (.) that I posed the question with TC-8 was I (.) in fact I posed the question

54

Reconsidering Patient Centred Care

122

is it ↑appropriate to bring (.) to bring

123

Stephen to the meeting and I got the response yes

124

(0.4)

The chair begins this extract with a reference back to the way that he has opened this meeting, and a reiteration of Stephen’s centrality to it. Stephen is subsequently offered the chance (lines 106–108) to place items on the agenda. There is a pause, and then Stephen produces a filler (E:r) which is followed by another lengthy pause; taken together these suggest both that Stephen is having some difficulty formulating a reply and that his eventual answer is likely to be a dispreferred one (Pomerantz, 1984; Schegloff, 2007). Eventually in line 112 the opportunity is declined. The social services team manager then offers the same opportunity to Stephen’s parents (this time stressing the ‘any’). As Heritage and colleagues’ analysis of the use of the word ‘any’ would suggest (Heritage et al., 2007; Heritage & Robinson, 2011), Stephen’s father also declines this opportunity, but he produces an account for his declining which goes beyond the negative polarity of the word: it is rooted in the fact that he is not quite sure what the purpose of this meeting is. This extract, then, provides a clear example of how difficult it is for a lay person to take control of an agenda ‘in the moment’ (Pilnick et al., 2010), in a context where there is both limited time to consider the issues and limited access to what might be within the remit of this particular opportunity. Of course, an inability to contribute to an agenda at the outset does not mean that this opportunity is lost forever, and proponents of PCC would rightly argue that opportunities can be created for clients or patients to add issues to the discussion at a later point. However, this does not resolve the problem of unequal access between participants as to what will be considered appropriately within the remit of the meeting. In fact, in the continuation of this meeting (and in data not shown here) Stephen’s father does attempt to add an issue to the agenda. There have been difficulties with the transport which, as part of his package of care, is supposed to be provided for Stephen to enable him to attend day centre and healthcare services. However, when he attempts to raise this, Stephen’s father is told that it is beyond the scope of the current meeting. The analysis of this extract, therefore, points to a specific issue as well as a more general one. In specific terms, it appears that it is unlikely to be fruitful for a professional to offer the opportunity to set or contribute to an agenda (and to cede the interactional control that follows from this), in the absence of formulating a prior framework for that process. Across all my data, I do not have a single example of a patient or client being asked to set an agenda and where they are able in the moment to produce an item which is both considered relevant and adds anything new to what has already been put on the table to be discussed or considered. The fact that when asked to do so, people mitigate their answers through the use of qualifiers such as ‘not at the moment’, or else produce accounts for why they are unable to do so (such as being unsure of the purpose of the meeting) shows the potential face implications (Goffman, 1955) this inability may have. The general conclusion

Analysing Patient Centred Care in Practice

55

which can be drawn from the analysis of this practice is significant: things that in the abstract look like inclusive practices, and which may work that way in ‘ordinary’ interaction, or even in specific primary care contexts, may operate differently in other care contexts which involve professional–client interaction. Rather than patients or clients experiencing them as empowering, they may instead experience them as a face-threat. These examples, then, are an empirical demonstration of the way in which PCC, through the use of specific discursive practices, constitutes an ideal patient as a particular kind of subject with particular kinds of attributes (Armstrong, 1983; Silverman & Bloor, 1997). Where this ideal is not realised, the empowerment promised by PCC is not easily achieved. Once again, this finding underlines the importance of looking at what happens in actual healthcare interaction, rather than importing external principles or assuming the transferability of interactional practices across settings.

Tying Additional Items to an Existing Topic of Discussion I have used the examples considered so far in this chapter to show how practices that may be seen as patient centred in the abstract do not necessarily function in this way across any or all healthcare contexts. A further example of a practice that functions well in ordinary talk but can become problematic in a healthcare context is the way in which we link items for discussion. In ordinary talk, we design our contributions to be relevant to the prior talk; they are context shaped (Heritage, 1984). Therefore, if one speaker is speaking on a particular topic, a second speaker may in their next turn share their similar experiences, observations or views. However, the operation of this principle in healthcare settings is more complex. The example below is taken from a study of smoking cessation advice provided by general practitioners (see Pilnick & Coleman, 2003, 2006, 2010). When interviewed about giving cessation advice, GPs felt that they had strong reasons for preferring to discuss smoking with patients presenting with smoking-related problems (Coleman, Murphy, & Cheater, 2000). This preference was informed by the recognition that smoking was a delicate issue to raise in consultations, and that participants foregrounded the importance of good doctor/ patient relationships and described wanting to preserve these. In the example below, the patient’s presenting complaint is chest pain when carrying out normal daily activities, and he has mentioned a cough as an occasional symptom. The doctor’s subsequent questioning picks up on this. Doctor 252, Patient 10 D252: Are you coughing a lot? P10:

I can’t cough at all like, you know what I mean? It’s like someone has put a knife into it. Even when you try and blow your nose like. D252: Do you feel you needed to cough though? P10:

Yes you know what I mean?

56

Reconsidering Patient Centred Care

D252: Do you normally cough? P10:

Well yes

D252: So you’re a smoker? P10:

Yes. It’s a habit like, you know what I mean?

D252: I’ve got you recorded in 1993 as a twenty a day man. P10: Yes, well, it’s about the same thing. Yes D252: And you have a bit of a cough associated with [y P10:

[I haven’t got a cough now like, no. The point is it’s painful when I do anything like, you know what I mean?

D252: Slip your shirt off ((GP examines patient)) D252: Your lungs sound okay and there’s not many secretions, not much rattling around in there. Obviously, the fact that you are a fairly heavy smoker [y P10:

[Yes, the cough isn’t a problem. Every one gets a tickle in the back of their throat, you know what I mean and it helps you clear it. I’ve had it ((the pain)) when blowing my nose and sneezing. Yesterday I sneezed and it brought tears to my eyes.

D252: Most non-smokers you know don’t cough up phlegm. P10:

Well I don’t really like, you know. I don’t think I’m that bad, I’m only trying to get help.

The doctor’s questioning in this section of the consultation addresses the cough that the patient has reported, asking first about its frequency, and secondly trying to establish whether this is a normal or regular event for the patient. Having established that it is, the doctor explicitly links this to the patient’s smoking behaviour (‘so you’re a smoker’), which the patient confirms. However, in the next turn the doctor begins to reframe the patient’s symptoms as ‘a cough associated with smoking’, and this is strongly resisted by the patient; he interrupts, minimising the relevance of the cough for the current problem and also suggesting that the proposed link between his coughing and his smoking behaviour is not ‘the point’ of the consultation. A similar resistance is seen later on in the extract, when the doctor has finished examining the patient’s chest. As the doctor begins an utterance ‘Obviously, the fact that you are a fairly heavy smoker…’, the patient once again interrupts to dismiss the cough as the relevant problem to be focused on, and accounts for this dismissal by a claim to normalisation of the symptom: coughing is something ‘everyone’ does. In turn, the doctor resists this normalisation, with an assertion that most non-smokers do not routinely cough up phlegm. This time the patient’s response is to minimise the occurrence of this symptom for himself, before producing a statement with an explicitly moral dimension: ‘I don’t think I’m that bad, I’m only trying to get help’. The moral nature of an utterance like this is twofold. Firstly, it attends to the patient’s moral responsibility towards his own health, while minimising his role in creating any healthcare problem (by describing himself as ‘not that bad’). Secondly, it attends to the moral responsibility incumbent on the doctor to try and help the patient to

Analysing Patient Centred Care in Practice

57

re-establish a healthy state (Freidson, 1970, 1973; Gill & Maynard, 1995; Parsons, 1951). Given the fact that the attempt by the doctor to introduce a discussion of the patient’s smoking behaviour is so strongly resisted here, when it is has a clear relevant linkage with the matter under discussion, how do we explain this? Drawing on everyday talk practices would suggest that linking the patient’s presenting or existing problem to smoking behaviour is an appropriate way to topicalise it, because it then becomes both relevant and legitimate to discuss. The pervasiveness of this principle in everyday talk appears to be the basis doctors in this setting are working on when they report in interviews that they see this as a preferred approach to initiating smoking cessation discussions. However, there might be two reasons for the interactional difficulties that occur here following the doctor’s introduction of smoking: one is to do with the fact that, although there is a shared topic, there has been no establishment that smoking is viewed as a shared problem (Pilnick & Coleman, 2003). There is also a second reason which has wider implications: patients recognise the consequences of accepting the doctor’s assertions, and indeed they orient to these explicitly in utterances such as ‘I’m not that bad’. In Parsons’ (1951) conceptualisation of the sick role, patients are not held to be responsible for their illnesses. However, if action the patient has taken (or failed to take) becomes directly linked to their health concern, this lack of responsibility becomes less clear cut, and there is a potential moral implication for the patient. A second component to Parsons’ (1951) conceptualisation is that the legitimacy of being sick is conditional on a patient trying to get better, and complying with medical advice in order to do so. It follows that any attribution of illness that implicates smoking points to either the desirability of giving up, or to question marks over the legitimacy of the illness. Any notion of getting well in this context will therefore also include the desirability of giving up smoking. This analysis shows that the consequences of being labelled as a smoker in this context are far reaching, and also helps to explain why other patients in this data set respond to questions about smoking behaviour with utterances such as ‘I don’t smoke smoke’. Such utterances acknowledge the technically correct attributions made by doctors about smoking behaviour whilst at the same time resisting the label of ‘smoker’. They also resist the authority of the doctor in proposing a way forward that involves quitting smoking, by accepting the content of a prior turn but rejecting a claim about its relevance for the participant’s situation or actions (Versteeg, 2018). Conversation analytic work which examines labelling in healthcare contexts underlines that labelling is a social activity, and points to two possible scenarios. The first is that, in circumstances where the availability of additional support and resources are at stake, such as autism diagnosis, receiving an official label may be seen as a positive or desirable outcome (Hayes, McCabe, Ford, & Russell, 2020; Hollin & Pilnick, 2018). However, in other scenarios, labelling may be resisted by patients because they recognise the consequences of having a label, and the fact that the moral or social implications it carries may bring with it differential social responses and treatment (Gill & Maynard, 1995). There are no gains of the first kind to be made in these consultations but the moral implications are problematic. In the dataset collected for the smoking cessation

58

Reconsidering Patient Centred Care

project, there are examples of even more overt attempts by patients to dissociate the presenting problem from their smoking behaviour, such as a patient presenting with symptoms of a chest infection who tells the doctor who raises smoking ‘I’ll give up smoking when you get me better’. Understanding this helps to explain why, in none of the cases in this corpus does the raising of the topic in this way lead to any agreement or commitment by the patient to modify their smoking behaviour. In fact, the resistance that making the link engenders renders the topic difficult to pursue, so that it generally becomes quite quickly abandoned. The commonsense understanding of ordinary interaction that doctors bring into these consultations, then, has negative effects in this context. And from a PCC perspective, an attempt to ground discussion of the patient’s presenting problem in the wider context of their life results in defensiveness and a closing down of the topic.

The Difficulties of Distinguishing Between ‘Good’ and ‘Bad’ Healthcare Practice As the examples in this chapter have shown, there may be good organisational reasons for what may appear on the face of it to be ‘bad’ interactional practice in healthcare. These may be tied to the larger organisational objective of the institution, for example in prioritising patient or client basic needs or safety. They may also be tied to more contextually specific reasons, linked to the particular purpose of specific medical interactions, whereby a practice that functions unproblematically outside of a healthcare context or in a limited healthcare context only, becomes problematic when transposed. In this chapter I have added to the list of interactional ‘dysfunctions’ that Heritage (2011) proposes, but I certainly do not suggest that this is now an exhaustive list. What I have included are examples of additional dysfunctions that occur in the contexts that I have researched, but there will undoubtedly be others that arise in different contexts. This analysis begins to show, then, that distinguishing between good and bad practice in healthcare cannot easily be done by measures or checklists of present or absent features, because these are so context dependent. It also follows that one-size-fits-all policy recommendations for healthcare practice that are grounded in moral values are likely to be problematic because of their failure to consider the local context. The analysis I have presented in this chapter has also problematised the notion of ‘patient choice’ as either easily deliverable or unambiguously desirable. Since choice does not operate in a vacuum, the examples shown here highlight the difficulties associated with using patient or client choice as an indicator of patient centred care or as a marker that paternalism has been removed from the healthcare encounter. Mol (2008) counters the argument that choice liberates patients from the passivity that ‘old fashioned’ medical practice has forced them into, suggesting that if we think instead of care practices, patients are far from passive. To make care happen, patients must jointly act with practitioners, and they must also continue to act since care is an ongoing collaborative process.

Analysing Patient Centred Care in Practice

59

From this perspective, patients are therefore crucial members of the care team; care tasks are shared in varying ways, or as Mol (2008, p. 92) puts it, ‘the action moves around’ between patient and care team. It follows from viewing healthcare in this way that actors do not have fixed, a priori tasks, and the action is more important than the actor. In other words, it matters more that an action is taken that has a positive impact than who carries out that action. While Mol’s work does not consider the details of interaction, these words seem particularly salient in the context of the example of Alec, who appears destined for a longer-term outcome that he has no desire for because of a reluctance to step outside of the norms of client centred interaction in the short term. And this analysis also returns us to a consideration of the importance of relational autonomy (Keller, 1997; Young, 2001) rather than the absolutist version which appears bound up in conceptions of PCC which emphasise control. In Heritage’s (2011) analysis of interactional dysfunctions, he observes that the interaction order has not evolved to be designed for the practice of medicine, because the constraints and objectives associated with organised healthcare are much more recent than our need to communicate with one another. Two things follow from this. The first is that, as this chapter has further illustrated, some of our routinised everyday interactional practices become problematic in a healthcare context, or even when transferred between healthcare contexts. The second is that that we cannot hope to simply impose new collections of interactional practices – such as those proposed as part of patient centred care – on healthcare without a better understanding of the norms that underpin this interaction. This is the issue I will address in the next chapter of this book.

This page intentionally left blank

Chapter 3

On Good Interactional Reasons for ‘Bad’ Healthcare Practice The previous chapter illustrates a variety of ways which might lead to individual ‘failure’ to enact PCC, rooted in competing organisational imperatives, and ‘dysfunctions’ (Heritage, 2011) in the transfer of practices from ordinary conversation to healthcare interaction. However, in this chapter I will argue that there is a more fundamental reason why policies such as PCC are not enacted in practice in the way we might expect, and that this reason may not be as amenable to improvement through the further training for healthcare professionals that Heritage (2011) suggests as a way forward. I return here to the fact that healthcare policies often have to be talked into being at the point of care delivery, although policymakers do not generally ground their policies in an understanding of how interactions actually work. As a result, and regardless of their intentions, sometimes policies direct HCPs to go against interactional norms, or to navigate interactionally between competing moral ones. In other words, while they make organizational sense, or make moral sense from a particular perspective, they do not make good interactional sense. This argument will again be grounded in and illustrated by data collected from a wide range of health and social care settings over the last 20 years, alongside reference to the wider CA literature. In the previous chapter I drew on Garfinkel and Bittner’s (1967) foundational work to produce a wider explication of the ways in which there can be good organisational reasons for what may look on the face of it like ‘bad’ healthcare practice, as judged by examining interaction. Examples in the chapter showed how an overarching organisational objective (for example, managing antimicrobial resistance) may be privileged over immediate manifestations of an individual patient or client’s choice or control, or how a local context such as smoking cessation may disrupt ordinary assumptions about how we structure conversation. For this chapter, I have subverted the title of the famous paper, and using the idea of ‘good interactional reasons for “bad” healthcare practice’, I will show some of the ways that PCC policies may transgress ingrained interactional – and sometimes other moral – norms against which practitioners’ actions are likely to be judged. I will continue to set this analysis against a wider consideration, and a critique, of the importance attached to ‘patient choice’ and patient control as part of PCC in healthcare. I will conclude by arguing that this analysis both deepens our understanding of why policies might not have the impact we expect, and underlines the importance of understanding the interactions that these policies depend on.

Reconsidering Patient Centred Care, 61–84 Copyright © 2022 Alison Pilnick Published under exclusive licence by Emerald Publishing Limited doi:10.1108/978-1-80071-743-520221004

62

Reconsidering Patient Centred Care

The Problem of Who Knows What In order to set this analysis in context, I will first consider an issue that arises from the psychotherapeutic roots of patient centred care, as discussed in Chapter 1. It is generally accepted that participants to an interaction treat one another as having privileged access to their own personal experiences – in other words, that no one can know these better than they do – and also that they have specific rights to narrate these experiences (Pomerantz, 1980; Sacks, 1984). This is a key principle for the operationalisation of person-centred psychotherapy. What this means for healthcare professional/patient encounters more generally is that patients are treated as the experts on their symptoms. Yet, CA research shows that patients offer diagnoses to doctors rarely and under specific circumstances. As Heritage and Raymond (2005, p. 16) assert, ‘the distribution of rights and responsibilities regarding what participants can accountably know, how they know it, whether there have rights to describe it, and in what terms, is directly implicated in organised practices of speaking’. These organisational practices are reflected in healthcare professional/patient interaction, and we will return to this issue at greater length as part of a broader consideration of expertise in Chapter 4. However, it is worth noting here that Heritage and Raymond (2005, p. 36) frame this as part of the wider the sociological dilemma of involvement/detachment, which has at its heart the fact that we generally seek affiliation and alignment from others, but at the same time the other must avoid too deep an incursion ‘into territories of knowledge, feeling and relational ownership’ that recipients may defend as their own. In practical terms, then, the rights and responsibilities assigned to what healthcare professionals know and the rights they have to describe this are generally seen in a technical, medical reformulation of a patient or client’s issue, which enables the offering of a medically grounded solution or amelioration. Though patients are the experts on their condition in the context of the way it affects their everyday life, they rely on professionals to make medical meaning for them, which is (an often large) part of the reason why they seek medical help in the first place. However, as I will show below, the enactment of PCC sometimes appears to conflate an absence of paternalism with an absence of any kind of expert assessment, thereby potentially transgressing this foundational interactional norm.

The Interactional Difficulty of Non-directiveness The practice of non-directiveness, like the concept of person-centred interaction, originates in the psychotherapeutic approach of Carl Rogers (1951). For Rogers, the central notion was that the client should guide his or her own life, although it is interesting to note that he largely stopped using the term in favour of person or

On Good Interactional Reasons for ‘Bad’ Healthcare Practice

63

client centredness. Nevertheless, this concept has been transferred from the field of psychotherapy to the field of medicine and healthcare, and perhaps the place where it is are most commonly found is in genetic and reproductive medicine. The concept of non-directiveness is widely understood to mean that the role of the practitioner should be that of information provider, rather than decision-maker, and that information should be presented in a neutral, non-judgemental manner. As such, its successful enactment represents a clear example of a scenario where ‘control’ is handed to the client. This is a principle which makes sound moral sense, set against a historical context of the abuse of biological information (see Kerr, 2004; Mazumdar, 1992; Pilnick, 2002c). It also makes organisational sense, as a defence against specific accusations of institutional eugenics. What it means in practice is that in order to foreground personal choice and reproductive autonomy, and defend against any such possible accusations, practitioners generally refrain from offering advice to patients or clients about the decisions they should make based on results of screening and testing. However, the principle does not make such good interactional sense. As I have already noted, patients are generally treated as having authority in relation to their own feeling states and experiences, but in healthcare interactions they also orient to a wider distribution of rights and responsibilities in terms of who is entitled to know what, and how. It follows that simply transporting norms of interaction from a psychotherapy setting into a non-psychotherapeutic healthcare setting has wider implications for how that interaction unfolds. As a result, and as I will show, HCPs are put in a position where a policy imperative sets up an interactional conflict. In my extensive dataset on reproductive medicine, comprising both specialist genetic service provision and general reproductive screening and testing, collected in collaboration with colleagues in both the United Kingdom and Hong Kong (see e.g. Pilnick, 2002a, 2002b, 2004; Pilnick, Fraser, & James, 2004; Pilnick & Zayts, 2012, 2014, 2016, 2019; Zayts & Pilnick, 2014), there are frequent presentations of test results, where these tests are undertaken to assist patients or clients with future decision-making about reproductive matters. Sometimes these tests are conducted on a person themselves, where a reproductive decision may be at that stage hypothetical. In other scenarios they are conducted on a foetus, when a woman is already pregnant. There are also two distinct types of test: diagnostic and predictive. For the former, a health professional will (in almost all cases) be able to give a definitive answer as to the presence or absence of a particular genetic condition. For the latter, the result will come in the form of a risk factor, expressed as a probability that a particular condition will be present. Regardless of these differences, across my dataset healthcare professionals typically refrain from offering advice to patients or clients about the decisions to be made based on test results. This is what would be expected given the ethos of non-directiveness, and it certainly works to foreground personal choice and reproductive autonomy. However, it is also interactionally problematic. This is because in most everyday contexts, as well as in most medical contexts, where some kind of news or results

64

Reconsidering Patient Centred Care

are delivered, there follows an elaboration which suggests what status the results have, and in medical contexts, some kind of proposal which suggests future alternatives (Maynard & Frankel, 2006). As Pomerantz (1984) highlights and Edwards and Potter (2017) emphasise, with specific regard to descriptions and assessments in everyday talk, a description of an event is not complete with the reference to it alone. Instead, a conclusion or a point is needed which gives a summary of the sense or experience of the event. Otherwise, a listener has to work out for themself how the description of the event should be interpreted, in the absence of the additional knowledge the teller has. Research specific to healthcare (Maynard, 2006), describes how in a primary care context, doctors normally announce news and then follow this announcement with an elaboration of this news, which gives the patient some indication of how it should be interpreted. Maynard describes a news delivery sequence in which doctors begin by citing or explicating the evidence, which is frequently, as is the case in my data, a test result. This is often done as a first turn announcement and is generally followed by a proposal of what that announcement means. Meaning-making, then, is commonly seen as part and parcel of diagnostic activity, following directly from the announcement of the result. When, in the context of genetic or reproductive medicine, patients do not receive this, they orient to its absence, often doing so repeatedly in an interaction. The result is a conflict between a morally-derived policy imperative and a well-established interactional norm. The example below comes from a dataset of antenatal screening consultations collected for a series of projects conducted collaboratively with colleagues in Hong Kong. These projects have variously examined the communication of risk; the use of English as a lingua franca in antenatal screening; and the impact of second language use on decision-making (Pilnick & Zayts, 2012, 2014, 2016, 2019; Zayts & Pilnick, 2014), funded by the General Research Fund of Hong Kong. This example is taken from the Hong Kong corpus and is a consultation between a nurse specialist and a pregnant woman, conducted in English. The nurse has just been explaining the different screening and testing options that are available to the woman to establish the risk of foetal anomaly. One option is a predictive (or screening) test: nuchal translucency screening. This is non-invasive (so there is no risk to the woman or the foetus in carrying it out) but is less accurate than the second option, amniocentesis. Amniocentesis is a diagnostic test, so will give a definitive result, but it is invasive (it involves drawing amniotic fluid from the uterus) and therefore carries a risk of miscarriage. Having given this information to the woman (this part of the extract is not shown here), the nurse then begins to enquire about a decision (N 5 nurse, W 5 pregnant woman). HK: WTX 1. 2.

N: Or, can you decide today? Or you want- you prefer to talk to your husband first?

On Good Interactional Reasons for ‘Bad’ Healthcare Practice 3.

W: It’s ok, which is the best for me? [I(h)] (h)don’t k(h)now.

4.

N:

5.

65

[Naa,] see, (.) Naa, see now whether you want to have the, the first thing, is (.) you decide you want to have

6.

test, or no test, ok? And the second thing is whether you want to have

7.

direct test, or indirect test. Whether you want to have an accurate test,

8.

or whether you want to have a safe test. If you want to have a safe test,

9.

then you need to undergo the screening test. But if you think that I

10.

want to have an accurate test, then you need to undergo the, the::, the::,

11.

the other test.

12.

(10) ((W is looking at the papers in front of N))

13. W: ˚huh huh huh .hhh˚ ((chuckle))

The framing of the pregnant woman’s question in line 3 of this extract makes explicit her difficulty in making a decision. However, it also suggests that her difficulty relates to ‘not knowing’ what would be the best decision for her to make in her circumstances, as opposed to ‘not understanding’ the information she has already been given. Despite this, the nurse specialist avoids answering the question directly, and instead responds by producing a lengthy summary of the key characteristics of the tests that have already been offered to the woman. The available choices are restated (including the option not to test at all), alongside a generalised presentation of the potential advantages and disadvantages of each test. However, no direct advice is given about which to choose, and the nurse also repeatedly states that the decision lies with the woman, through the use of formulations such as ‘if you want’ and ‘if you think’. Nonetheless, this repetition of information and reiteration of agency does not lead to the woman’s decision. Instead, there is a very long pause (10 seconds in line 12 where the woman stares at the information leaflets and medical records on the desk) followed by a wry laugh that appears to indicate that she is still encountering difficulties in reaching a decision. This consultation, then, is conducted in accordance with the principles of non-directiveness: the nurse does not recommend any particular course of action, and it is clear that the woman is given autonomy to make her own decision. However, it also brings to mind Bosk’s vivid metaphor that ‘the dark side of patient autonomy [is] patient abandonment’ (1992, p. 158). Whilst it is clear that the woman here is given autonomy to make her own decision, it is far less clear that this autonomy is helpful or desirable to her, and her question about what is best makes clear that she is orienting to a norm where she can expect some kind of professional guidance. But the corollary to this is that where professionals do make recommendations in these scenarios, they are potentially at odds with the non-directive ethos under which this service operates (and researchers, including sociologists, have been very critical of such direction where they find it (see Press

66

Reconsidering Patient Centred Care

& Browner, 1997; Thomas & Rothman, 2016; Williams, Alderson, & Farsides, 2002)). Returning to Bosk’s (1992, pp. 40–41) ethnographic work on paediatric genetic counselling, he describes how ‘The commitment to a nondirective style leads counsellors to respond to the question ‘What should we do?’ with a didactic lecture about how only the couple can know what is right for them, and to respond to the follow-up question, “What do others in our situation do?” by dismissing the relevance of the question because each individual couple is in its own unique situation’. I have previously documented the occurrence of precisely these kinds of questions in UK genetics clinics (Pilnick, 2002a). Bosk (1992) goes on to describe the way in which these decisions are deflected by framing them as personal or familial, which therefore places them outside of the professional relationship. Across the data I have collected from both reproductive medicine and clinical genetics settings, there are numerous similar occurrences of this kind of deflection, where not only is the meaning-making that is generally found in medical practice absent but the decision is also placed outside the realm of medical expertise (and I will return to this issue in a wider consideration of expertise in Chapter 4). However, while Bosk’s ethnographic work richly illuminates the difficulties that arise in the setting, his analysis is based on field notes that are not intended to capture interactional detail. The examples reproduced here and in Chapter 4 from my dataset illustrate precisely how patients or clients expect and seek that meaning, and the considerable amount of interactional work that health professionals have to do in order to deflect the interactional norm of providing it. In considering this problem, it is helpful to return to the work of Mol (2008). She argues that the idea of ‘Westerners’ as autonomous members of free societies is an oversimplified clich´e, although it also one which has received much political traction. The fact that people in similar circumstances often end up choosing remarkably similar things illustrates the limitations of the idea, highlighting that choices are not made in a social vacuum. This point has been made by many others as part of wider critiques of neoliberalism (e.g. Harvey, 2005; Hall, 2011; Honneth, 2014). However, and particularly relevant to the argument about abandonment that has been made here, Mol (2008) goes on to argue that even in free societies, making choices consumes time and energy that some members of a society might not have or wish to spend. Material resources impact on (and in some cases determine) choice, but even where there is a real choice to be made it will necessitate weighing up advantages and disadvantages, probabilities and risks. As Mol notes, there is a large body of psychological research which suggests that, from a rational perspective, people find this process difficult, and are susceptible to differences in framing or presentation. This difficulty is well established specifically in relation to genetic information (see Pilnick & Zayts, 2014) but it also highlights a moral conflict. Whilst ‘the logic of choice’ assumes that choice is a normative good, because it offers individuals autonomy, the same logic then avoids making normative judgements as to which treatment or goal is best, or as Mol (2008, p. 85) puts it ‘the autonomy that (competent) individuals are entitled

On Good Interactional Reasons for ‘Bad’ Healthcare Practice

67

to within the logic of choice is precisely the autonomy of attaching their own value to just about everything (except autonomy)’. For some patients, choice may well be experienced as control, but for others it may be experienced as a burden. However, the possibility of abdicating this choice in order to avoid the kind of abandonment illustrated in the extract above is not generally to be found in PCC checklists, or in related measures of patient engagement or empowerment. One problem therefore with the adoption of non-directiveness as a guiding moral principle in the service of patient autonomy is the potential that it can be experienced interactionally by patients as abandonment. Another problem with the principle, however, regardless of any moral framing, is that it is interactionally very difficult to achieve. Landmark et al. (2017), in their examination of shared decision-making in a range of (non psychiatric) secondary care settings in Norway, use CA to examine instances where doctors present a hypothesis about patient preference in the form ‘if you think X, we/you can do Y’. They note that while this kind of formulation (which is also used in the example with the nurse and pregnant woman above, although with the singular ‘you’ rather than the collective ‘we’) may seem to embody the ideals of shared decision-making, allowing for exploration of options in a non-biased manner, it is in practice incredibly difficult to produce a truly neutral utterance. Different ways of framing different possibilities (e.g. from their data (2017, p. 2084), in relation to the timing of test results ‘“If you are very impatient” we can do X’) result in them being heard as more or less desirable, or even to delegitimise certain options completely. Quantitative scoring scales for SDM such as OPTION5 do not generally distinguish between these framings, meaning that as well as Landmark et al.’s (2017) demonstration that the same interaction can receive completely different scores by applying different scales, even when the same scale is used utterances that may be received very differently by patients or clients can be given the same score. In addition, the positioning of utterances sequentially can affect how clients hear their options, and I have already noted in the previous chapter that in examining antenatal screening data elsewhere I have shown how it is not only the offering of choice but also the positioning of that choice that is important in how it is experienced. When a choice about foetal anomaly screening is presented to women at the end of a sequence in which routine tests (e.g. for blood group and iron levels) have been presented, this positioning works to undermine women’s perceptions of choice (Pilnick et al., 2004). Once again, the impossibility of making judgements on talk without considering that talk in its sequential context is underlined. The discussion so far shows that, despite its moral grounding, the principle of non-directiveness creates interactional difficulties for two reasons. Firstly, a practitioner has to try to do something that is interactionally almost impossible, in a context where clients may either resist or may infer direction regardless. Secondly, if they do succeed in achieving this, then what they have handed over may not only be control, but also the (potentially unwanted) burden of choice. Mol (2008) suggests that this burden is added to by the presence of guilt, because having a choice necessarily implies that the person who chooses bears responsibility for what follows the choice. Such guilt is potentially magnified when, as in

68

Reconsidering Patient Centred Care

antenatal screening, the choice is being made by one party on the behalf of another. However, there are further layers of complexity to add to this already complex interactional environment.

Dealing With Uncertainty Given the prevailing norm of meaning making outlined above, it is perhaps unsurprising that, even in a context underpinned by non-directiveness, healthcare professionals do in some circumstances find themselves going some way to providing this meaning. The example below is taken from the same Hong Kong data corpus, and is extracted from a consultation between a doctor and a pregnant woman. This time the consultation is taking place when the woman has already undergone nuchal translucency screening tests. All women with a ‘high risk’ or ‘screen positive’ result (for this clinic a threshold of either 1 in 250 or 1 in 350, depending on the combination of tests that are undertaken) are called back to the clinic for a consultation to discuss possible next steps. Therefore, the woman in this consultation knows, by virtue of having been called back, that her result must be officially ‘high risk’. However, women do not find out the actual result until it is given in the consultation. In this case, the woman’s result is 1 in 200. The conversation is conducted in English (D2 5 doctor, W 5 pregnant woman). HK 2014 17. 18.

D2:

The chance that the baby have Down syndrome is (.) about one in two hundred.

19.

W:

°(exclaims in Tagalog)°, okay.

20.

D2:

Okay? Not- not a very high [risk.]

21.

W:

[Yeh,] °I know.° ((while nodding))

22.

D2:

So it’s:: sort of less than point five percent

23. 24.

W: D2:

°Yeh.° [°Yes. Sure.°] [So less than ] one percent

25.

W:

Yeh, yeah,

26.

D2:

Okay?

D2:

.h so um it’s up to you. (.) We- we- we still

27. 28.

((W nods))

29.

call it screen positive because there is a small

30. 31.

possibility that the baby might have Downs W:

Um hmm

In this extract, the doctor presents the overall result as a ‘1 in x’ figure (‘about one in two hundred’, line 18); despite the fact that the printed sheet from which the doctor reads the information gives a definitive number, it is transformed by

On Good Interactional Reasons for ‘Bad’ Healthcare Practice

69

her utterance into something less certain, through the use of the mitigator ‘about’. The woman’s response to this is an exclamation of shock made initially in Tagalog, her first language, and the video recording shows that the shock is evident in her face. This utterance is followed immediately by the doctor’s qualitative assessment that this is ‘not a very high risk’ (line 20), so that the doctor makes some meaning from the numerical result which situates this within her wider knowledge of these results. Maynard (2006) shows how in cases where patients show emotional reactivity to news, affirming positive aspects of it and disconfirming negative aspects are ubiquitous devices which respond to this reactivity. This is clearly evident in the doctor’s utterance here, where the negation contained in her assessment acts to counter any implicit assumptions that might be made about this test result. As Pilnick and Zayts (2014) have noted, contextualising results through the use of formulations such as ‘not so high’ draws on the notion of range (Sarangi, 2002), which in turn situates this specific risk in the context of the health professional’s wider knowledge of risk factors and thus reduces its impact (Michie et al., 2003). In this instance, then, the news delivery sequence is followed by some explication of what this news means. There is a further minimal response from the pregnant woman in line 21, before the doctor follows this up in lines 22–24 by producing what O’Doherty and Suthers (2007) refer to as a ‘‘loss framing’. Through this framing, the 1 in X risk factor is converted to a percentage chance of the undesired event occurring, but such framing serves to emphasize its unlikely nature. In the course of this short extract, then, there is a shift away from the official ‘problem’ that the woman’s recall to clinic indicates and towards a position where the risk is subsequently framed – in both qualitative and quantitative terms –as uncertain and small. Minimal, and in some cases non-verbal, acknowledgements of this transformation are produced by the pregnant woman over lines 23–27, before the doctor offers an explanation of the apparent contradiction at work here. This explanation is made with recourse to the official category of ‘screen positive’ (the terminology used by the clinic for results that exceed their threshold, and what the woman will have been told when she was recalled to the clinic) and is also linked to uncertainty, because although on the one hand the woman has been told not to worry, the result cannot rule out a ‘small possibility’. In the remainder of the consultation (not shown here) amniocentesis is then topicalized as a possible option to address this uncertainty. Whilst the extract above shows the way in which a healthcare professional might engage in some form of meaning making for a patient or client, it also shows another interactional norm in operation. As Maynard (2006) observes, there is a tendency for what he calls ‘auspicious interpretation’ in medical settings—downplaying the negative aspects while emphasizing the positive ones; this tendency is built on the norm in ordinary conversation to move towards ‘optimistic projections’ (Jefferson, 1988). Per¨akyl¨a (1991) gives an illustration of this in what he calls ‘hope work’ in the care of seriously ill patients. Buetow (2016) develops this argument further, to suggest that because too much information may frighten the patient and therefore reduce patient welfare in and of itself, so clinicians may communicate hopeful prognoses. In part, this tendency to offer hope may be a way of dealing with the difficulty of intersubjectivity that clinicians

70

Reconsidering Patient Centred Care

face, in that what they might perceive as ordinary or routine as a result of repeated exposure might be perceived as an emergency or crisis by a patient for whom it is a unique experience (Hughes, 1951). In relation to the specific extract above, it might be argued that this auspicious interpretation responds to the evident shock that it is displayed by the pregnant woman on receipt of the news of the test result; research in other healthcare settings shows that where patients are distressed, other activities are typically suspended until the distress can be resolved (Pilnick et al., 2021). However, Maynard’s (2006) analysis shows that this auspicious interpretation is more widespread than as an immediate distress response, and in my data it is commonly seen when dealing with uncertainty. As Pilnick and Zayts (2014) have argued, following Davis’ (1960) observation that uncertainty can be used by doctors to serve functional purposes, the analysis of consultations in which uncertainty cannot be avoided show how it can be used as an interactional resource. To put it another way, uncertainty can be talked with as well as about, and this means it can be used to facilitate positive framing. As the example here shows, since the result is uncertain, it can always and truthfully be claimed that it might be nothing to worry about. This analysis of the way (partial) meaning is made from uncertainty is not intended to be critical of practitioners; rather, I have argued that there are good interactional reasons for this kind of presentation. Some of these may be specific to the presentation of individual patients or clients (such as visible distress) but others are more pervasive, linked to what Maynard (2006) calls ‘auspicious interpretation’ and which itself is linked to the preference in ordinary interaction for moving conversation towards something positive to facilitate closure of a topic or interaction. In everyday talk, discussion of troubles is normally exited with optimistic projections (Jefferson, 1984, 1988; Stivers & Timmermans, 2017): what Maynard (2003) calls ‘good news exits’ and what Holt (1993) calls ‘looking on the bright side’. Stivers and Timmermans (2017), in their CA analysis of presenting genetic testing results, suggest that presenting challenging results with a ‘silver lining’ allows parents of children with genomic disorders the hope to continue a challenging parenting journey. In the same vein, they may function in the antenatal clinic to give hope to continue a pregnancy if that is the course the woman chooses to take. However, this does not remove or remediate the practical problem that is faced in this particular setting. As Pilnick and Zayts (2014) have noted with specific reference to the clinic from which the extract above is taken, the cut-off point for what is considered to be ‘high risk’ or ‘screen positive’ results are 1 in 250, or 1 in 350, depending on the combination of tests. These figures equate to percentages of approximately 0.4 or 0.29, and a figure of this kind of magnitude (less than 1%) may generally be viewed as small in all circumstances. The ‘essential tension’ (Mazeland & ten Have, 1996) that results is that women are placed in a high-risk category and recalled to the clinic on the basis of these figures, but the same figures are then used interactionally to emphasise the small mathematical nature of the risk. The conflict between official policy and interactional norms is therefore also overlaid by organisational issues. To adopt and persist with the clinic’s official high-risk narrative is to go against the ‘bright side’ interactional norms for both ordinary and medical interaction. However, to

On Good Interactional Reasons for ‘Bad’ Healthcare Practice

71

follow these norms is to create the ‘essential tension’ described above, where there is no consistency between organisational and interactional framings of results.

Telling People What They Already Know A further example of the way in which interactional norms or imperatives can collide with well-intentioned policy comes from a dataset collected for a UK Department of Health funded project, examining communication between pharmacists and attendees at an outpatient paediatric oncology clinic where children under the age of 18 were treated for cancer and leukaemia. The project focused on the communicative competencies required of pharmacists in this role, and on the particular challenges of communicating in the context of an often long-lasting treatment regime which is also subject to regular change (see Pilnick, 1998, 1999, 2001, 2003). In this context, although the child is the patient and some older patients came to clinic alone or were the primary participant in the consultation, many of the interactions are primarily between parents and pharmacists. Pharmacists in the UK are bound by a Code of Ethics (Pharmaceutical Society of Northern Ireland, 2016; Royal Pharmaceutical Society of Great Britain, 2021) which, against a backdrop of providing patient centred care, lays out their responsibilities, including (1) that they should ensure the safe and effective use of medicines, and (2) that they must not, either by their actions or omissions, create a risk to patient care. In practice, these professional obligations are operationalised by ensuring patients know how to take their medication, at the point at which it is handed over to the patient. In the extract below, this is what the pharmacist begins to try and do. The participants are the pharmacist (Ph) and the patient’s mother (M). The patient is present but does not speak: dc/op/be 1

Ph:

2

M:

3

(name)? (.) OK (.) [I’ll go through it all with you [Yes (0.6)

4

M:

5

Ph

It’s alright (.) I know [it anyway [Oh yeah (.) I apologise for the bottles

In this extract, the pharmacist’s statement ‘I’ll go through it all with you’ clearly orients to professional responsibility, but it also allows for little or no knowledge on the part of the mother of a long-term patient. This potentially has moral implications, which we will consider in more detail below, and is immediately countered with what can be heard as a rather defensive response from the patient’s mother. CA work in other settings clearly shows that there is an interactional norm that we do not tell people what they already know (indeed this is one of the interactional ‘dysfunctions’ Heritage (2011) identifies, though in a

72

Reconsidering Patient Centred Care

different context where it is a problem which faces patients when asked for their medical history and where they may not be clear what is already known about them by a particular practitioner). However, in this context the management of the norm becomes interactionally embedded in a professional role, where it potentially conflicts with the ethical requirement incumbent on the pharmacist to ensure that patients are familiar with the dosage instructions for their medications. This creates the obvious difficulty of either attempting to give advice to already knowledgeable parties, thereby running the risk of undermining their competence, as above, or else assuming knowledge on the part of the recipient which might not be warranted. While other approaches can be used, for example asking a patient or parent if they would like an explanation, these are also potentially fraught, since a rejection of an explanation will likely only assert, rather than demonstrate, understanding (Pilnick, 2003). It may therefore still remain unclear what understanding someone actually has. Pharmacists in other settings, for example HIV clinics, have sought to address this issue by asking patients to talk through their medication regimes (Watermeyer & Penn, 2009); however, this approach also has possible downsides in that there is the potential for loss of face (Goffman, 1955) if a patient is then unable to so correctly. What even this short extract illustrates then, is the problematic interactional contingencies that can arise when professionals use approaches to advice or information giving that are deemed inappropriate by patients or clients. There are particular sensitivities to be managed when those approaches do not allow for the degree of knowledge and/or competence that long-term patients or their representatives acquire. As I have previously noted (Pilnick, 2003), while much of healthcare practice depends on the giving of information or advice, this is always a delicate interactional practice because, by its very nature, the relationship between advice or information giver and advice or information recipient is inherently asymmetrical. The act of giving advice or information defines one party as having significantly more expertise than the other (Heritage & ¨ Lindstrom, 1998). This asymmetry of expertise is not generally treated as problematic for activities such as diagnosis, where the reason for consulting with a healthcare professional is for a patient to solve a problem they cannot solve for themselves. However, this situation becomes much more complex for patients with chronic illness or their carers, when ‘what I know’ about an illness may come to be seen as equally significant as ‘what is known’ (Macintyre & Oldman, 1977). An extra dimension is added where the patient is a child. As Strong (1979) observed, in his study of paediatric hospital clinics, the carers of paediatric patients are largely treated as well intentioned, but not necessarily competent. Whilst the pervasive accuracy of this statement may have changed since the 1970s, it is still the case that there is a moral dimension to these encounters, in the sense that the parental obligation to look after young children may be accompanied by an attendant desire to demonstrate that this obligation is being properly fulfilled (Voysey Paun, 1975/2006). Since it is difficult to do this overtly by rejecting healthcare professionals’ ‘officially produced’ versions of reality outright (Pollner, 1987), clients instead take opportunities to demonstrate their knowledge where they can.

On Good Interactional Reasons for ‘Bad’ Healthcare Practice

73

This scenario, then, is an example of how, in some healthcare settings, in order to discharge their professional obligations, practitioners need to either run the interactional risk of telling people what they already know, or have to ask them to actually demonstrate their competence. Either of these go against interactional norms. Uninvited offers of advice are commonly oriented to as problematic (Jefferson & Lee, 1992; Heritage & Sefi, 1992) even in institutional contexts where advice is expected. However, assertions of knowledge from patients may not demonstrate understanding, and asking them to provide this demonstration can be face threatening. This, then, is another illustration of the difficulties of managing the interactional tensions inherent in PCC: attempting to ensure patient action is well-informed can result in defensive or dissatisfied responses, and there is no way of addressing this that is likely to be universally successful, given that alternative ways of approaching this issue also have potential attendant disadvantages. It therefore becomes clearer why resolving communicative problems in delivering care is not simply a matter of more or better training in PCC.

Patient ‘Troubles Telling’ Another area where there may be good interactional reasons for what may appear on the face of it to be ‘bad’ healthcare practice is in relation to what happens when patients or clients tell troubles to a healthcare professional. There is a significant body of CA work on troubles telling, developed from Gail Jefferson’s (1984) work and her collaborations with John Lee (Jefferson & Lee, 1981). In terms of what might be considered a ‘trouble’ in interaction, Sacks (1995, p. 59) noted that, even where people do not know each other, there are certain items which we hear as unambiguously bad. The example he gives is of asking ‘How’s your appetite’ and hearing ‘not too good’. In this example, it is not necessary to know what this means specifically in terms of food intake, in order to recognise it as an undesirable state. Jefferson and Lee’s (1981) work on troubles telling in ordinary conversation highlights the asynchrony between different kinds of interactional roles. When a party begins to tell a trouble in conversation, Jefferson and Lee show through their data how the preferred role for the person who is listening is that of ‘troubles recipient’: in other words someone who listens to the trouble and empathises with it. This means that the production by the listener of any kind of advice sequence has the potential to disrupt this alignment. Jefferson and Lee (1981) found that, when one party attempted to give immediate advice in response to a trouble expressed by another party, dispute was common. As they explain, there is a fundamental interactional reason for this. In terms of the general conversational categories of ‘speaker’ and ‘recipient’, both ‘troubles teller’ and ‘advice giver’ occupy the same category. By giving advice, the recipient of the trouble shifts themselves into the ‘speaker’ category. It follows, Jefferson and Lee (1981) suggest, that the rejection of advice that recurs in talk about a trouble may have little to do with the content of the advice per se, and more to do with a speaker’s attempt to preserve the status of the talk as a troubles telling, and

74

Reconsidering Patient Centred Care

therefore to preserve their position as the incumbent speaker with the conversational rights this entails. This interactional misalignment of the category of ‘troubles teller’ with the category of ‘advice giver’ is particularly pertinent for healthcare practice. Patients often begin a healthcare encounter with a problem presentation (in fact this may be specifically solicited through initial questions like ‘What seems to be the problem’?). However, for patients, healthcare problems are unlikely to exist in isolation, and are instead often intertwined with problems which have personal, social or economic origins which may fall outside of a professional’s domains of expertise and ability to offer a solution. Indeed, this relationship of health concerns to the ‘lifeworld’ is at the heart of Mishler’s (1984) critique of medical practice. The example below comes from the project examining the provision of opportunistic smoking cessation advice in primary care in England (Pilnick & Coleman, 2003, 2006, 2010). The patient has come to the doctor for injectable contraception (D 5 doctor, P 5 patient). Doctor 114, Patient 8 Dr:

Are you smoking again?

P:

yes

Dr: P:

How many? Only five

Dr:

That’s hardly worth it isn’t it?

P:

I know…But…when you’ve got the young ‘un running around…

Dr:

It helps with the stress?

P:

It’s that or take it out on her. It’s that or take it out on her

Dr:

yes that’s true

In this extract, the doctor topicalises smoking with a direct question that is not linked to any prior conversation in this consultation. He then seeks to quantify the patient’s smoking behaviour, and on receipt of her answer suggests that her smoking habit is so minimal that it is not worth sustaining. The patient agrees with this proposition, but goes on to provide a rationale for her continued smoking, though she does so implicitly. This implicit rationale is subsequently made explicit by the doctor: he asks a declarative question which also presents a candidate understanding (Pomerantz, 1988): that the stress of looking after a small child is relieved by the smoking. While the patient does not provide a direct yes/no answer that accepts or rejects the proposed understanding, she does produce an account which acknowledges the presence of stress and presents an alternative to smoking: taking the stress out on her child. The doctor accepts this account, and the issue of smoking is not raised again in this consultation. In this example then, and despite the fact that smoking behaviour is a relevant consideration for the prescription of hormonal contraception, the topic of

On Good Interactional Reasons for ‘Bad’ Healthcare Practice

75

smoking is closed with the doctor apparently reinforcing the patient’s expressed justification for smoking. Undoubtedly, there are broader sociological reasons for why doctors may respond in this way, as Pilnick and Coleman (2006) have noted, drawing on Strong’s (1980) work on alcoholism. In considering the discussion of alcohol in medical encounters, Strong highlights that if heavy drinking is a cultural norm for some groups of people, it is unlikely to be oriented to by those in that group as a medical problem. In a similar vein, the issue of mothers of young children smoking as a coping strategy has been well documented, and could also be argued to be a cultural norm among some socio-economic groups (Graham & Der, 1999; Sperlich et al., 2013). Strong’s analysis also underlines the difficulties inherent for doctors in attempting to handle a problem for which they have no medical solution to offer, where he suggests they run the risk of ‘stepping into a morass...quite unrelated to their normal medical remit’ (Strong, 1980, p. 41). Since the doctor in this consultation can offer no solution for managing the stress of caring for young children, he is also unable to pursue the issue of smoking. This example presents clear evidence of a ‘trouble’ in Sacks’ (1995) terms; stress is a state that is unambiguously bad, whatever the context, and taking this stress out on a child is to be avoided at all costs. Smoking behaviour, in this scenario, cannot be separated from the non-medical issues with which it is intertwined. However, there are also good interactional reasons why the doctor not only declines to pursue the topic, but also emphatically accepts the patient’s account. By receiving this account as a trouble, a doctor cannot pursue advice giving because, as we have seen, the roles of ‘troubles recipient’ and ‘advice giver’ are not interactionally compatible. By contrast, empathising with or condoning the patient’s actions are the appropriate conversational response of a troubles recipient. What this analysis suggests, then, is not only that ‘lifeworld’ concerns may not be as readily dismissed in medical encounters as Mishler’s (1984) critique suggests, but also that there are good interactional reasons for the way they are received. This example, then, is a further illustration of the tension that is inherent in PCC and picked up in studies which attempt to evaluate it: the potential mismatch between patient satisfaction with a consultation in the short term, and ultimate health outcomes. So far in this chapter, I have examined some of the ways in which interactional norms conflict in the pursuit of patient centred care. I have identified four specific examples of this: the desire to be non-directive vs the norm of meaning making; the problem of uncertainty vs the norm of auspicious interpretation and good news exits; the need to make sure patients are appropriately informed vs the norm that we do not tell people what they already know or give unsolicited advice; and the norm that a recipient to a trouble does not pursue advice giving. For the remainder of this chapter, I will turn to address a slightly different problem: the problem of conflicting moral norms and the difficulty of managing these interactionally.

76

Reconsidering Patient Centred Care

The Problem of Managing Conflicting Moral Norms in Interaction As I have argued from the outset of this book, PCC is largely treated as an unassailable moral good. However, as I have suggested in the brief consideration of bioethics in Chapter 1, the problem with moral principles is that, in practice, they can conflict, and moral analyses generally recognise this conflict. For the delivery of PCC, the conflict between autonomy and beneficence is particularly problematic, and this must ultimately be resolved at the level of interaction between professional and patient or client. However, this conflict is rarely recognised in assessments of the satisfactory or otherwise delivery of PCC. Instead, it is assumed that failing to enact items on a checklist represents failure – or at least a need for further training – on the part of healthcare staff. Below, I will present in more detail an example which illustrates how two moral imperatives may compete, returning to the idea we have already touched on in Chapter 2 about how the concept of patient centred care has to be weighed in both the long and the short term. This is significant because research to assess the adequacy of the delivery of PCC is usually focused on the short term of an interaction. However, in healthcare settings the patient’s longer term well-being may be equally, if not more, significant. This example also further illustrates the complexity of the way in which choices and decisions over what will happen next are managed interactionally, and the problems that arise through seeking to judge interactions by identifying or observing isolated components. The example is taken from data on inpatient dementia care collected for the NIHR funded VOICE project, where a CA approach was used in order to produce and implement a communication skills training intervention for healthcare staff working on acute medical wards (Harwood et al., 2018; O’Brien et al., 2018). Communication impairments are common with all types of dementia (e.g. Royal College of Speech and Language Therapists, 2014), and hospital environments can highlight or exacerbate communication difficulties, because of background noise and unfamiliarity of people and routines (Borbasi et al., 2006). The extract is therefore taken from a context where the usual norms of interaction may not always be observed. The day-to-day delivery of healthcare in any setting often involves tasks which may have clear moral motivations (such as encouraging someone to eat or drink), and where patient refusal may have longer term consequences. In the wider dementia care context, and in our dataset, patient refusal of these tasks is common. The policy context for dementia services has also been influenced by changing philosophical positions on what constitutes appropriate care. The drive towards patient centred care which now pervades many health contexts has been particularly influential in dementia care (see Kitwood, 1997), with an aim to recognise and value the personhood of the patient with dementia. The National Institute for Health and Care Excellence has recommended since 2006 that all acute NHS trusts should adopt a person centred approach for patients with dementia (NICE, 2006: CG42). Set alongside this, a number of studies have shown that healthcare professionals report experiencing the process of communication with people living

On Good Interactional Reasons for ‘Bad’ Healthcare Practice

77

with dementia (PLWD) as challenging (Alzheimer’s Society, 2009; Griffiths, Knight, Harwood, & Gladman, 2014). In the context of patient centred care, what is at stake in this example is how to deal with a proposed activity which is judged to be morally necessary, but where the activity is in conflict with the patient’s expressed wishes. In this specific example, the patient is at risk of developing pressure sores, which are painful, present an infection risk, and will prolong the hospital stay. To avoid pressure sores, the patient needs to avoid staying in one position for an extended period of time. In theory, whether to move or not is the patient’s choice, and, as we saw with the example of Alec in Chapter 2, it is not generally considered patient centred to override a patient’s clearly expressed views. In practice, however, health professionals are aware of the longer-term unpleasant consequences that may ensue for a patient who develops pressure sores (and they will also be aware of institutional policies e.g. those to minimise inpatient stays in order to return patients to their own homes for convalescence). The participants below are a nurse (HP) and a patient (PT), and after a greetings sequence the interaction begins as follows:1 Extract 1 133_206 5

I ↑was ↑just

HP:

6

wondering if I could help you with (0.4) relieving some pressure

7

on ye:r ↑botto:m

8 9

(1.0) PT: °°no::°° hhh

10 HP: no::?

This request, as initially formulated by the nurse, is what CA researchers would describe as a low entitlement, high contingency request (Curl & Drew, ¨ 2008; Heinemann, 2006; Lindstrom, 2005): the nurse in this instance does not assume the authority to make the patient stand, and way the request is formulated allows for the possibility of non-cooperation. Setting this request in the context of previous research on requests for action, both within and beyond healthcare, what has been demonstrated is that accepting is usually the preferred response, and even more so when the person who is asking has some position of authority relative to the person who is being asked. As a result, refusals are usually mitigated in some way (e.g. an account of why something is not possible is provided for the refusal) (Pomerantz, 1984; Pomerantz & Heritage, 2013; Schegloff, 2007). However, our dataset does contain a number of similar direct refusals, produced without any mitigation at all, which suggests that this may be a more common

1 This example is analysed in greater detail in O’Brien et al. (2020) in the context of a broader analysis of types of refusal and the ways these are dealt with by healthcare professionals in this setting.

78

Reconsidering Patient Centred Care

occurrence in people living with dementia (PLWD) than in the general population. The central analytic issue here is the issue of competing moral (and policy) imperatives: practising patient centred medicine in the immediate interaction vs preventing pressure sores. The patient has clearly said that she does not want the pressure on her bottom relieved (although the action associated with this – standing up – remains unspecified and it is possible it might not have been understood). In some instances in our data, we have observed that professionals do not pursue potentially difficult topics after an initial rejection by the patient (O’Brien et al., 2018). Here, however, the healthcare professional continues to orient to the clinical need to relieve the pressure, and so she pursues the request. Extract 2 10

HP:

11 12 13

(0.6) PT:

14 15

no::? (0.6) .hhh (0.4) we- I don’t want you to get a numb bu:m °(and our) charlie said °°because er I don’t always get the right number I don’t know°°

HP:

no:::?

In this continuation of the extract, after acknowledging the patient’s refusal with a questioning articulation of ‘no’, the nurse offers an explanation for why the activity is necessary. She frames this in colloquial terms – ‘avoiding a numb bum’ – and she also produces this explanation in a jokey tone. As O’Brien (2019) has noted, this colloquial explanation is in direct contrast with the initial request which was framed in clinical terms (‘relieving pressure’), suggesting the healthcare professional may be orienting to the possibility that the patient’s initial refusal is grounded in a lack of understanding of the initial request. Following the production of the explanation, the HCP pauses, perhaps for agreement- but none is forthcoming. What happens instead is that the patient produces some hard to hear talk in lines 13–14, and the fact that that it is unclear on the recording means it was likely also to be difficult for the HCP to understand. The reference to numbers possibly suggests that the patient has heard ‘numb bum’ as ‘number’. Nevertheless, and despite the unclear nature of the referent (and possibly the relevance) of the patient’s turn, the HCP acknowledges it, before going on to reframe the request, as follows. Extract 3 16

PT:

cos with me eyesi:ght

17

HP:

yea:h (.) well (.) ↑how ↑about .↑I just, stand you up

On Good Interactional Reasons for ‘Bad’ Healthcare Practice 18

79

he:re for a minute or two:. (0.8) just to get you

19

[off your

]5

20

PT:

°[(I think there)]°

21

HP:

5bottom (0.8) would that be o↑ka:y?

The patient’s utterance in line 16 has features of a mitigating account for why action is not possible, beginning as it does with ‘cos’, but the HCP does not respond specifically to this. Instead she acknowledges it but then produces another request for action, this time describing the specific action in more detail. Once again, this request is produced in a low entitlement format: it is a suggestive proposal (‘how about’) rather than directive, and explicitly seeks permission (‘would that be okay?‘). However, this version of the request is much more specifically formulated, and locates agency for the action with the nurse, ‘I stand you up’, thereby addressing any difficulties the patient may forsee with carrying out this activity themselves. It also specifies exactly where the action will take place (‘here’) and for how long (‘a minute or two’), and embeds the reasoning for the action (‘to get you off your bottom’). The nurse also uses ‘just’ twice, which downplays the action required and presents it as potentially less onerous for the patient. The contingencies displayed around this request, then, present a contrast with the presentation of highly entitled requests which have been examined in the CA literature (Antaki & Kent, 2012; Craven & Potter, 2010). In spite of these efforts by the nurse to reformulate the request, the immediate response from the patient does not suggest an acceptance of the proposal. The interaction continues as follows: Extract 4 22

PT:

[°(if you got we can)° (0.4) put me i:n (0.6) he::re,]

23

[PT tapping tissue box with right hand

24

(0.4) °.(for) I, can’t get, [(see the-)]°

25

[↑the:re,

27 28

[I ↑don’t] think you’d fit in

HP:

26

]

]

[HP touches tissue box] PT:

[°I then u::m,° (0.6) I told well I

29 30

PT:

[HP looks under PT’s table, picks up handbag] (0.4) [take ] that5

31

HP:

32

PT:

5an- (0.6) [stuck it (0.6) in the ↑pa:th,]

34 35

HP:

[HP put bag behind PT’s chair ] yeah, (0.4) ↑mauree:n, (0.6) can I ↑just ↑move ↑this -table out the way just for a min↑u:te,

36

PT:

°°yeah°°

37

HP:

is that al↑right?

33

]

[right]

80 38

Reconsidering Patient Centred Care PT:

39 40

°°this o:ne°° (0.8)

HP: 5whoop,[(.)↑let’s leave that on the table just for a]5

41

[HP touches tissue box, stopping it falling]

42 43

5minute, [(2.0)

44

[HP moves PT’s table to side]

45

HP:

]

tha::t’s ↑it, [(0.8) s’pull that do:wn (0.6)

46

]

[HP moves PT’s nightgown to cover knees] right, [(0.4) ↑am ↑I ↑alright just to ↑stand you up (.)]

47 48

[HP takes blanket from PT’s hands

49

]

just for a mo↑me::nt,

As the interaction continues, the patient orients to a box of tissues on her bedside table, tapping it with her hand, so that it appears her ‘put me in here’ (line 22) relates to this box. Once again the nurse uses humour to respond to this, but while she is ostensibly responding to the patient’s talk it is clear from the video that she is also preparing (by moving the position of the patient’s bag, asking permission to move the furniture, and rearranging the patient’s nightgown) ‘as if’ the patient is going to carry out the requested action. This ‘staging’ of the desired action displays a high sense of entitlement, in the sense that it assumes the activity will eventually take place, but it also serves to make it easier for the patient by removing any physical contingencies, therefore not just assuming but also making it more likely that it will happen. This section of the extract ends with another reframing of the request, again with low entitlement: the HCP asks permission for her actions (‘am I alright to’); again she locates the agency with herself so bypassing the question of the patient’s unaided ability (‘stand you up’); again she gives a specific duration (‘for a moment’); and the overall effort of the task is again minimised, with ‘just’ used twice in one sentence. However, even now the patient does not respond in the affirmative. Instead, she produces a mitigated no; her actions rely on some other factor which she doesn’t state. Extract 5 50

PT:

51 52

all ter depe::nds, (0.6)

HP:

53

[↑it ↑all de↑pe::nds? (0.4)

] on ↑wha:t,

[HP puts blanket down off camera]

54

PT:

ho:w big it i:s, (0.4) how wi:de it i:s

55

HP:

-o::h (0.4) [just standing

57

] ri:ght he:re

[HP points to floor by PT]

56 PT:

°e:::r° (0.4)well I don’t know til I’d sta::rt do I,

On Good Interactional Reasons for ‘Bad’ Healthcare Practice

81

Following the patient’s production of what the HCP repeats, and treats as ‘it all depends’, if the HCP wishes to continue to pursue the action, she needs to try to establish what the limiting factor is here: on what does it all depend? The patient’s response to this question is ambiguous, but the HCP treats it as a question about where the action will take place. She then specifies the action with ‘standing right here’, emphasising this with her gesture to the floor directly in front of the patient. She also once again mitigates the scale of the task with the downgrade ‘just’. At this point, the patient responds as if she has been asked if she is able to carry out the action – (‘well I don’t know til I start do I9)?, which projects that she will start it. The HCP capitalises on this projected willingness and the interaction continues as follows: Extract 6 58

HP:

-shall ↑we give it a go:: then,

59

[(0.4)

60

[PT puts hand out to HP who takes it]

]

61

PT:

ah yeah but which, (0.8) (↑karvens are here)

62

HP:

it’s just gonna be [right

64 65 66

PT:

[right he- yea:h,

HP:

[HP places one hand behind PT, holding other] -yeah? [(0.6) are you ↑ready?

HP:

69 70

] ]

[HP takes firmer grip on PT, bobs slightly]

67 68

] he:re

[HP points to floor]

63

[(0.6) one two three ↑u:p we go, (0.6)

]

[HP and PT move upwards together into standing] HP:

o:::h lovely,

This time, the HCP’s request is more direct, using the modal request form ‘Shall we give it a go’. However, the request is also collaboratively framed, so that the HCP retains agency in it and the formulation ‘give it a go’ provides for the fact that it is agreement to trying rather than a commitment to success that is being sought. By constructing the activity both as a joint one, and one which is to be attempted rather than necessarily completed successfully, she avoids any difficulty over the patient’s assessment of her own competence, and also any face threat that might ensue from a failure to complete it. As O’Brien (2019) observes, these kinds of exhortations to ‘give it a go’ or to ‘try’ are common in this dataset, and work towards removing possible grounds for not complying with a request. As this interaction continues, the patient puts out her hand to the HCP, who takes it quickly. However, there is still some possible reluctance expressed in line 61. This talk is difficult to understand, containing a non-word that sounds even on repeated listening like ‘karvens’, but the HCP again treats this hard to interpret talk as a question about where the task is to take place. In line 64 the patient

82

Reconsidering Patient Centred Care

produces an assent to the request and the HCP continues to physically support the patient to stand. ‘Up we go’ once again frames the action as a joint one, and once the patient is fully standing the HCP produces an affirming assessment (‘o:h lovely’). This interaction has been unpacked at length here because it both illustrates and problematises a number of issues in relation to the delivery of patient centred care. As noted above, unmitigated refusals in response to a request occur in this dataset, but they are rare in everyday interaction, with the exception of interaction with children. They are also particularly rare in healthcare interaction, where there is the additional issue of institutional authority or entitlement to make a request (see Stivers, 2006). Probably because these kinds of refusals are so rare in other kinds of interaction (meaning that professionals get little practice in dealing with them), they can be hard to manage interactionally. However, as we have seen in this example, professionals in this context sometimes have to make a judgement that respecting an initial refusal would result in greater harm for the patient, so they have to revisit the patient’s decision by pursuing the request. As in this example, when a healthcare professional wants or needs to pursue an action, the way they manage this is often to treat the initial ‘no’ as an indicator of ability, rather than willingness. This enables the request to be reformatted, to minimise the ability that is required, by working to shrink both the size of the task and the effort that is required from the patient: the use of ‘Just’ and ‘try’ are common, as is the reframing of the action as a collaborative task, or one that the HCP will lead on. On the face of it, this example may seem similar to those presented in the previous chapter, where the autonomy of a patient or client is also constrained. However, in those examples this constraint is carried out in the service of a larger order service objective: alternative courses of action are not in the end treated as equivalent because of the priority that an institution affords to one over the other, and organisational priorities win out. This means that, ultimately, the negotiation over these is limited. In this example, however, the tension is a fundamentally moral one, where self-determination, in the form of a choice not to cooperate with necessary care, has the potential to cause harm to the patient themselves. This difficulty of reconciling competing moral imperatives in interaction is one which was brought into stark relief by the formal evaluation of the VOICE study. Low entitlement, high contingency requests are the norm in interactions which seek to respect the autonomy of the person being asked, through formulations such as ‘Do you mind if…?’, and they are typically held up as examples of good practice and as components of seeking a patient or client perspective. However – and of course relatedly – they are also interactionally much easier to refuse. As the example above shows, this is a problem where the request is only being made in the first place because the action is seen as morally necessary from a beneficence perspective, since the refusal has to be managed and then pursued. This moral tension can be seen in the way that, as in this example, healthcare professionals in this setting often make initial requests in a way that suggest autonomy; however, this autonomy cannot necessarily be upheld if necessary care is to be delivered, and the negotiation over the request can then take considerable time. In the

On Good Interactional Reasons for ‘Bad’ Healthcare Practice

83

training intervention we developed from our study, we were able to use examples like these from our data alongside our knowledge of the interactional norms of requesting to successfully train healthcare professionals to frame essential healthcare activities in ways that make patients more likely to cooperate with them (Harwood et al., 2018; O’Brien at al., 2018). However, this work has also shown that Patient and Public Involvement observers are then more likely to view these interactions as ‘bossy’ or ‘controlling’. The obvious reason why observers make such an assessment is because they recognise the interactional manifestations of this underpinning moral conflict, meaning that the interactions do not follow the ingrained norms that are part of our ordinary social habitus. But understanding the management of this moral conflict requires understanding the interactional norms that underpin it. Unpicking these shows that practitioners are in a no-win situation: they can observe the interactional norms of requesting, but will likely fail to achieve patient cooperation, with the potential responsibility for the health risks or suffering that this brings. Alternatively, they can transgress the interactional norms of requesting, asking in a way that makes cooperation more likely and avoids longer term health complications, but will also be judged negatively if they do so. This analysis therefore helps us to better understand the challenge to effective healthcare delivery that is presented by the privileging of an absolutist version of patient choice or control, and to see that there can be good interactional reasons for what may be viewed from some perspectives as ‘bad’ healthcare practice.

The Limitations of Training in PCC as a Solution to Interactional Conflict in Healthcare In the previous chapter I considered the work of Heritage (2011) on what he calls ‘dysfunctions’ in healthcare practice. These dysfunctions occur when practices that work well in ordinary interaction are transported into healthcare interaction, but contextual aspects of healthcare interaction mean they are ill-suited. He demonstrates three scenarios where these occur in healthcare, and using examples from my own data I demonstrated a range of others, illustrating some of the reasons why both the delivery of patient centred care, and its consequences, may be more complex than its advocates have recognised. Heritage (2011), in his consideration of these dysfunctions, suggests that they can be addressed by better training of healthcare professionals. However, I argue here that there are two problems with this proposed solution. The first is that all such possible dysfunctions would need to be identified, in order to train professionals to recognise and avoid them. In the previous chapter, I have added to Heritage’s list, but also argued that there are undoubtedly more to be found in different healthcare contexts and interactional settings. In any case, even assuming that all of these dysfunctions could be identified, there is still a more foundational problem. The difficulties highlighted in this chapter involve more fundamental interactional conflicts, grounded in basic interactional norms that underpin all of our interactions. The interactional problem caused by the lack of alignment between being

84

Reconsidering Patient Centred Care

a troubles recipient and an advice giver, or of feeling unable to make meaning from something someone is struggling to assimilate, or of being judged for transgressing norms of requesting while navigating between competing moral principles of autonomy and beneficence, cannot be easily trained away, because these conflicts are embedded in what policy asks of healthcare professionals. For example, the offering of expert advice that a client is requesting potentially conflicts with the policy of non-directiveness, and the offering of choice and privileging of control potentially conflict with a duty to avoid harm. Rather than continue the cycle of training and retraining for patient centredness, it may then be more realistic and more honest to accept at the outset that there are interactional limitations to the application of the principle. This, I argue, is the inevitable consequence of the application of a one-size-fits-all moral principle to the complexity of healthcare work across a wide array of settings, in the absence of any grounding in an understanding of how healthcare interactions work in practice. Policies ultimately have to be talked into practice between healthcare practitioners and their patients or clients, and as the analysis in this chapter have shown, sometimes policies fail to work as they were intended to because, while they may make good organisational, or even moral, sense, they do not make good interactional sense. What my data also show is that sometimes policies explicitly direct professionals to interact in ways that go against interactional norms. Viewed from this perspective, it is easier both to understand why policies such as PCC might not have the impact we expect, and to see the importance of understanding the interactions, and the interactional norms, that those policies depend on. It also follows that any training initiatives to improve communication in healthcare might better be directed at helping professionals to understand why and how interactional difficulties can occur, and to think about the different ways that might be appropriate to manage these in different contexts, rather than aiming for their avoidance or providing blanket criticisms of their presence. However, there is also an additional piece to this jigsaw, which so far has only been referred to in passing. The issue of who knows what, and who is treated as having the right to know, is fundamental to PCC. Issues such as agenda setting, advice giving and deciding on an appropriate course of action in the short vs the long term are all bound up with the issue of expertise, and when and how it is seen as appropriate to exercise or invoke this. It is to examining the problematic relationship between expertise and PCC that I now turn.

Chapter 4

Rehabilitating Medical Expertise for the Twenty-first Century Famously, according to the then Secretary of State for Justice, Michael Gove, in a TV interview in June 2016, the British public has had enough of experts ‘saying that they know what is best and consistently getting it wrong’. While Gove’s quote has often been misrepresented, and was produced in the context of a criticism of the wider failures of economic forecasting, it can nevertheless be argued that a view of claimed expertise as potentially problematic can also be seen in the consumerist, choice-centred conceptualisation of medicine that underpins PCC. The challenges an absolutist version of individual choice presents to the delivery of effective healthcare are made visible in other current contexts in the United Kingdom, such as declining vaccination rates for some childhood illnesses such as measles, and the variable uptake of the COVID-19 vaccination amongst different sections of the population. This chapter will begin with a consideration of the way medical authority and expertise have been conceptualised sociologically since the 1950s, to further understand why it is that the healthcare professional/patient encounter has come to be conceptualised as a struggle for control. I will show that this conceptualisation is not borne out by the examination of actual healthcare interaction, which shows how patients and professionals construct expertise collaboratively. From this perspective, control is not a discrete property that can be ‘returned’. To unpack this further I will consider the distinction between epistemic authority (in simple terms, the right to know something) and deontic authority (the right to decide what should happen about something). I will argue that the rejection of paternalistic medical practice has, rightly, problematised medical deontic authority, but that it has at the same time problematised epistemic authority, and therefore medical expertise. However, as the data I present in this chapter will illustrate, sidelining epistemic authority in this way ultimately leads to patient abandonment. PCC places an emphasis on the ‘epistemics of experience’ (Heritage, 2013), but this runs the risk of creating a hierarchy which gives this primacy, turning healthcare issues into private concerns; the logical endpoint of this emphasis is then either in practices of patient affirmation which may fall short of care standards, or in placing patients in a situation where they cannot access expertise they want or need. Practices designed with the aim of empowering patients and improving care can therefore have the opposite effect.

Reconsidering Patient Centred Care, 85–109 Copyright © 2022 Alison Pilnick Published under exclusive licence by Emerald Publishing Limited doi:10.1108/978-1-80071-743-520221005

86

Reconsidering Patient Centred Care

As briefly discussed in Chapter 1, Parsons’ (1951) conceptualisation of medical practice, and the socially prescribed roles for patient and doctor that comprise this, envisages the doctor as a disinterested professional, using complex knowledge to act at all times in the best interests of his or her patient: what Parsons calls ‘affective neutrality’. Stevanovic (2021) notes that, while Parsons linked this specifically to the role of the physician, researchers examining the work of other professions have demonstrated that practitioners commonly refrain from taking an openly affective stance towards a matter under discussion (Drew & Heritage, 1992); indeed this is often seen as one of the hallmarks of professionalism. For Parsons, the doctor’s ability to act in this way was possible because the role of the doctor incorporated not only medical expertise, but also medical – and by implication social – authority. Parsons’ conceptualisation, however, glosses over the way that authority has to be achieved in practice, and I have already noted in Chapter 2 that part of Garfinkel’s impetus for the establishment of ethnomethodology, and the related development of conversation analysis, was to establish empirical foundations for previous theoretical abstractions (Rawls & Turowetz, 2021). In this vein, Stevanovic (2013, p. 20) reminds us that ‘authority is not primarily about someone claiming authority but about others accepting someone as an authority’. The authority of a doctor or other healthcare professional is therefore a relational, collaborative achievement between professional and patient. However, as Parsons recognised, and importantly for the argument that will be made in this chapter, authority is distinct from expertise. Expertise is commonly viewed as a professional competence held by an individual in a specific field; this sets up a distinction between an ‘expert’ and a ‘lay person’, who does not hold this specific field-based competence. As Abbott (1988:323) observes in his work on the sociology of the professions, ‘professionalism has been the main way of institutionalising expertise in industrial countries’, and medicine is a classic case of this institutionalisation. Nevertheless, just like authority, expertise is also a phenomenon which has to be talked into being in and through healthcare professional/ client interaction, and the examples we have seen in the previous chapters show how it may be invoked, negotiated and sometimes (such as in the case of linking smoking with a presenting illness, or advocating for antibiotics) contested. The straightforward picture of mutuality that Parsons’ role-based theory presents, then, becomes more complicated when examining actual interaction in contemporary healthcare. As I have already observed, some of these limitations can be attributed to Parsons’ structural-functionalist perspective which paid little attention to the specifics of how roles were enacted in practice, but it also seems reasonable to assume that there is a link to the way in which (the practice of) medicine and healthcare has changed since the 1950s. Writing shortly after Parsons, Szasz, Knoff, and Hollender (1958) recognised three factors driving changes in ideas and practices of medicine. Firstly, the influence of psychoanalysis and modern Western psychiatry, which as I have already described in Chapter 1, has been responsible for a change in the language used to describe healthcare as well as the ways in which it is delivered. Secondly, increased medical and social emphasis on chronic illness such as diabetes and

Rehabilitating Medical Expertise for the Twenty-first Century

87

arthritis; in these scenarios a doctor cannot rely on a simple exercise of authority or any conceptualisation of roles that assumes an active professional and a passive patient. Finally, a wider global change in social relations towards democratic patterns of behaviour, which has inevitably exerted a similar pressure on medicine. These changes have led other sociologists to critique the unproblematic mutuality that is assumed by Parsons, and to seek ways of explaining why there may be a failure to achieve this. Freidson (1970) detailed the inherent differences doctors and patients bring to the relationship as an inevitable source of conflict. One such difference is the fact that for the doctor this may be a ‘typical’ case, whereas for the patient it is specific and personal. Freidson (1970) argues that it is the function of the doctor to apply general knowledge to a particular patient, whereas patients seek to retain an acknowledgement of their particularity, and hence an element of control over their future. This mismatch of perspectives, Freidson argues, necessarily results in conflict, which has to be remediated by trust. Through the lens of the sociology of professions, the development of the medical professions is often viewed as a ‘collective social mobility project’ (Parry & Parry, 1976), which has been achieved through the successful pursuit of the occupational strategy of professionalism. Doctors have been able to raise their income and status in the community and to also to dominate the institutions and organisations of medical care. The result is that, in terms of their relationship with the state, they have been generally successful in securing a recognition of their privileged position. However, as Bury (1997) argues, the recognition that the world was shifting towards more democratic patterns first made by Szasz et al. (1958) has been seen in an erosion of hierarchical relationships more generally in late modern cultures. In addition, a focus on the control and assessment of risk, proliferation of screening programmes, and the creation of categories such as ‘prediabetic’ have widened the scope of the category of patient so extensively that Buetow (2016) argues that society itself has been made a form of patient. One consequence of this is the loss of individual relationships where the trust that Freidson (1970) identified as key to the minimisation of conflict could prosper. Set in the context of these developments, it is easy to see why the doctor– patient encounter specifically, and health professional–patient encounters more generally, have come to be conceptualised as a struggle for power and authority. Patient centred care speaks directly to this struggle, with its promise to return control to patients (Health Foundation, 2013) and to centre their choices and preferences. As Mol (2008, p. 46) observes, ‘the logic of choice is drawn into health care with the promise that it will free patients from the patriarchal rule of professionals’. However, she argues persuasively that professionals cannot simplistically be equated to feudal lords. While the state affords considerable power to professionals in some specific circumstances (in the UK, for example, the power to detain under the Mental Health Act (1983, 2007), or to decide on someone’s fitness to drive), Mol highlights how failing to observe medical advice rarely leads to sanctions, and a doctor cannot compel a patient to comply with advice or a course of treatment. The fact that people generally do so is because they want to get better (or at least to avoid getting worse). However, Mol’s work

88

Reconsidering Patient Centred Care

does not examine the details of healthcare professional/client interaction, and it has been my contention throughout this book that this detailed evidence is fundamental to understanding the failure of PCC to deliver improved health.

The ‘Struggle for Control’: What We Know From the Literature on Doctor–Patient Interaction As I have argued in the preceding chapters, the distinctive methodological orientation of CA and the findings that stem from this have important implications for understanding medical encounters and also for any attempts to reform them. Stuart Hall (2011) has observed the way in which neoliberalism casts the state as tyrannical and oppressive, and through the lens of PCC, the behaviour of health professionals appears problematically cast in the same way. However, and significantly, detailed analysis of interactions underlines that we cannot begin by treating aspects of social context (e.g. the respective power, social status, knowledge or authority of healthcare professionals and their patients) as exogenous and immutable factors that will always govern participants’ behaviour in predictable ways. Instead, what CA research shows is how this work is ongoing and is negotiated through the sequences of talk from which interactions are built. Boundaries of expertise, of knowledge, and of the right to determine what course of action should follow, are negotiated and determined in the moment as part of these interactions. In other words, the social reality of medical encounters is talked into being in real time, and through this process interactional dilemmas and their solutions emerge. This is the fundamental reason why, in contrast to much other research evaluating healthcare communication or the impact of PCC, CA research in healthcare settings focuses just as much on the interactional contributions of patients as it does of professionals. As Bogoch (1994) has argued, the Parsonian patient was largely one who could not understand or appreciate professional opinions, and therefore had no choice but to accept or comply with professional diagnosis and treatment recommendations’ (Bogoch, 1994, p. 66). However, as CA research has shown, and as the data I have presented in the previous chapters has illustrated, ‘It is by acting together that doctor and patient assemble each particular visit with its interactional textures, perceived features, and outcomes’ (Heritage & Maynard, 2006a, p. 19). Approaching the healthcare encounter from this perspective has resulted in some important findings about the way patients or clients participate in them, including the ways in which they are able to demonstrate agency (e.g. Gill, 2005; Gill et al., 2010; Stivers & McCabe, 2021; Toerien, 2021), that could not have been predicted or uncovered from a starting assumption that patient status removes this agency. Specifically, CA research has shown how patient ‘lobbying’ for a particular outcome can be done through the way problems are presented by the patient, how they respond to a clinician’s questions, and whether and how they contest the clinician’s characterisation of their problems (Stivers & McCabe, 2021). However, and despite these findings, conceptualisations of PCC as giving or returning control to patients continue to undermine this agency. It is somewhat

Rehabilitating Medical Expertise for the Twenty-first Century

89

ironic that a perspective which aims to empower fails to recognise this, or to understand the sophisticated ways in which patients choose to use their agency at different points in healthcare encounters. This latter point is important because patients are generally interacting with healthcare professionals since, as I have previously observed, there is something they want or need. It follows that they orient to what CA analysts have called ‘progressivity’ (Stivers & Robinson, 2006), in order to move the interaction along to its desired conclusion. And it also follows from this that another important feature of professional–patient interaction often missed by both ‘code and count’ and critical interactional analysts is that ‘interactional dominance’ by doctors or other healthcare professionals is not necessarily always and exclusively problematic, and that there may be good organisational reasons for what appears to be this dominance. The fact that professionals speak more than patients, or that they drive the questioning agenda, is to be expected given that the patient is generally presenting the professional with a problem that they cannot solve themselves, and which they would like assistance with solving. As Maynard (1991) contends, asymmetry in the form of physician control cannot be considered as an automatic effect of institutional processes; analysis of consultations shows that both parties to the consultation constitute and enact this asymmetry throughout the interaction. In part, he suggests, these patterns develop as a way of handling the interactional difficulties and doctor–patient encounter creates. These difficulties have long been recognised in medical sociology as a ‘doublebind’ situation for patients (Bloor & Horobin, 1975), in that they are expected to use their own judgement as to when it is appropriate to seek medical advice, but later to defer to the professional’s judgement when undergoing treatment. Heath’s work on diagnosis in general practice explicates the way in which ‘patients’ accounts of their illness or behaviour, and in particular the ways in which they attempt to justify having sought professional medical help, reveal a deep sensitivity to the asymmetries in the relationship between patient and doctor’ (Heath, 1992, p. 261). He goes on to consider the practical ways in which patients preserve, carefully and systematically through their talk, the differential understanding between their own assessment of their complaint and its professional medical assessment, or to put it more bluntly, between medical expertise and lay opinion. The ways in which patients do this is varied: they may for example, emphasise the subjective nature of their own understanding, or produce qualified accounts which mitigate against any claims to expertise. He concludes that patients display a ‘central concern to avoid any response which could serve to imply that the participants’ versions and assessment of the condition had an equivalent status’ (Heath, 1992, p. 262). This concern follows through into CA research on the receipt of diagnosis. A notable CA finding from the analysis of diagnostic encounters is that patients commonly withhold any assessment of a diagnosis when it is produced. The example below is taken from Heath’s (1992) data, and shows the way in which patients preserve the differential status between their own understanding of their complaint and its professional assessment; as a result, they also preserve the difference between medical expertise and lay opinion.

90

Reconsidering Patient Centred Care

Physical Examination (Heath, 1992, p. 242) 1

Dr

2

Yeah (0.3)

3 4

Dr

5

Dr

That’s shingles (1.2)

6

,that’s what it is: (0.2)

7

P

Shingles

8

Dr

Ye[s

9

P

10

Dr

11 12

P Dr

that it (.) it’s not itched ↑a lot No:::

13

P

But: and I couldn’t stand to touch it.

[Oh: I (.) did won: (.) be[cause the fact that it [no:

In his analysis of this extract, Heath points out how the doctor has to reconfirm the diagnosis in line 5 before any response is forthcoming from the patient. However, even after this prompting, the patient’s initial response at line 7 is simply a recycling of the diagnosis. It is only after a further reconfirmation by the doctor (line 8) that the patient initiates any discussion herself. This extract is an example, then, of how the assumption that a dearth of patient participation in consultations can unilaterally be attributed to undesirable behaviour by the doctor is an oversimplification. And whilst the lack of patient response to diagnoses might seem surprising, it is in fact exactly what would be expected given the nature of the double-bind. Any response to diagnosis which challenges the asymmetry of knowledge that has produced it will also inevitably undermine the patient’s grounds for seeking professional medical help in the first place. In a similar vein, Sharrock (1979) criticises mainstream medical sociology for acting as an indictment, since by describing the professional–client relationship as ‘oppressive’, it is both finding fault and apportioning blame. By constituting the doctor–patient relationship as a struggle for dominance, or for control, the implication is that every healthcare encounter is a struggle, and that if only patients were given the opportunity they would have much more to say and would be much more active participants. However, as Sharrock points out, ‘if the medical professional is reliant upon the way in which he [sic] structures his talk with the patient … for his control, then he [sic] is indeed dependent on the very weakest constraints which could not contain or control anyone who genuinely wanted to raise the topic’ (Sharrock, 1979, p. 142). What Sharrock’s argument points to, and what over four decades of CA studies in healthcare have shown is that, rather than always struggling for dominance and being doomed to failure, in most cases patients do not really try to contest the authority of the professional by seizing control. Instead, they recognise that a degree of what might be termed

Rehabilitating Medical Expertise for the Twenty-first Century

91

‘interactional submission’ is the fastest and most effective way to bring about a solution to the problem they present with. This ‘interactional submission’ is underpinned by what ten Have (1991) calls an asymmetry of topic, in that it is the patient’s condition, rather than the professional’s, that is under review. But this is only the first asymmetry: there is also a related asymmetry of task, in that – for example – a doctor’s work of diagnosis or management necessarily involves questioning, investigating and decision-making, in order to establish the possible courses of action and navigate between them. While patients take the initiative for the encounter – in that they decide which practitioner to consult at what point in the development of their problem – they ‘lose’ this initiative when the professional’s questioning takes over. The following example is from ten Have’s (1991, p. 142) data: 9.

D:

What’s up?

10.

P:

yes I have so much diarrhoea and it won’t go away

In his analysis, ten Have (1991) notes how the patient’s statement stops short of formulating any kind of request or establishing any kind of direction for the subsequent interaction. It gives a reason for the visit and provides the doctor with material on which further questioning could be based, but it also returns the questioning initiative to the doctor. By responding to the doctor’s initial invitation in this way, patients defer to the professional to decide what should be done next. Ten Have (1991, p. 145) concludes that ‘patients seem to disguise their interactional initiative by refraining from formulating requests, by giving the initiative back to the physician rather quickly, or by using quite subtle and covert devices to hold off the doctor’s questioning interventions’. However, this orientation to authority is far from one-sided: Per¨akyl¨a’s (1998) analyses of primary care data show how doctors demonstrate accountability for their diagnoses, by showing evidence for their conclusions through talk and action. In other words, they orient to a need to show patients not only what they judge to be the problem, but also why. What CA research in primary care has shown, then, is that although the encounter is likely to happen on the patient’s initiative, the ultimate goal of diagnosis means that the distribution of tasks involves, or in fact necessitates, quite ‘natural’ interactional dominance by the professional, which is enacted through questioning, investigating and navigating possible outcomes, and this is complied with by the patient. The implication then, at least for encounters that are based around diagnosis or treatment is that it would take specific and deliberate effort on the part of the patient to counter the interactional contingencies leading to asymmetry. This is rarely seen in practice, for good reason: it would hinder both the progressivity of the encounter, and the accessing of the medical expertise that is needed for the patient’s problem to be resolved. As Mol (2008) notes, what information is worth gathering is not a matter of choice for an

92

Reconsidering Patient Centred Care

individual patient in the consulting room. This, alongside Sharrock’s (1979) observations, highlights the problem of assuming paternalism on the basis of assessing interactions that do not take into account the dynamic nature of the talk they contain, or understand its structural properties. Conceptualising shared decision-making or patient centred care as a ‘gift’ from the active clinician to the passive patient (Berger, Galasinski, Scalia, Dong, Blunt, & Elwyn, 2022) is both inaccurate and, ultimately, re-establishes the assumption of the patronised patient that reform purports to overcome. At first glance, this foundational CA work on diagnosis in primary care may seem less obviously related to current debates about PCC. However, as I have argued throughout this book, choice is seen as a cornerstone of PCC approaches, and understanding how patients can be better facilitated to exercise this choice is seen as key to its successful implementation. There is a more recent body of CA work which has explicitly and specifically aimed to investigate the operationalisation of choice in healthcare treatments in a broader range of settings (e.g. Barnes, 2018; McCabe, 2021; Stevenson et al., 2021; Stivers et al., 2018; Stivers & Barnes, 2018; Stivers & McCabe, 2021; Toerien, 2021; Toerien, Shaw, & Reuber, 2013; Toerien, Reuber, Shaw, & Duncan, 2018). What this work shows is that approaches to treatment recommendations from doctors that may on the face of it seem very similar linguistically can be received very differently by patients, and hence have different outcomes. As Stivers et al. (2018, p. 1343) note, ‘when physicians recommend treatment, they adopt a position of greater...authority relative to the patient’, but the way this recommendation is actually enacted encodes this authority to a variable extent. Linguistic formulations such as pronouncements, suggestions, proposals, offers and assertions all reflect a different authority gradient, and have different implications for patient choice. In these terms, a simple assertion of possibilities by the doctor, which in Stivers et al.’s (2018) data typically involves some kind of generalisation about treating an illness and can therefore be treated simply as information provision rather than a concrete recommendation, represents the flattest authority gradient. And as Toerien and colleagues argue (Toerien, 2021; Toerien et al., 2013), by listing options rather than recommending a course of treatment, doctors relinquish a little of their authority, and a scenario presented as one where there is no single best option gives added weight to the patient’s right to choose. This work, then, adds further weight to the call to recognise the importance of the details of healthcare professional/patient interaction, and to understand that control is not a discrete property that should be considered as present for patient centred care and absent elsewhere. However, examining whether patients respond to the right to choose as universally desirable is an issue to which I shall return at greater length later in this chapter.

Who Knows and Who Decides: Epistemic and Deontic Authority In order to understand more why the notion of control is problematic in healthcare encounters, there is another body of research to which I will now turn. Recent work in conversation analysis has focused on the area of epistemics in

Rehabilitating Medical Expertise for the Twenty-first Century

93

interaction: examining how claims to knowledge are produced, understood and contested through talk (Heritage, 2012; Heritage & Raymond, 2005). This literature makes distinctions between epistemic status (who has access or primary access to an element of knowledge) and epistemic stance, which encompasses the position that a participant takes vis-`a-vis another participant’s knowledge. An example is the way in which a questioner may design a question to convey an unknowing epistemic stance, despite the fact that they may know or suspect some part of the answer (Heritage, 2012). A third element to such analyses is epistemic authority: whether participants have the right (or primary right) to particular elements of knowledge. For example, as I have already observed in relation to psychotherapy, and as Heritage (2012) describes more generally, participants are usually seen as having both primary knowledge of and primary rights over their own internal feeling states, while (as the examples at the outset of this chapter have illustrated) professionals are generally seen as having a primary right to diagnostic knowledge. Of course, having knowledge is only one element of a professional role, and the enactment of the role also includes the authority to do certain things with this knowledge (for example for a teacher to grade an assessment, or for a lawyer to draw up a contract). However, and as I have already argued, from an interactional perspective, such authority is not set in stone: a first speaker can make a suggestion or proposal, and a second speaker can acquiesce to or resist these, but also acquiesce to or resist the speaker’s right to make them. As Stevanovic (2018) puts it, a person can only direct action, not simply by claiming the right to do so, or the right to know best, but if this is accepted by other participants to an interaction. Stivers and Timmermans (2020) provide an empirical demonstration of this in their examination of instances where doctors engage in persuasion with the parents of paediatric patients, highlighting that such persuasion would not be necessary if authority could simply be assumed. In considering the exercise of authority, Bochenski (1974) makes an important distinction: between ‘epistemic authority’ and ‘deontic authority’. Stevanovic and Per¨akyl¨a (2012, p. 298) define these as follows: ‘epistemic authority is about knowing how the world “is”; deontic authority is about determining how the world “ought to be”’. This distinction is also critical for understanding the enactment of expertise. Stevanovic (2021) suggests that while expertise is typically expressed through the exercise of epistemic authority, via a demonstration of knowledge, it can also be articulated through the assumption of deontic authority, through assuming the right to decide. It follows that deontic authority suggests one has the right to expect compliance from others: as Stevanovic (2021:3) puts it: ‘deontic authority may be enabled and underpinned by epistemic authority, which is the case when other people are impelled to act according to the will of a person as they assume that his or her expert knowledge in the matter at hand goes together with good reasons to promote a specific line of action’. It is important to note here that Stevanovic is not writing specifically about medical interaction; her work is set in the context of church officials (pastors and cantors) taking part in planning meetings to make decisions about orders of service. Nevertheless, it has relevance for medical practice, and particularly in understanding the rejection of the

94

Reconsidering Patient Centred Care

paternalistic model of medical interaction that PCC is rooted in a reaction against. Stevanovic (2021) argues that construction of expertise through ‘a mere display’ of deontic authority is a paradox, where someone implies knowledge by claiming the right to decide unilaterally, and their actions entail the assumption that their expertise is taken for granted. I will argue here that this may well be an accurate depiction of how medicine has historically constructed expertise, so that it was often hidden. A paternalistic model does not require a display of the epistemic authority, and the expertise that underlies this, in order to arrive at a decision. The historical assumption that ‘doctor knows best’ overrides this requirement. In modern society, characterised by a more flattened hierarchy (Bury, 1997) it is easy to see why the exercise of deontic authority in this way would be rejected. However, I will go on to argue that in problematising deontic authority, we have also problematised epistemic authority, and therefore expertise.

Constituting Expertise in Interaction: Whose Rights and Whose Responsibilities? To display expertise in interaction requires more than just telling people what to do; as Goodwin’s (1994) careful analysis of expertise in action has shown, an expert is not only in possession of a particular body of knowledge, but also has specific ways of handling and utilising this. Though each will bring their own specific expertise to bear, individuals from different communities will see different things in the same object or event, just as a farmer will see different things in a patch of soil than will an archaeologist. Goodwin shows how participants build a particular ‘professional vision’, a socially organised way of seeing and understanding events that pertain to the specific interests of a particular professional group. And this difference in vision is precisely why patients bring problems for medical assistance: they can perceive something as a problem, but lack the ‘professional vision’ to understand the significance of this or the action that should be taken as a result. This process is then played out in the way that healthcare interactions unfold. As I have shown at the beginning of this chapter, patients actively defer to doctors in putting their problem on the table for diagnosis, since they recognise that they are authorities about their experiences but not about why their health problems occur. It follows that they exhibit caution when they offer explanations for their presenting problems, as Gill (1998) and Gill and Maynard (2006) have shown. Through all talk in any context, Gill (1998, p. 342) argues, ‘Individuals interactively construct a distinct social order where knowledge and authority are distributed unequally’ (Gill, 1998, p. 342). In healthcare consultations, this manifests in the sensitivity both doctors and patients display towards differential social roles and entitlements. Patients commonly show that they are entitled to know about the aspects of their health problems that they experience and observe (what Gill calls the empirical realm), and analysis of consultations from a CA perspective shows that doctors ratify patients’ authority in this area. However, when patients offer their own explanations for problems,

Rehabilitating Medical Expertise for the Twenty-first Century

95

they do so in ways that downplay their own knowledge about the causes of the problem, for example by prefacing explanations with explicit uncertainty, such as ‘I don’t know if…’. The data Gill presents clearly illustrate this (and are consistent with the work of others e.g. Heath (1992); Drew (1991)). The example below is taken from her dataset of US consultations between US primary care physicians and patients, as published in Gill (1998). Dr. Curtis with Patient Ives [1:46] 1

Ms.I:

I still ha::ve my: ordinary::: migraine

2

headaches, which I’ve ha:d for yea::rs?

3

(1.3)

4

Ms. I:

*And ah:* (.) they come and go so badly I: just

5

have to really wonder what triggers that.

6 7

I know I do have some *allergies*. So [that’s ] s:ome possibilities5

8

Dr. C:

9

Ms. I:

[.M hm,] 5an I know (.) .hh the do bother me

In this example, as Gill shows, the patient’s authoritative description of what she experiences (‘ordinary migraine headaches’) contrasts with her uncertain and cautious talk about what the cause or trigger might be: she ‘wonders’ about this, and presents some ‘possibilities’, but reserves the state of knowing for a description of her previous diagnoses, and her own mental state in relation to the migraines: that they ‘bother’ her. These kinds of evidentially qualified claims (for example ‘It feels like...’ or ‘It looks like…’) are common in primary care data, and as Heritage and Raymond (2005) note, they are downgraded claims, since while they do offer a possible interpretation, they are also compatible with this preferred explanation not being the case. Gill’s work and the work it builds on is important not just in its own right, in terms of showing how patients present their concerns to medical professionals, but also because it reinforces Drew’s (1991) observation that there is a distinction to be made between the cognitive state of knowing, and displays of entitlement to that knowledge. As Gill (1998, p. 345) puts it, ‘Patients’ displays of uncertainty are not necessarily evidence that patients lack knowledge about what is wrong. Rather, they can be conversational devices that patients use to display their lack of entitlement to a particular type of knowledge-that is, knowledge about causation’. In other words patients display the ordinary entitlement to knowledge in relation to their own first-hand experience but, as the double-bind identified by Bloor & Horobin (1975) suggests, are (generally) careful to avoid portraying any claim to authority in relation to analysis of what this means. It is seen as the healthcare professional’s job to make meaning from the information, and this

96

Reconsidering Patient Centred Care

explains why patients find it difficult when they do not, as we have seen in Chapter 3. Another manifestation of this orientation to differential entitlement is that patients also generally treat medical terminology as belonging to the other, so that even when this is known to them it is produced with an orientation that it is outside of their domain. Gill (1998) presents an example where a patient enquires about a blood test as an example of this: From Gill (1998) 1

Ms. B:

An’ also do you- do they do: uh .what is

2

it-, (.) Doctor Waters called it the

3

sma:c (0.2)

4

Ms.B:

S M A C hh

As she notes in her analysis, the patient here produces the medical acronym ‘SMAC’ with hesitation and self-repair, and also attributes her knowledge of the term to having heard it produced by another doctor. The fact that patients do this so hesitantly, if at all, shows their sensitivity to their relative lack of expertise. These interactional findings are supported by other work in the sociology of health and illness: for example, Henwood, Wyatt, and Smith (2003) in their study of women’s use of HRT, describe how their participants reported great concern not to appear to overstep the boundaries between ‘patient’ and ‘expert’. And based on her clerical data, Stevanovic (2021) argues that the act of a non-expert participant making a proposal is in itself a suggestion the matter would be worthy of joint discussion, which is why such delicacy is required. The fact that this phenomenon is found in her clerical data also suggests that this is an interactional norm which transcends the healthcare setting. However, other more contemporary examples of the way in which patients orient to their lack of expertise relative to healthcare professionals can be found in studies of the way in which patients introduce information from the internet into their consultations (Rapley, 2008; Stevenson et al., 2021) and how they manage the presentation of knowledge that this gives them. For example, Rapley (2008) presents an account from a juvenile arthritis clinic where, once a doctor describes a course of proposed treatment, a patient’s mother then comments she has already researched this on the internet, and retrieves a printout from her handbag. However, she uses her knowledge from the internet search to mark agreement with the doctor’s decision rather than introducing it prior to this to drive an agenda. This distinction between the right to know and the right to decide what should happen is fundamental for the evaluation of patient centred care. As Stevanovic (2021) shows, in non-medical contexts, expert participants balance compliance and resistance, invoking past or future decisions of their own, or patterns beyond both participants’ control, as grounds for other participants’ compliance with their decisions. In this way a non-expert speaker is excluded from those who have

Rehabilitating Medical Expertise for the Twenty-first Century

97

deontic authority in a particular matter, and so is excluded from a final decision about what should happen, on the basis they do not have access to the distinct experiential perspective that characterises an expert perception of things. It is hard to argue, if you are the person who this decision concerns, that this does not represent disempowerment. Medicine has therefore rightly moved on from this point. Epistemic and deontic authority are linked in that, as Heritage and Raymond (2005) assert, epistemic authority provides the basis of justification for deontic authority in a professional setting. However, this does not have to mean that the assumption of deontic authority inexorably follows from epistemic authority, or that epistemic authority is in itself fundamentally problematic. As we have seen in the examples above, as part of the fundamental way they organise their interaction, patients generally engage in substantial interactional work to orient to and preserve the expertise of the professional. Discarding or downplaying epistemic authority is therefore no solution to this problem: instead, reform would better be directed at considering how this epistemic authority can be exercised civilly and productively, for the benefit of patients. Simplistic assessments of paternalism, to which PCC is presented as the solution, are grounded in a lack of understanding about how healthcare interactions work in practice.

Expertise in the Context of Chronic Illness It is worth pausing here for a moment to reflect on the fact that the examples I have drawn on so far in this chapter are related to acute illness. Sociological research into those patients who live with chronic conditions suggests a very different picture in relation to expertise. For example, West’s (1976) foundational work on the management of childhood epilepsy examines interaction between doctors and ‘experienced’ patients or carers. In his analysis he suggests that, where these carers perceive that information is not being volunteered, or even withheld, this ‘leads parents to initiate challenges to the doctor, opening up the agenda and creating the situation where the doctor has to ‘work’ to construct his claim to competence’ (West, 1976, p. 28). Interactional evidence from consultations suggests that the way in which expertise is negotiated and constructed with patients with chronic illnesses looks quite different from those with acute illness; nevertheless, it gives further weight to the idea that ‘control’ is not a fixed property in these interactions which can be held by one or other party. One illustration of this is the use of jargon in medical encounters. In Meehan’s (1981) examination of the use of medical terms by doctors and patients, he draws on Barnlund’s (1976) suggestion that the use of jargon between members of a group can increase efficiency of communication, cultivate a rapport amongst members and provide a sense of common identity. However, from a patient centred perspective the use of jargon in communication is usually problematised, on the basis of the assumption that the use of this language precludes the possibility of patient understanding. For example, the UK National Institute of Clinical Excellence Guideline on improving the experience of care for people using adult services (updated 2021) instructs healthcare professionals to ‘avoid

98

Reconsidering Patient Centred Care

using jargon’ (2021: Section 1.5.6). However, in some settings in my dataset patients and/or carers themselves commonly use technical terms, and healthcare professionals appear to treat this body of knowledge as something to which the patient has access. The example below is taken from the data collected from the UK Department of Health funded project to examine pharmacist/patient/carer communication in a hospital outpatient setting. In contrast to the extracts reproduced from Gill’s (1998) primary care data above, there is little use of mitigators, qualifiers or questioning intonation around the medical terms that are used by the patients’ mother here, or any other interactional contingencies which might serve to suggest that she has limited access to this technical vocabulary (M 5 mother of patient, Ph 5 pharmacist). 38: jb/nc/op (simplified transcript) 1 Ph:

Fifty per cent then (.) for a week

2 3 M:

(0.3) Yeah

4 Ph:

Count’s up again is it?

5

(0.2)

6 M:

No (.) it’s down (.) He was on 150 last week

In this extract, the participants are discussing the tailoring of a chemotherapy regime to the patient’s white blood cell count. Following the results of regular blood tests, the medication dosages are adjusted up or down accordingly. However, if this extract were to be considered without the context provided above, there are few features to mark it as a professional/client encounter as opposed to a discussion between two professionals. In this context then, jargon may be conceptualised less as a problem to be avoided, and more as a claim to expertise by patients. Indeed, in other examples in this dataset, patients or carers react negatively to pharmacists’ attempts to avoid jargon, by responding to generic utterances about ‘medicine’ by producing the correct drug name, or reacting to attempts at extended explanation by waiting for the conclusion or upshot, and then asking ‘Why didn’t you just say that?’ (Pilnick, 1998). In this setting, this use of technical as opposed to vernacular vocabulary by both parties is one obvious way in which the asymmetry of expertise of the ‘typical’ professional/client encounter can be flattened, and is a clear demonstration of the everyday negotiation rather than imposition of expertise. Of course, there has been a longstanding recognition in the wider sociological literature that patients with chronic illnesses are of a different order from episodic patients in their interactions with the health care system, and this is a common criticism of the sick role model. As Freidson (1970) notes, in terms of Parsons’ (1951) description of the sick role, the sick person’s exemption from the duties of

Rehabilitating Medical Expertise for the Twenty-first Century

99

everyday life ‘is temporary, and its legitimacy conditional on trying to get well’ (Freidson, 1970, p. 234). This kind of temporary exception, however, is applicable only to acute illnesses; in chronic illnesses legitimacy is not conditional in the same way. Indeed, as Freidson indicates, it is generally believed impossible to recover from most chronic ailments. Instead, legitimacy comes from following medical advice, and from refraining from making demands on others that are seen as disproportionate to the illness. One way of displaying this legitimacy in terms of following medical advice is to demonstrate the understanding of it, including its technical language, which is what the mother of the patient does in the example above. It is important to recognise, then, that expertise can be used by patients not only to negotiate medical authority, but also to demonstrate cooperation with it; this is a further reason why its blanket problematisation can disadvantage rather than empower patients.

Expertise, Autonomy and Abandonment So far in this chapter, I have suggested that the problematic paternalism of medicine can be linked to the deontic authority that clinicians have traditionally assumed. As Stevanovic and Per¨akyl¨a (2012) have shown, normally in interaction, participants within an epistemic domain of expert knowledge use this as a basis to defend deontic rights: simply expressed, the person with the knowledge assumes the right to decide. I have argued that a reaction against the medical paternalism of the 1950s, and the problematisation of the decision-making rights embedded within this, have also led to a problematisation of expert knowledge which has found an expression in PCC’s focus on the patient perspective. However, data from actual healthcare interactions shows that patients continue to orient to this expert knowledge, displaying a sensitivity to their own lack of it or (as in the case of chronic illness) using it as a basis to demonstrate their engagement with medical perspectives. Viewing healthcare interaction from this perspective sheds further light on why patients may sometimes experience the abandonment described by Bosk (1992) rather than the autonomy aspired to by PCC. In interactional terms, the introduction of patient centred care has weakened the link between epistemic and deontic rights in healthcare settings, but has also undermined the epistemic authority of the professional. There is therefore a danger that we leave patients or clients unable to make the meaning they require for and from their situation. If this seems on the face of it unlikely, then this is not borne out by the consultations in my dataset. It is important to remember, in the context of healthcare decision-making, that clients do not simply ask questions of healthcare professionals because they are having trouble making their own minds up about what to do. They ask because, where a client is trying to decide upon a desirable course of action, a healthcare professional is seen by their client as having expertise that they want or need (Dingwall & Pilnick, 2020). The extract below is taken from the genetic medicine corpus I have previously described (Pilnick, 2002a, 2002b) and is a very straightforward illustration of this point, showing how clients do not treat all sources of information as equal in their decision-making

100

Reconsidering Patient Centred Care

process. The extract is taken from a consultation with a client with myotonic dystrophy, a type of muscular dystrophy that causes progressive muscle loss and weakness, and for which there is no cure. Immediately before the talk reproduced here, the client has told the clinical geneticist about a conversation with her mother, in which her mother has suggested that dietary change might help her condition. Rather than rejecting this suggestion outright, the geneticist has said that he has not seen any evidence to support it, and the consultation continues as follows:

Extract 5 C31197 917

C:

918 919

Cl:

920 921 922

I mean (.) it’s (0.2) because I’m of your mum’s generation so I have to be careful [what I say [yeah b-but you’re a learned man (.) you know more about it she just [doesn’t (.) I mean

C:

[well people like me too would like to have an (.) explanation that suits us

This extract clearly demonstrates how clients orient to the information gained from healthcare professionals as occupying a privileged position in relation to other sources. It also shows how exhortations to patient centredness can act at cross purposes in terms of expertise. While the clinician in this extract is extremely careful in his rebuttal of the mother’s suggestion (and indeed does not refute it definitively), the client is much more definitive in her rejection of it. She makes a clear statement in lines 919–920 about how she perceives the validity of the clinician’s statements relative to her mother’s. It is also notable that in making his rebuttal the clinician downplays his own expertise, responding to the client’s description of him as ‘a learned man’ by aligning with the client’s mother in line 921, and putting himself in the same group as her in terms of seeking an explanation for the disorder. In this extract then, there is a mismatch between what the client attributes to the clinician and what the clinician is prepared to acknowledge. Nevertheless, it is a good illustration of how an individual might clearly prefer a health professional to make or guide a decision for them, based on their expert knowledge in the area. As Mol (2008) observes, professionals, through the course of their work, collect patient experiences, and pass these on from one person to the next. Information such as particular side effects or how easily a treatment can be integrated into everyday life enable consideration of advantages and disadvantages relative to what is important to a particular patient. Doctors do not always follow the results of clinical trials or clinical epidemiological research because the parameters explored in trials do not necessarily map to the ends patients or their doctors want to achieve. Such expertise is often bracketed under the term ‘clinical

Rehabilitating Medical Expertise for the Twenty-first Century

101

judgement’, but this bracketing represents a significant epistemic resource. Mol (2008) argues that one problem with a focus on choice as something to be exercised in a consultation is that it hides the work that needs to be done to maintain that choice, which is an active engagement in care. However it is my contention here that what it also does is hide the expertise which underpins it. The work of Landmark and colleagues (Landmark, Gulbrandsen, & Svennevig, 2015) unpicks this issue in more detail. In a Norwegian secondary care context, they use conversation analysis to examine consultations where patients are offered a choice between invasive and non-invasive treatment options. They build on the work of Toerien et al. (2013) in examining the way doctors offer choice to patients, noting her observation that when doctors choose to use option listing rather than recommending a particular treatment, they relinquish a little of their authority to decide what should happen next. In their corpus, Landmark et al. (2015) demonstrate that a physician’s presentation of treatment options is often tilted in favour of one of the proposals that is on the table, but still allows the patient to make the final decision. However, this does not generally have the outcome of the patient unproblematically exercising this choice. Conceptualising this in terms of epistemics and deontics, whilst physicians give patients the deontic rights to allow them to select their preferred option, patients respond by themselves orienting to the physician’s epistemic and deontic rights. They therefore resist making a choice by claiming a lack of epistemic authority (e.g. ‘I don’t know anything about it’) or by making the decision contingent on the physician taking a stronger deontic stance (e.g. ‘If you think so’). In either case, the result is that patients avoid committing to the physician’s propositions and instead attempt to hand back not just the interactional initiative but also the initiative for the decision to the clinician. Landmark et al. (2015, p. 55) describe this scenario as a ‘two way struggle of avoiding sole responsibility’: patients resist assuming deontic authority based on lack of knowledge, whereas doctors resist assuming deontic authority because of a context – shared decision-making specifically, and patient centred medicine more generally – in which it is taken for granted that patient preference or experience should be the deciding factor. The end result is that, as Keevallik (2011) has observed, patients may be reluctant to choose even where there is an option they seem to favour, and account for this reluctance by foregrounding their lack of (‘expert’ medical) knowledge. These analyses point to a situation where the ‘struggle’ that PCC was designed to address has been replaced by a struggle of a different kind: to recentre professional expertise. This is, of course, not to suggest that patients will do whatever doctors ask, regardless of their own preferences. Patients can and do take deontic authority where they feel they have grounds to do so. However, CA research also shows that when they do, they orient to the need to account for this. In the example below taken from Landmark et al.’s (2015) data, the physician is discussing the possibility of uterine removal with a woman who has severe bleeding disturbances. A less invasive procedure has been suggested by the doctor as a possible first step, before the consultation continues as follows (the original data are

102

Reconsidering Patient Centred Care

presented in Norwegian and English translation, but for reasons of space only the English translation is reproduced here). Landmark et al. 2015: 65 GYN (uterus) 1

D:

2

mm, e:: there are some things that are related to surgery? complications with bleeding,

3

P:

m,

4. 5

D: P:

infection right? mm,

6

(0.5) D:

.hh h:ave that as a las:st resort. (2.0)

9

P:

.hhhhhhh[hh ((cross arms animatedly))]

10

D:

11

P:

7 8

12 13

[trust me. ]5 5I’m ve:ry dete(H) [rmi(hh)ned I can’t ha:ndle it any mo:re? eH:e.] [HeHeHe.hehehehehe] h .h oka:y,

In this extract, as Landmark et al. (2015) note, there are strong markers of dispreference throughout before the patient eventually assumes deontic authority. These include minimal responses and long pauses following the physician’s listing of complications of surgery and the suggestion it should be a last resort, as well as the crossing of the patient’s arms. When the patient finally produces an extended response, Landmark et al. (2015) show how she does so with a twofold account that is grounded in her presentation of both a strongly held stance, and longstanding personal experience of the difficulties her condition causes her. They argue that the laughter particles in the word ‘determined’ display awareness of potentially inappropriate patient behaviour, in rejecting a medically grounded recommendation. In other words, while patients can on occasion take a strong deontic stance, it requires significant interactional work to do so. This is in line with Versteeg’s (2018) work on the negotiation of scientific expertise that shows how, to resist deontic claims, a second speaker may accept the content of a prior turn but reject a claim about its relevance for the participant(s)’ situation or actions. On this basis a HCP’s conclusion about ‘what is best’ can be accepted as a general principle, but resisted in the context of a specific patient’s circumstances. This extract shows, then, how a professional claiming general medical expertise is not inherently problematic, in that it does not prevent the patient from trumping this with personal expertise related to their own circumstances. However, what the collected examples here also show is that clinicians who are reluctant to present this medical expertise in the service of choice may be actually

Rehabilitating Medical Expertise for the Twenty-first Century

103

undermining it. Patient centred care initiatives commonly reconceptualise the healthcare professional/patient encounter as a meeting with professionals from two different fields-the field of medical knowledge, and the field of personal experience and preference. This conceptualisation is grounded in the sociology of Mishler (1984), who argues that the patient is an expert in their life world, and that conflicts in the doctor patient relationship arise because of conflicts between the voice of medicine and the voice of the lifeworld. However, if the expertise of the life world is emphasised at the expense of medical expertise, rather than alongside it, it is easy to see that the result is potential abandonment. In Chapter 3, I demonstrated the way in which non-directive approaches in reproductive medicine can lead to the dismissal of questions from those explicitly seeking advice about choices. Bosk (1992) describes how, in his data, these decisions are framed as personal or familial, and so set outside the professional relationship. The data extract I present below is taken from the Hong Kong corpus of antenatal screening data (see Pilnick & Zayts, 2012, 2014, 2016; Zayts & Pilnick, 2014). Following initial screening, the pregnant woman in the consultation has received a risk factor for fetal anomaly of 1 in 244, which for the combination of screening tests she has had, is above the clinic’s ‘high risk’ or ‘screen positive’ threshold of 1 in 250. She attends the consultation with the doctor together with her husband (This consultation is conducted in Cantonese, but for reasons of space only the English translation is reproduced here. For more information on transcription and translation of these data please see Pilnick & Zayts, 2016) (D 5 doctor, P 5 pregnant woman).

LTMA 52. D: .h so but based on this(0.4)the cut off,(0.2)is set like that.h this 53.

(0.2).h is the cut off,which(0.3) means(0.5),the report is positive..

54. P: Um. 55. (0.4) 56. D: It depends on whether you and your husband::.)are(0.3) worried (0.4) 57.

and want to go for amniocentesis. Or(.)we can have an ultrasound,to see

58.

if roughly (0.9) eh::: we don’t see anything special,

59. P: Um.

In this extract, as is usual in these consultations, the results from the screening tests are presented as the driver for a decision with regard to future action (lines 52–53). On this occasion, only two of the three possible options in this scenario are explicitly presented: amniocentesis and further ultrasound scanning (lines

104

Reconsidering Patient Centred Care

57–58). The third option, to continue the pregnancy and take no further action, is not explicitly stated, though it arguably could be heard implicitly. However, what is more significant for the discussion here is the basis on which the practitioner in this instance suggests the decision should be made between these options – whether the woman and her partner are ‘worried’ (line 56). In Mishler’s (1984) terms, worry clearly belongs to the domain of the lifeworld. As Heritage (2012) describes in his consideration of epistemic authority and epistemic status, the rights to know and articulate inner feelings, thoughts and hopes are generally treated as a person’s to know and to describe. Worry is not only an inner feeling but also a subjective feeling, and so the doctor in this extract treats it as known by the pregnant woman and her partner, but not by her; in Heritage’s terms she takes an ‘unknowing’ epistemic stance towards the couple’s feelings. The implication of this epistemic stance is that any decision on which option to be take here cannot be grounded in medical expertise. Instead, it can only be made on the basis of feelings. This example is similar to those presented by Landmark et al. (2015) in their data, in the sense that the choices the patients in their research context are being asked to make (between invasive and non-invasive interventions) are placed within the epistemic domain of the patient and their subjective experiences of quality of life. However, I would argue that the implications of this particular extract are wider. In Landmark et al.’s (2015) data, the handing over of deontic rights to the patient generally occurs after the physician has presented their views of the benefits and risks of different options, drawing on their experience of the outcomes of these options for other patients. Different degrees of ‘sharing’ of a decision may be present dependent on how the doctor frames different proposals or options. However, what other people may have done is not considered relevant in this reproductive medicine context, lest it lead to direction by implication: that what ‘most people’ do is also seen as the ‘right’ thing to do. Whilst this may seem a pragmatic approach to dealing with a situation which is otherwise professionally irreconcilable for the doctor, it also has significant implications for shared decision-making. If it is suggested that a medical decision should be made solely on the basis of an internal feeling state to which a practitioner has no direct access, it follows that the decision is treated neither as one that the doctor has rights over, nor as one that can be ‘shared’, but instead one that the woman alone can make. Rather than seeing what Heritage (2013, p. 392) calls a struggle between the ‘epistemics of expertise’ and the ‘epistemics of experience’, this is instead a scenario where the epistemics of expertise are completely disavowed, leaving only the epistemics of experience. As I have shown in Chapter 2, the epistemic resources that healthcare practitioners have access to are commonly shared in the delivery of test results, in order to help patients understand what different results should be taken to mean (Maynard, 2003; Maynard & Frankel, 2006). However, any such contextualisations are absent in the extract above and almost entirely absent from the rest of this antenatal corpus, where they are usually limited to generic auspicious

Rehabilitating Medical Expertise for the Twenty-first Century

105

utterances such as ‘not so high’ or to repeats of clinic-level statistics for risk levels. The end consequence is that where, following the process of option listing (Toerien et al., 2013, 2018), women seek help in working out which is the best option for them, such help is not forthcoming. By setting the decision purely in the context of a patient’s feeling state, the epistemic resource held by the professional is completely sidelined. This example can also be used to expand the consideration of epistemic vs deontic rights presented here. In a context where non-directive practice is held as an ideal, it is easy to see why professionals do not generally assume deontic rights over decision-making; indeed, as I have argued above, to do so in any context would run the risk of returning to an outmoded paternalistic approach. Nonetheless, it is worth noting here that there is an important difference between proximal deontic rights and distal ones, where the former is defined as the right to give advice in a local context but not ultimately to decide what a person should do (Stevanovic, 2021). In practical terms, moving things beyond a here-and-now joint decision-making agenda of the kind seen commonly in acute primary care is one way to sidestep any problematic claim to deontic rights. However, on the evidence of my data, by refraining from giving any kind of advice, practitioners also avoid any engagement with proximal deontic rights as well as distal ones. Returning to Mol’s (2008) distinction between the ‘logic of choice’ vs the ‘logic of care’, a central argument that she makes is that good care has little to do with patient choice, and therefore focusing on patient choice will not improve healthcare. While choice and good care may sometimes complement each other, she argues that they more often clash, and that practices designed to foster patient choice in fact act to erode existing practices that were established to ensure good care. While Mol is arguing at the philosophical level, certainly it could be argued that the extract above is an interactional example of this, highlighting the way in which advantages and disadvantages of possible healthcare avenues can be turned into private concerns. The reframing of complex problems as simple matters of choice (Henwood, Harris, & Spoel 2011) does not necessarily bring with it the empowerment that PCC assumes. Indeed, as Mol (2008) asserts, offering choice can be as much a way of neglecting as forcing an action. I am not arguing here that such neglect is in any way intentional, but rather that examples like the one above are stark illustrations of how an emphasis on personal choice can lead not to an experience of autonomy but instead one of abandonment. It could of course be argued that the example I have discussed at length above is from a very specific setting, and therefore represents a very specific set of tensions around expertise in the context of non-directiveness. However, there are indications of the impact of the tension between epistemic and deontic authority, and the attendant difficulties of finding a place for professional expertise, in CA analysis of interactions across a range of healthcare settings. As I have noted earlier in this chapter, Stivers et al.’s (2018) CA work on treatment recommendations in primary care shows how treatment possibilities may be delivered as pronouncements, suggestions, proposals, offers or assertions. In their analysis of this, the authors note that all of these formats result in doctors abrogating one or other dimension of authority, either epistemic or deontic, depending on the

106

Reconsidering Patient Centred Care

formulation that is used. What makes the extract I have presented above such a stark example is that professionals are abrogating both dimensions. In addition, Landmark et al.’s (2015) work in secondary care which I have drawn on above also illustrates how, in that setting, the epistemics of experience are often used as a means to increase the patient’s deontic responsibility. They observe how in more high-stakes medical decisions, patients are given greater deontic authority- but paradoxically empirical evidence shows patients seem to require objective medical facts in order to make these decisions. This leads to a situation in treatment decision-making where ‘the patients claim that it is the physician’s assessment of medical necessity that grounds deontic authority, whereas the physicians claim that it is the patient’s subjective experience or preference’ (Landmark et al., 2015, p. 63). These wider findings also have relevance for the categorisation of consultations as ‘unilateral’ or ‘bilateral’, a distinction which is often made in relation to shared decision-making and where it is assumed in general terms, that the former is undesirable whereas the latter represents best practice. SDM generally assumes a bidirectional dialogue about action/treatment (Peek et al., 2010) and it is generally assumed that it is the patient’s perspective that is missing from this dialogue. In considering this uni/bilateral distinction, Collins et al. (2005) have emphasised the importance of external context, noting that in general, a more unilateral format was observed in their recordings of oncology consultations than in primary care diabetes consultations. In other words, clinical urgency can have an impact on how far a decision is treated as one to be shared. However, antenatal screening also offers a very limited time span for decisions, and for subsequent possible termination of pregnancy; in this setting the ethos of non-directiveness prevails over clinical urgency, with the result that women are left to make their own decisions on the basis of their feelings. This scenario also suggests that the distinction between unilateral and bilateral decision-making is less straightforward than has previously been assumed. In particular, the assumption that the professional is always the agent of unilateral decision-making does not hold. In this context decisions are not so much shared by doctors as devolved to pregnant women to take unilaterally, and the evidence from my data suggests that this autonomy of choice, and the associated authority to determine future action, can be experienced as a burden rather than a means of empowerment. This conclusion also has consequences for the conceptualisation of professionalism that was considered at the outset of this chapter, since it points to a double bind that rivals the well-established double bind that patients find themselves in (Bloor & Horobin, 1975). When epistemic resources that professionals have (in relation to testing processes, outcomes, etc.) get sidelined, this results in the conundrum that professionals have this knowledge by virtue of being professionals, but also by virtue of being professionals they decline to use it to impact on decision-making. As in many of the other examples I have drawn on throughout this book, they therefore find themselves in a position where however they act can be judged as unsatisfactory.

Rehabilitating Medical Expertise for the Twenty-first Century

107

From Professional Expertise to Patient Affirmation? The conclusions I have drawn so far in this chapter give weight to Landmark et al.’s (2015) observation that we have likely underestimated the outcomes of what constitutes a patient-centred approach, in the sense of assuming that these are always favourable. They also underline Toerien, Shaw, Duncan, and Reuber’s (2011, p. 319) assertion that the concept of choice ‘is not as simple as the literature may suggest, and that the simple course of telling clinicians to “offer patients more choice” may not achieve its objective’, or at least not in the way which was intended. The ultimate end point of ‘giving control to the patient’ is seen in affirmative care models, e.g. in transgender healthcare in the United States, where the clinician’s role is conceptualised as empathetically supporting the assertions of the client (Edwards-Leeper et al., 2016). From this perspective, clients’ understanding of their situation, and in particular their dysphoria and/or desire for medical treatment to address this, is not to be challenged or questioned. As Silverman and Bloor (1997) have observed, this kind of approach has its roots in a wider cultural movement where truth or authenticity is unproblematically associated with the revelation of inner experience. However in practical terms, such an approach officially removes dimensions of the epistemic resources clinicians might otherwise bring to bear, including information about the wide range of past experiences and life history narratives that clients bring to transgender health clinics (see Zottola, Jones, Pilnick, Mullany, Bouman, & Arcelus, 2021) and how these may relate to future trajectories. I have noted above the potential for patient abandonment in the absence of these epistemic resources, and the double bind professionals are placed in as a result. In the United Kingdom, the alleged practice of affirmative care in a transgender healthcare clinic for the under 18s has led to an application for Judicial Review from a previous patient, on the grounds that professional duty of care has been breached because clinicians should have conducted further exploration into possible reasons for her presenting distress rather than taking her own presentation of this at face value (Helyar, Jackson, Patrick, & Hill, 2022). The case also has implications for consent to treatments such as puberty blockers and cross-sex hormones which have long term consequences into adulthood (Bell and Mrs A v Tavistock and Portman Trust, 2020). The application for Judicial Review was upheld by the High Court, but subsequently overturned by the Court of Appeal. However, the subsequently published interim report of the Independent Review of Gender Identity Services for Children and Young People (The Cass Review) (2022), commissioned by NHS England, concluded that a fundamentally different service model is needed in this clinical field. In particular, it noted some staff concerns over pressure to take a purely affirmative approach, and recommended there should be agreement and guidance about the appropriate clinical assessment process that should take place before any treatment can be offered or provided. This situation illustrates once again that the assumed moral rectitude of patient ‘control’ is more complex than some conceptualisations of patient centred care allow. As Mol (2008) argues, making a decision is not like making a calculation, where all the variables are fixed. And fully informed consent to treatment, the importance of which is

108

Reconsidering Patient Centred Care

highlighted by the Bell vs Tavistock case, depends on a clinician sharing their expertise on the advantages and disadvantages of possible courses of action, both in the short and longer term. Without this, the fact that PCC and SDM have been shown to improve patient satisfaction in the short term, but do not produce a positive impact on health outcomes (Dwamena et al., 2012; Shay & Lafata, 2015), needs to be viewed with caution, and certainly more critically than is usually the case. While this measure may shed some light on the ‘success’ or otherwise of PCC, what is implicated in this finding deserves further consideration. This chapter began with a consideration of power and authority in the doctor/ patient encounter, set in the context of some of the major ways this has been theorised from the 1950s on. Returning to this theme, it is worth considering from an interactional perspective that power struggles are not necessarily about the boundary between voluntary submission to authority and resistance against illegitimate power, but, as Stevanovic and Per¨akyl¨a (2012) argue, the boundary between voluntary submission to authority and framing this submission as something that comes from the subject’s own will. As they suggest, in some contexts we may be less concerned about our ability to bring about consequences, and more concerned about our capacity to maintain our views about ourselves. Importantly, as Lukes (1978, pp. 639–640) argues, evaluating a possible course of action before agreeing (or disagreeing) with a proposal made by someone else demonstrates there is no ‘surrender of private judgement’ and therefore no face threat (Goffman, 1967); this does not represent any kind of ‘struggle’. This perspective is grounded in Goffman’s theoretical conception of ‘face’ as situated within ‘the flow of events in the encounter’ (1967, p. 7), rather than in isolated components of it. Heritage and Raymond (2005) describe how, ‘Participants’ concerns with face can be found in the management of rights and responsibilities related to knowledge and information’ (2005, p. 16), and the examples that have been presented in this chapter show some of the specific ways in which this is managed in and through healthcare talk. The end result is that in most contexts patients orient to professionals’ rights to healthcare knowledge and see this as a resource on which they wish to draw. While features of PCC and SDM, such as establishing a patient perspective and considering the advantages and disadvantages of different courses of action may be seen as valuable by patients, I have shown that adopting an approach of focusing on control and the choice that underpins this places patients in a situation where they cannot access expertise that they want or need. The arguments I have made in this chapter also show the wider problem with conceptualising PCC as fundamentally about a struggle for control. In any healthcare context absolute control is an illusion, because of the limits of what can be appropriately provided within a particular system, in terms of availability, resources etc. While healthcare has undoubtedly become more consumerised since the 1950s, this is not a straightforward kind of consumerism, and this is a theme which I will consider in more detail in the final chapter of this book. Affirmative care models are the closest thing to consumerism that the UK health service offers, and here the control that patients or clients are given in the short term has been retrospectively problematised by some of these clients as not being in their

Rehabilitating Medical Expertise for the Twenty-first Century

109

long-term interest. Control is therefore a problematic concept for both participants in the healthcare professional/patient interaction; we might do better to reject it as a principle relevant to healthcare than to frame it as a central plank of healthcare reform and as a result offer something that is ultimately either unattainable or with unwanted long-term consequences. The analysis I have presented here has also provided further empirical demonstration of the problems that arise in the context of medicine’s wider borrowing of concepts from psy-disciplines (Rose, 1998, 1999) that was first discussed in Chapter 1, highlighting some of the issues that a straightforward transposition of this kind produces. These difficulties crystallise around expertise: while psychotherapy may be able unproblematically to consider the patient/client the expert in their own condition, the examples presented here have shown how this is much more complicated for healthcare more broadly. The end result is that practices introduced with the intention of supporting patients or clients can ultimately have the opposite effect. However, it is one thing to identify these problems, but it is another to suggest solutions. In the final chapter of this book I will begin to consider what the way forward might be.

This page intentionally left blank

Chapter 5

Looking to the Future: Moving Beyond Patient Centred Care? In the previous chapter, I have argued that advocates for reform of doctor/patient interaction have grounded their reforms in both a misunderstanding of the way medical expertise is constituted, and the role it plays in healthcare encounters, locating this argument in sociological analysis from Talcott Parsons onwards. The consequence of this misunderstanding is illustrated by reviews of PCC and its components, which provide evidence for the fact that it makes people feel more satisfied with their healthcare consultations, but not that it improves their health outcomes (Dwamena et al., 2012; Shay & Lafata, 2015). Despite the widespread adoption of patient centred medicine in Western healthcare, then, the evidence for its success is at best mixed. Instead, we have arrived at a circular situation where the characteristic of consultations that training in PCC most improves is PCC as measured by the same checklist (Dwamena et al., 2012; Lewin, Skea, Entwistle, Zwarenstein, & Dick, 2001). I have argued that this is because PCC represents an a priori moral principle, applied in the absence of any understanding about how interaction in general, and healthcare interactions specifically, actually work in practice. As the analysis I have presented in the previous chapters of this book has shown, there are good organisational and interactional reasons for why PCC has proved so difficult to achieve in practice, and not only for why the expected impact has not been seen but also why its implementation may be problematic or undesirable for patients. Elwyn, Stiel, Durand, and Boivin (2011) (writing specifically about shared decision-making) identify the presence of a theory/practice gap, and the need to address this in order to bridge the difference between what is prescribed by policy and what is described by participants. Their proposed solution is that decision-making tools and aids should be better grounded in existing theories and models of decision-making. However, there are two problems with this assessment: the first is that it assumes the solution is to address this gap within the existing framework of SDM, and by implication PCC; and the second is that from a CA perspective, we need to rely not only on these participant descriptions or on idealised models, since they do not necessarily give us the level of detail we need to understand how health professional–client interaction both does and does not work in practice. The more generally espoused solution to this proposed theory/practice gap is to propose more, better, or different training for healthcare professionals. Failure to

Reconsidering Patient Centred Care, 111–125 Copyright © 2022 Alison Pilnick Published under exclusive licence by Emerald Publishing Limited doi:10.1108/978-1-80071-743-520221006

112

Reconsidering Patient Centred Care

train healthcare professionals appropriately in PCC would be one reading of the fact that close examination of medical interaction reveals it seems to have changed very little over the last 30 or so years (see Pilnick & Dingwall, 2011). Instead, I have argued that this failure was inevitable, as it is also grounded in a lack of understanding of what patients orient to as the purpose of medicine, and the ways in which expertise is talked into being in healthcare encounters. Of course, none of this is to suggest that doctors should not take time and care in talking to their patients, or that good communication is unimportant. In fact, what the evidence presented in this book shows is that it is perhaps even more fundamentally important than proponents of PCC recognise. However, because current conceptualisations of PCC are not built on an understanding of how communication works in practice, it is inevitable that it will fail. In this final chapter, I will consider in more detail the relationship between PCC and consumerism, before relating this to debates about autonomy that have been an ongoing theme in my analysis throughout this book. I will also consider how the delivery of PCC is problematised not just by the kinds of local contexts the examples in this book have illustrated, but also by wider sociocultural contexts where the availability of choice may be no more than an illusion. This raises the question of whether it is possible to usefully rehabilitate the principle of PCC, or whether it is better to abandon it altogether. I will argue that answering this question involves recognising the way in which social science has sometimes been co-opted in the advocacy for healthcare reform, in the absence of a critical sociological engagement with that reform. This co-option has led to an acceptance that ‘better’ communication skills training in PCC is the solution to problems identified in its delivery, rather than a wider reflection on what we consider or want the purpose of medical expertise in modern society to be. Finally, I will return to the problem first raised in Chapter 2: that many healthcare policies depend on interaction between individuals for their enactment; as with PCC, they have to be ‘talked into being’. This means that understanding why policies may not work as intended requires an understanding of human interaction. CA analyses illustrate that values-based, ‘one size fits all’ philosophies underpinning policy are problematic because local context is incredibly important in determining what might or might not be considered to be appropriately patient centred, but they also show that some of the components of PCC contain embedded interactional conflicts that are difficult or impossible for practitioners to resolve. This in turn illuminates why studies repeatedly find that practitioners ‘fail’ at delivering PCC, because crucially it also shows why they find it so hard to do what it expects of them, and why this cannot simply be addressed with more or ‘better’ training. I will argue that this ultimately points to a need to consider to what extent an understanding of interaction can be used to inform policy making from the outset, rather than simply using it as a tool to measure the success or otherwise of its implementation.

Looking to the Future: Moving Beyond Patient Centred Care?

113

The Limitations of Consumer Choice as Applied to Healthcare The analysis presented throughout this book has resonance for debates around consumerist medicine, in the sense that from a PCC perspective, medical treatment is seen as a commodity to be freely chosen by individuals. Though intended as a means for patient empowerment, the development of PCC (and the subsequent industry that has developed around it) can itself be seen as a deeply consumerist endeavour. Indeed, as noted in Chapter 1, the substitution of the term ‘patient’ for ‘consumer’ in Dwamena et al.’s (2012) updated review of patient centred interventions demonstrates the shift that has taken place over the last 30 or so years. Neoliberal thought is generally seen as defining consumerism around the wishes and choices of the individual, so that consumer choice is equated with individual choice, and as I have shown, PCC centres the idea of patient choice, through its narratives of ‘returning control’. However, O’Hara (2013) traces the way in which this particular understanding of consumerism was promoted by neoliberal think tanks, including the Institute of Economic Affairs in the United Kingdom, drawing on the ideas of the economist Alan Enthoven. Enthoven’s vision was of an internal market where a plurality of healthcare providers were paid to provide different medical services, but as O’Hara goes on to persuasively demonstrate, consumerism had a different meaning in the United Kingdom in the 1960s and 1970s; fundamental to this was this it was a collective rather than competitive concept, with a focus on the need to reform practice overall to improve patient care and to avoid the scandals and controversies associated with poor care that were otherwise inherent to the system. This conceptualisation also foregrounded the importance of patient participation, for example in the Community Health Councils set up in the 1970s. O’Hara (2013, p. 289) asserts that discourses of choice are ‘rooted in ideological pre-commitment rather than observation’, and this in turn helps to explain why the actual exercising of choice in healthcare interaction has proven to be rather less straightforward than reformers might have imagined. As Allsop and Jones (2008) have observed, the term ‘consumer’ is not a term derived from state-citizen relations, but from market economics, where consumers satisfy their wants in line with their ability to pay. While we may take individual choice as a consumer for granted in other settings, since NHS services in the UK have been free at point of service since inception, the straightforward application of the concept is obviously problematic. This links to the fact that the operationalisation of PCC in this way is also grounded in a misunderstanding of the purpose of healthcare professional/patient interaction in modern society: for the most part patients or clients are not consulting a professional simply because they can provide a particular service or treatment, but because they require the expertise of the professional in determining which service or treatment will best aid them to resolve their problem. O’Hara (2013) observes that the notion that choice is what patients want has partly become popularised through findings of opinion polls which ask voters whether they would like a choice of hospital or a choice of consultant; while these invariably show that they would (Le Grand, 2007) and while CA studies show these kinds of preferences freely asserted in consultations

114

Reconsidering Patient Centred Care

(Stevenson et al., 2021), this is a much more simplistic version of choice than the one that is currently centred in policy. Patients are happy to choose which hospital they are treated at, because those choices are likely to be driven by pragmatic factors such as location or length of waiting list, but the evidence suggests they are often much more equivocal about exercising choices in relation to more directly medical matters such as specific treatment regimes or courses of action. However, the two kinds of choice are often conflated, and as Mol (2008) notes, the concept of choice as a single moment in a consultation hides the work that needs to be done to maintain that choice, which is an active engagement in care. One obvious observation to be made in relation to this is that choice has become a kind of a sticking plaster, at least in a UK context. This sticking plaster can be applied to an underfunded, overstretched healthcare service, where an emphasis on what is available to the individual detracts from the structural issues. There is also, of course, the fact that the language of consumerism ultimately pushes responsibility for health problems back to the individual, so that PCC depends on the construction of a particular kind of patient for the empowerment it promises (Armstrong, 1983; Silverman & Bloor, 1997). As I have shown in Chapters 3 and 4, some patients clearly orient to choice as a burden which they feel ill equipped to carry, and in the context of a wider discourse of increased responsibility for one’s own health, it is easy to see why. Almost all of the data extracts I have presented in the previous chapters of this book show that professionals orient to the desirability of patient or client choice, and therefore devolve decisions to them to some degree. As Fotaki (2014) has argued, consumerist models of medicine are based on the economic premise of rational individuals who, when provided with the necessary information, will make a rational decision to maximise their welfare. However, as Stimson’s (1974) foundational work on non-compliance shows, real human behaviour has a complexity that is not captured by these models. There is also the problem, identified in Chapter 4, that the ultimate expression of consumerism in healthcare can be problematic for both patients and the medical profession. Offering support is not the same thing as doing what patients want, and as Mol (2008) notes, in a market there is nothing to limit futile trading. Affirmative care models which formally problematise aspects of professional expertise leave professionals open to at best criticism or complaint, and at worst legal action. How then might the issue of patient choice, and its operationalisation within the healthcare system, be reconceptualised beyond this consumerist framing? Williams (1976) has argued that consumption is an illusion created by producers to encourage purchase and waste; since healthcare is not easily commodified, its application is particularly problematic in this setting. O’Hara’s (2013) response to this is that choice does not have to mean an economistic concept of individual choice, and that consumerism does not have to be conceptualised in neoliberal terms. As he observes, consumer pressure groups in healthcare, which operate on behalf of those with a particular disease or condition, are often ‘consciously campaigning political groups’ (2013, pp. 290–291). It is therefore possible to reconceptualise consumerism as a social movement rather than an ideology, with costs and benefits, and rights and responsibilities, shared among all citizens rather

Looking to the Future: Moving Beyond Patient Centred Care?

115

than located in the act of individual choice. This perspective leads to the reframing of choice as being involved in your own care rather than necessarily determining it. This is echoed in Mol’s (2008) contention that ‘good care’ is not a description or an evaluation, but an intervention. Care is bad when people are neglected and not listened to, not when they do not choose. As she describes, for many patients who report episodes of deep dissatisfaction with healthcare services, the point of their complaints is not that others have bossed them about or told them what to do, but that they feel nobody cares about their predicament. As a result, she argues persuasively that we should see ‘making a choice’ not as an automatic prerogative of specific people (in this case, patients) but as a characteristic of specific situations.

The Limits of Autonomy? Unpacking the issue of choice in this way necessarily also requires a consideration of how we conceptualise and privilege autonomy. Privileging autonomy in relation to patients’ health choices potentially calls a fundamental tenet of Parsons’ (1951) sick role model into question: the idea that patients are not responsible for their illness as long as they comply with medical advice. As I have shown, the flipside of this can be the experience of choice as a burden, or as abandonment. In theoretical terms, this finding points to the continued relevance of Parsons’ work in understanding health and illness. However in practical terms, it also shows that the notion of the ‘expert patient’ risks overstating patients’ (desire for) independence and their knowledge levels (Fox & Ward, 2006; Timmermans, 2020). Consideration of autonomy as a moral principle also highlights the way it potentially conflicts with other ethical principles which underpin medical practice, notably beneficence. As Kaldjian (2013) has argued, the operationalisation of the principle of autonomy tends to focus on the patient as a participant in a consultation, but the principle of beneficence focuses on the patient as a person. Considerations of short-term vs long-term gain, and of relational rather than individual autonomy (Keller, 1997) therefore become relevant. And as Buetow (2016) contends and as the World Medical Association Declaration of Cordoba (2020) states, professional autonomy is also a key component of the physician’s obligation to alleviate suffering. Armstrong (2014, p. 171) suggests that ‘the active patient emerged phoenix-like from the ashes of professional autonomy’; that patient autonomy could only be achieved by the destruction of professional autonomy. In other words, where Parsons (1951) had seen professional autonomy as a force for good, in the sense that it was applied in a disinterested (or un-self-interested) way for the benefit of the patient, subsequent observers viewed this as self-interested authoritarianism. However this professional autonomy is viewed, there is certainly an argument to be made that an emphasis on the autonomy of patients has increased at the same time autonomy of clinicians has decreased. Montori et al. (2006, p. 34) recognise this when they write that ‘clinical practice guidelines coupled with quality audits (of processes and outcomes) and a system of penalties and incentives may push

116

Reconsidering Patient Centred Care

clinicians to ‘ensure’ that patients decide to use and adhere to ‘best’ therapies’. Their concern in highlighting this is the way that such guidelines and processes might impact on choice, and they ask the question ‘Would patient outcomes be better if the course of action is negotiated rather than dictated?’. However, as I have already observed, the evidence from the evaluation of PCC interventions (Dwamena et al., 2012) suggests that this is not the case. Instead, the failure of PCC to bring about the expected improvements in healthcare, and the fact that decades of research examining doctor/patient interaction does not show much change in how these interactions are actually conducted suggests that reformers’ desire to protect patient autonomy in the face of a growing emphasis on clinical practice guidelines may have been grounded in a misunderstanding of the purpose medicine serves in society, particularly in relation to the importance of medical expertise. Consideration of the limits of autonomy also returns us to O’Hara’s (2013) ideas around the rehabilitation of choice. Armstrong (2014) points out that we have tended to focus on conceptualising autonomy in medicine as a question of capacity (in relation to conditions such as dementia where such capacity may be impaired or lacking), so that we seek exceptional case-by-case reasons why autonomy should not be exercised rather than considering the problem more widely. However, it could equally be framed as a question of collectivism, thinking instead about relational autonomy (Keller, 1997) and interdependency (Young, 2001). As Mol (2008) observes, we do not live as individuals – we are family members, colleagues, neighbours etc. – and our actions impact on others. Similarly, healthcare is not delivered in a vacuum: there are competing claims from other patients, perhaps from family members, and from society as a whole. In the United Kingdom, treatment decisions are ultimately limited by guidelines from the National Institute for Health and Care Excellence (NICE) as well as local clinical governance protocols. Outside of the United Kingdom, in fee for service systems such as the United States, choice is inextricably bound up with finances. In either of these models, the privileging of patient autonomy therefore has the potential to divide physician loyalty to individual patients vis-`a-vis society as a whole (Buetow, 2016). One possible way forward is to shift our focus from patient autonomy and choice and instead focus on patient agency. Rapley (2008, p. 436) notes that ‘agency emerges in and through a web of intersubjectivity and relationality’, and it follows that decisions are made on basis of what is possible and practical in a given moment (what Flyvbjerg (2001) calls phronetic judgements). ‘No decision about me without me’, as the Kings Fund (2010) suggest as a manifesto in their report on how to deliver high quality patient centred cost-effective care, does not have to equate to the returning of control to the patient. It is possible to exercise agency in, and be engaged in, decision-making without seeing control as a property that can rest with only one party; this returns us to the importance of Goffman’s (1955, 1967) work on understanding how ‘face’ is constructed collaboratively in interaction, and to Silverman and Bloor’s (1997) contention that the discursive practices of PCC can ultimately result in an unhelpful redefinition of the ‘ideal’ patient. Accepting this shift would in turn indicate that the

Looking to the Future: Moving Beyond Patient Centred Care?

117

language used to describe the delivery of healthcare should also be changed, so that the ‘engaged patient’ (Timmermans, 2020) is offered patient-oriented or (as Buetow, 2016 proposes) patient-focused, rather than patient centred, care.

Choice and Expertise in Context: The Problem With One-Size-Fits-All As Fotaki (2014) notes, and as I have shown in the examples presented in Chapter 4, an inherent and inevitable imbalance of medical knowledge between healthcare professionals and their patients or clients means that trust in a doctor’s expert judgement may not easily or straightforwardly be replaced by choice. This observation highlights the importance of context. It is erroneous to assume that the only rival to PCC or SDM is outmoded paternalism (see Seale, Chaplin, Lelliott, & Quirk, 2006); instead there needs to be a recognition that no one model of practice can fit all settings. This is echoed in the work of Frank et al. (2010) who suggest that current avenues of research which apply checklists to consultations are likely to make little progress in researching how treatment decisions might best be shared between practitioners and patients. Instead, we need to recognise that this is a distributed process which can be affected by multiple other mediators beyond client participation, including local policies, protocols and the use of technologies. As the examples from different projects and settings used throughout this book have shown, healthcare consultations encompass a substantial variety of activities with different local goals. Whereas doctor-patient consultations in primary care may be fundamentally concerned with issues of diagnosis and treatment, other sites of practitioner–client interaction have very different purposes. For example, interactions may be administrative (e.g. admitting a patient to hospital), they may involve the delivery of therapy as part of the interaction (e.g. in physiotherapy) or they may be for routine monitoring (e.g. taking the temperature or blood pressure of a patient on a hospital ward). In some of these instances the interaction will have been initiated by the patient, but in others it will not have been. Some interactions will also incorporate the necessary training of student or more junior healthcare professionals. This diversity of settings and activities raises a correspondingly diverse set of issues, within which choice and decision-making play different roles, and so we cannot assume models, principles or policies that originally developed based on doctor-patient interaction in primary care can be transplanted wholesale. This problem is illustrated by a range of studies using conversation analysis that demonstrate how ‘blanket’ recommendations for best practice can be problematic in the face of local contexts and contingencies. I showed in Chapter 2 how what Heritage (2011) calls ‘dysfunctions’ can occur when practices that work unproblematically in ordinary conversation are transplanted wholesale into healthcare settings. But I have also shown how recommendations for practices that work well in one specific healthcare setting cannot simply be transplanted into another, and that, for example, the practice of asking people outside of a psychotherapy setting to set an agenda for their consultation

118

Reconsidering Patient Centred Care

can cause confusion or even discomfort (see Allwood et al., 2018; Pilnick, 2002b; Pilnick, Clegg, Murphy, & Almack, 2010, 2010 for more examples of these transplanted practices). Other CA studies underline the importance of local culture in understanding practitioner–client relationships, for example Penn et al.’s (2011) work on adherence to antiretroviral drugs for HIV/AIDS in South Africa, and Wang and Liu’s (2021) work on parental entitlement and influence in antibiotic prescribing for children in China. These studies show that the level of patient engagement in a consultation cannot straightforwardly be attributed to the skills or deficiencies of the clinician; as Timmermans (2020) argues, it may also represent a broader lack of cultural health capital on the part of the patient, or a lack of desire to be held responsible for the outcome. A further problem with checklist-based approaches to PCC and the shared decision-making model that underpins it is that, as I have noted above, it generally fails to recognise the distributed nature of decision-making. As Rapley (2008, p. 438) puts it, ‘much in the same way that conversation analysts describe the work that a turn-at-talk does, a consultation is related to prior consultations, the ongoing consultation, and future consultations’. Despite this, assessments of care delivery usually focus on what he calls ‘orphan consultations’: one off dyadic encounters in consultation rooms. Rapley goes on to argue that a recognition of this distributed nature should lead us to a more plausible but also more mundane set of ideals for the doctor–patient relationship. He suggests that while current approaches to care seek to combine the dominant threads of contemporary medical practice, including ‘active’ forms of patienthood and the promotion of treatment decisions surrounded in empirical evidence, they rarely if ever engage with the everyday work patients undertake to produce autonomy and expertise. This links to the point I have made in the previous section of this chapter and which is underlined by Rapley’s analysis: these models are grounded in a Kantian liberal version of autonomy as cognitive autonomy, where a decision is generated from within, on the premise that we exist apart from our social relations. However, Rapley’s (2008) interview-based data shows how shows how relatives or significant others may make meaning for patients where doctors decline to do so. The end result is that whilst, as I have shown in Chapters 2 and 4 of this book, clinicians may decline to make this meaning in the service of patient autonomy, a decision is still likely to be influenced by others, outside of a framework of medical expertise. The concept of choice, then, is meaningless without context, and the extent to which a choice can be considered ‘free’ will vary widely according to context. On this basis, O’Hara (2013, p. 293) argues that consumer choice has become, in post structuralist terms, ‘a symbol without meaning’. I suggest here that this is another factor which is inextricably linked to the inability of studies of PCC to show any impact on health outcomes: we have arrived at a scenario where choice is being offered as an ideological device rather than because it has any positive impact on the health of the person doing the choosing. And importantly, as Lewin et al. (2001) have highlighted, it is difficult to see how PCC could be implemented in low/middle income countries or in low-income settings within high income countries, where ‘patient choice’ may be no more than an illusion. Taking all this

Looking to the Future: Moving Beyond Patient Centred Care?

119

into account, it is clear that in order to improve practitioner–client relationships in the twenty-first century, we first need to understand them better, and to do this we need to move beyond the restriction of traditional models rooted in Western doctor-patient encounters, and towards an analysis which recognises not only the interactional complexity, but also the significant breadth and diversity that healthcare delivery entails. This conclusion inevitably begs the question: Do we want to retain the principle of patient centred care at all? I have argued above that we should consider a linguistic shift to patient-oriented or patient-focused care. However, whatever terminology is used, rather than assuming that these properties can be measured and assessed as something that does or does not take place, I argue that they would be better conceptualised as a continuum for practice, with an acceptance that different healthcare contexts may require practitioners and their patients or clients to negotiate a different position on this continuum. Given the importance of distributed decision-making highlighted above, the link between a moment of choice or autonomy in a healthcare interaction and a healthcare endpoint that evaluations of SDM or PCC search for also becomes problematic. Failing to recognise these problems leaves us trapped in the circularity where the only aspect of healthcare that is routinely improved by PCC interventions is the practice of PCC as judged by that intervention (Dwamena et al., 2012; Lewin et al., 2001).

The Complicity of Social Science As part of his broader and incisive history of the changing doctor–patient relationship, Armstrong (2014, p. 169) argues that the notion of expertise, and the privileging of patient ‘expertise by experience’ is bound up with self-management, which he describes as a strategy ‘incited by medicine and complicit social sciences’ so that patients have ‘had little option but to take the road of choice’. This issue of complicit social sciences, and in particular my own discipline of sociology, is one that deserves further consideration. Andrew Abbott (2001) argues that the breadth and reach of sociology is both a strength and a weakness. He suggests that while more single-minded disciplines such as economics may have an accepted right and a wrong way to conduct research, sociology does not. This makes its scope greater, but also brings with it greater responsibilities. He warns particularly of what he sees as sociology’s inability to ‘keep judgements about the rightness of things separate from judgements of their actual nature’ (2001, p. 198) so that as a result value judgements become mixed with scientific ones. This is problematic because of the potential it creates for sociology to become co-opted by other disciplines for their particular purposes. As I have previously argued (Pilnick, 2013), there is already evidence that the danger Abbott warns of has come to pass in relation to the medical profession, and I argue here that it can be seen crystallised in relation to PCC. In considering this it is helpful to reflect on a longstanding distinction that has been made in the subfield of medical sociology, because it presages Abbott’s warning. Robert Straus in 1957 made the distinction between the sociology of medicine and

120

Reconsidering Patient Centred Care

sociology in medicine. According to Straus, the sociology of medicine is concerned with studying factors such as the organisational structure, role relationships, value systems, rituals and functions of medicine as a system of behaviours. On this basis, research and investigation in this paradigm can best be carried out by persons who are independent of the formal medical setting, and who operate outside of it. By contrast, sociology in medicine is a collaborative endeavour, which may include research or teaching and which often involves the integration of concepts, techniques or personnel from many disciplines. Importantly, Straus suggests that these two types of sociology are likely to be incompatible with each other, for very practical reasons: the possibility that the sociologist of medicine may lose sociological objectivity if s/he comes to identify too closely with medical teaching or clinical research: and the risk that the sociologist in medicine will lose a good working relationship if s/he tries to critically study colleagues and their practices. Though this is an old distinction, it is certainly one that has contemporary currency. The Blackwell Encyclopaedia of Sociology (2007) states that ‘Sociology in medicine is the label given to the collaborative work between sociologists and medical or health personnel within medical institutions or health care organizations. This distinction represents the applied work of medical sociologists in the pure vs applied dichotomy of the social sciences. In its most extreme form, sociology in medicine encompasses sociological work aimed at the provision of technical skills and problem solving for the medical community while neglecting contributions to the parent discipline’ (Usher, 2007). I have written elsewhere (Pilnick, 2013) about the problems this can cause: not only might sociology in medicine ultimately become parasitical, in the sense that it feeds off the host discipline without contributing to the strength of that discipline, but it also means that the original purpose of sociological enquiry is lost. This can be seen in the way that social science has been co-opted in an attempt to bring about the delivery of patient centred care, rather than considering whose interests this philosophy serves in the first place. As Mol (2008) reminds us, if the history of medicine teaches us anything it is that good intentions can have bad effects. While there is no suggestion here that PCC has been conceptualised and implemented with anything other than the best of intentions, it is certainly the case that it has the potential – as some of the examples I have presented here have shown-to be at best unhelpful to some patients in some scenarios, if it removes the ability for doctors to make clear where expert knowledge leads to. This is not to say that we should not also address the inappropriate exercise of medical power or the paternalistic practice of (some) professionals. However, in the context of the doctor–patient relationship Mol (2008, p. 103) asserts that, ‘if criticism goes on and on it becomes mechanical’ and therefore tells us nothing new, and the loop of PCC training to increase the occurrence of PCC characteristics appears to fall into this category. Mol’s view is not a new one: Szasz and colleagues observed in the 1950s that any critical examination of the doctor–patient relationship usually predisposes towards change (Szasz, Knoff, & Hollender, 1958). They go on to warn that an awareness of cultural relativity should make us ‘skeptical of the assumption that

Looking to the Future: Moving Beyond Patient Centred Care?

121

our current practices are “good” or the “best possible”’ (Szasz et al., 1958, p. 528). However, whilst this is not a new observation, Mol (2008) argues that in the contemporary context, an analytic framework that became too exclusively adapted to the task of criticism of medical professionals created a vacuum which individual autonomous choice came to fill. What this reflects, in Szasz et al.’s (1958) terms, is a congruence of social expectations and socially shared ethical orientations. However, evidence from the evaluation of PCC in practice leads to another question. The strongest evidence for PCC or its key component of SDM is that it improves patient satisfaction levels in some contexts (Dwamena et al., 2012; Shay & Lafata, 2015). However, this does not necessarily equate with actual improvements in care and health outcomes. This leads us to consider a more fundamental questions about what we think the purpose of medicine is in modern society, or indeed what we think it ought to be. The invocation of patient centredness has come to be seen as a kind of magic shield, to respond to successive scandals in the delivery of healthcare care (in the UK these have included the investigation into the Mid Staffordshire NHS Trust and the subsequent publication of the Francis Enquiry (Mid Staffordshire NHS Foundation Trust Public Inquiry, 2013), and the use of the Liverpool Care Pathway (Department of Health, 2013)) and to guard against similar future scandals. These scandals have resulted in the more general rise of values-based policy, where moral values such as compassion or dignity are presented as the foundations of healthcare delivery. To that end, PCC has provided a moral shelter for practitioners. However, we will need to step outside this moral shelter of PCC as currently conceptualised in order to rehabilitate professional expertise and allow patients to benefit from it. Such a reconsideration requires social scientists to engage in more critical thought about what counts as good or bad healthcare practice, in terms more sophisticated than the presence or absence of individual autonomy or control. As I have shown through the data presented here, if we problematise expertise, abandonment is an inevitable outcome for some patients. The rehabilitation of expertise – and the recognition of the important distinction between epistemic authority and deontic authority described in Chapter 4 – is therefore fundamental to this endeavour. Viewing the medical encounter in this way also takes us back to Balint’s (1957) original conception of PCC, which recognises the asymmetry in the clinician–patient relationship. However, it does not seek to remove this but takes it as a given and instead considers how it might be better (i.e. more civilly and productively) enacted. There is much to be said for returning to this position, not least the fact that many years of interactionist research suggests that attempting to get rid of this asymmetry is ultimately fruitless (Pilnick & Dingwall, 2011). What Rapley (2008) conceptualises as a need to ground reform in more mundane ideals could be realised through centring the importance of medical expertise and considering how best it can be civilly exercised. Individual components of PCC, such as active listening by professionals or their attempting to understand patients’ expectations, undoubtedly have value. But that does not have to, nor should it, equate to a handing over of control. The importance of patient concerns, values and desires (Guadagnoli & Ward, 1998) can still be incorporated in a framework which also recognises the importance of medical

122

Reconsidering Patient Centred Care

expertise. As I noted in the first chapter of this book, Balint (1969) occasionally used the term ‘patient-oriented’ instead of patient centred; the fact that the latter term is the one which has been popularised says as much about the ideological and political climate over the last few decades as it does about the practice of medicine. However, the difficulty of attempting to reframe an existing concept is that it is difficult to separate it from its existing baggage. Buetow (2016) concludes his critique of PCC with an assertion that, to move healthcare forward in the twenty-first century, we cannot just make improvements within the structure of patient centred care; we need instead to reframe the clinician–patient relationship. He argues that current tools and approaches focused on assessing the presence or quality of PCC potentially exclude rich insight; I suggest that this also limits our ability to think differently, or to step back and consider the wider picture. As I have noted above, this is partly because it also co-opts social scientists as part of a medical project, so curtailing their wider critical engagement. Buetow’s (2016) own proposals for this reframing rest on focusing on the reciprocity of relational care between clinician and patient and the importance of recognising a duty of care extended by the patient to the clinician, and it is difficult to see how this would be achieved in practice. However, he also calls for a shift away from an obsessive focus on the individual consultation and towards a focus on the importance of experience-based co-design of services and care pathways. This proposal would provide a significant and important way of recognising and incorporating the epistemics of experience in a broader context, rather than, as current formulations of PCC do, placing it into battle with the epistemics of expertise. In Honneth’s (2014) terms, it also acknowledges the value placed in modern societies on self-determination by expanding the idea of what we consider to be part of the self. Throughout the preceding chapters of this book, I have shown how the operationalisation of the principle of PCC is problematic. However, in this chapter I have also elaborated on why the principle itself is problematic. While it may have had the positive effect of cautioning against healthcare that is too paternalistic, and highlighting the central importance of patients’ perspectives of illness and treatment (Dieppe & Horne, 2002), I have argued that it has not necessarily had the inclusive effect that its proponents claim and that at its heart its claims of choice and control are underpinned by a neoliberal conception of consumerism which risks transforming health problems into individual private concerns. In addition, throughout the book I have illustrated that, regardless of the view taken on the principle, there are good interactional reasons why it cannot straightforwardly be implemented in the way that has been popularised. In the concluding section of this book, I will address the question of the contribution a more detailed analysis of healthcare interaction can make to the way we understand, but also the way we implement, healthcare delivery.

Looking to the Future: Moving Beyond Patient Centred Care?

123

The Specific Contribution of Conversation Analysis The empirical analysis that has been presented in this book, along with the literature that has been used to ground it, is rooted in a conversation analytic approach. This approach stands in contrast to the work of critical interaction analysts such as Waitzkin (Waitzkin, 1991; Waitzkin & Waterman, 1974) and Mishler (1984), for whom patients’ attributions and doctors’ responses are seen as outcomes of social structural arrangements such as an asymmetrical distribution of power and authority. As I have argued, the underpinning philosophy of PCC buys into this analysis, through an assumption that ‘control’ of the encounter has been improperly wrested from the patient and needs to be restored. However, detailed interactional examination of consultation data has shown a dynamic and ongoing construction of unequal distribution of knowledge and authority in doctor/patient interactions (Gill, 1998). Rather than being a priori imposed, asymmetry is socially produced through patterns of interaction that are used precisely to navigate different distributions of knowledge and authority. In the final pages of this book I will consider how a conversation analytic approach to understanding healthcare professional/patient interaction can move us beyond the current circularity of PCC. At various points in the preceeding chapters, I have drawn on Mol’s (2008) exposition of the ‘logic of choice’ vs the ‘logic of care’. As Mol explicates, she uses the term ‘logic’ to talk of practices, and as a philosopher her interest is primarily in the rationale for and coherence of the practices rather than the details. Nevertheless, she acknowledges that ‘the communication skills on which consultations depend…are extensive…good communication is a crucial precondition for good care’ (Mol, 2008, pp. 87–88). It follows that such ‘logic’ could be further explicated by attending in more detail than Mol does to interactional practices. However, for some CA researchers, what this means is that they see CA as a tool to improve the delivery of PCC as currently conceptualised. For example, in Landmark, Ofstad and Svennevig’s (2017) study of the implementation of shared decision-making as assessed by two different SDM scoring scales alongside a CA analysis, they argue for a closer examination of language in its context in order to shed light on how patient preferences are elicited, and the consequences for patient involvement. Nevertheless, they conclude that comparing CA and SDM measurements can contribute to specifying the communicative actions that SDM scores are based on, and that future research should combine CA with quantitative SDM measurements. Similarly, Heritage and Maynard (2006b) argue that CA can be used as way to improve the delivery of PCC. CA is therefore presented as a way to improve the existing model, rather than as a means for critically engaging with it. Leaving aside the strange irony that the more we try to protocolise patient centred medicine, the more we run the risk of losing sight of the individual, I have already noted that this approach runs the risk of making social science complicit in an agenda set by others and where it is at best unclear what the benefits to patients are. By contrast, Collins, Drew, Watt, and Entwistle (2005) argue that conversation analysis (CA) is an ideal method for research in the field of healthcare

124

Reconsidering Patient Centred Care

communication, because of its rejection of a priori models. As they highlight, CA studies are not based on preconceptions of what patient participation ought to look like, and so rather than seeking to identify the presence or absence of existing key concepts or elements, they focus instead on how decision-making is enacted, and how the interactional configuration shapes opportunities for patient involvement. This means a CA approach can lead us to a much more detailed understanding of not only how particular practices can be enacted through talk, but also why they might not be. De Kok and colleagues (2018) make a similar argument: that to understand decision-making in healthcare we need to see how responsibilities and entitlements are negotiated and identities produced through talk. Adopting this approach foregrounds the problems of checklist and measurement approaches to produce ‘good’ healthcare: as I have illustrated, what counts as ‘good’ is contextually bound, and may involve navigating between competing organisational/professional principles and/or competing moral ones, resulting in the disruption of foundational interactional norms. CA analyses therefore demonstrate the ways in which ‘blanket’ recommendations for practice struggle in the face of local contingencies. Different types of healthcare place distinct demands on practitioners and patients and there may be different agendas and asymmetries at work. As Per¨akyl¨a et al. (2007, p. 140) note with regard to standardised recommendations for patient involvement, ‘The relevancy of [different] forms of participation ultimately arises from the overall goal of the encounter, as well as from the theory of healing that guides the interaction’. And as Pilnick, Hindmarsh, and Gill (2009) have observed, even within a single setting for healthcare delivery, policies and recommendations can raise challenges to practitioners who are dealing with patients with various levels of knowledge, expertise and commitment. It follows that healthcare interaction can only be meaningfully evaluated within specific sociocultural, health system and disease contexts (Matthias et al., 2013; Pilnick & Zayts, 2016). What a CA approach cannot provide, then, is a checklist or a set of rules, Instead, it encourages reflection on the consequences of different interactional practices and whether these are more or less likely to be appropriate or effective in a particular setting. While this might at first glance seem problematic for being insufficiently precise, in fact this avoids the problem noted by Mol (2008): that general rules have the advantage of being easy to use, but lack the specificity for particular situations. At the same time, as Bosk (1992) suggests, the essence of professional judgement is to know when the rules must be suspended, and certainly the ability and legitimacy to bring professional judgement to bear is a necessary skill for all health professionals. We need therefore to change the focus from the definition of morally derived ethical principles in their abstract form to their practical application by practitioners in local contexts, and patients or clients’ responses to this. However we respecify the models or definitions of ‘good practice’ that healthcare professionals are asked to work within, we will not remove these underlying interactional contingencies and local negotiations. Changing the focus to issues of the delivery and receipt of information also moves the debate firmly from philosophical to social scientific questions (Pilnick, 2002a). Conversation analytic research can therefore help articulate what Mol (2008) calls

Looking to the Future: Moving Beyond Patient Centred Care?

125

‘doctoring’ actually looks like in practice, not by producing guidelines or checklists, or presenting itself as some kind of magic bullet to bring about patient centredness, but by explicating and demonstrating the ways in which care can be negotiated and adjusted to patient experiences. This chimes with Mol’s (2008) contention that it is not necessary to create suitable ways of attending to patients: they are already found in the consulting room but they can be highlighted, celebrated and improved upon. Examples of work which identifies and explicates these practices already exist (Harwood et al., 2018; McCabe et al., 2016; O’Brien et al., 2018; www.realtalktraining.co.uk) and show how CA work can be used effectively to bring about change. A CA analysis also enables us to move us away from ‘doctor blaming’, and the repetitive but ultimately futile cycle of criticism that Mol (2008) identifies. As I have shown throughout this book, what may appear simplistically to be examples of ‘bad practice’ by healthcare professionals may arise out of competing interactional contingencies that cannot be reconciled. Pilnick and Dingwall (2011) have previously critiqued the universal pursuit of patient-centredness, and I have developed that argument here, arguing that asymmetry is simply endemic to the medical enterprise as a result of medical expertise, and it may ultimately be our view of this asymmetry that is problematic rather than its continued presence. As participation in decision-making has come to be seen not only as a right, but also an expectation of the ‘good’ patient, so, as Rapley (2008) notes, rights and responsibilities have been redistributed. Specifically, providers are able to share the burden and transfer some of their responsibility to patients or clients. This aspect is illuminated by a CA analysis but is largely absent from any other considerations of the (assumed benefits of) patient centred care. A more detailed analysis of healthcare interaction also points to the importance of distinguishing deontic authority from epistemic authority, and thereby providing a means to rehabilitate medical expertise without reinstating medical paternalism. As Stevanovic and Per¨akyl¨a (2012) have argued, this interface between epistemic and deontic authority, and the way in which it is operationalised in interaction, requires further research. It is clear, then, that to properly understand the impact of an approach such as PCC requires a detailed interactional analysis of how it is deployed and treated across a range of healthcare settings. However, and perhaps more fundamentally, it also shows the importance of grounding the development of policies and recommendations in a solid understanding of actual healthcare practice. Broad moral principles used to underpin values-based policy may be attractive to policy makers, but as I have shown there, they may ultimately place practitioners in intractable and unworkable dilemmas, where however they act, they will potentially be judged unsatisfactory. In the longer term, then, the issue to be addressed is how we can advocate for an understanding of healthcare interaction to be used to inform policy making from the outset, rather than just providing a barometer for the success or otherwise of its implementation.

This page intentionally left blank

Appendix: Conversation Analysis Transcription Symbols (as described in Jefferson, 2004)

[Hi there] [Hello ] (0.6) (.) , Hi there 5 5Hello

Square brackets indicate onset and end of overlapping talk Figures in brackets indicate length of pauses (to tenths of a second) A dot in brackets indicates a pause of less than 0.1 seconds A comma indicates a gap which is too short to time

Equals signs indicate latched utterances between which there is no pause And so:o A colon indicates the prolongation of the previous sound or syllable So really? A question mark indicates a rising intonation (not necessarily a question) So really. A full stop indicates a stopping fall in tone (not necessarily a sentence end) And THEN Words in upper case are spoken more loudly Underlined words are emphasised or stressed comparative And then to other talk she said °no° Degree signs indicate talk which is quieter than the surrounding talk ↑ A shift to higher pitch immediately following arrow ↓ A fall in pitch immediately following arrow ,. Brackets denote an utterance delivered at faster pace than surrounding talk ((moves papers)) Transcriber’s description rather than transcribed talk (…) Unintelligible talk Audible aspirations (hhh) and inhalations (.hhh) are inserted in the speech where they occur.

This page intentionally left blank

References Abbott, A. (1988). The system of professions: An essay on the division of expert labour. Chicago, IL: University of Chicago Press. Abbott, A. (2001). Chaos of disciplines. Chicago, IL: University of Chicago Press. Allsop, J., & Jones, K. (2008). Withering the citizen, managing the consumer: Complaints in healthcare settings. Social Policy and Society, 7(2), 234. doi: 10.1017/S1474746407004186 Allwood, R., Pilnick, A., O’Brien, R., Goldberg, S., Harwood, R., & Beeke, S. (2017). Should I stay or should I go? How healthcare professionals close encounters with people with dementia in the acute hospital setting. Social Science & Medicine, 191, 212–225. doi:10.1016/j.socscimed.2017.09.014 Alzheimer’s Society. (2009). Counting the cost: Caring for people with dementia on hospital wards. London: Alzheimer’s Society. Antaki, C. (2013). Two conversational practices for encouraging adults with intellectual disabilities to reflect on their activities. Journal of Intellectual Disability Research, 57(6), 580–588. doi:10.1111/j.1365-2788.2012.01572.x Antaki, C., & Kent, A. J. (2012). Telling people what to do (and, sometimes, why): Contingency, entitlement and explanation in staff requests to adults with intellectual impairments. Journal of Pragmatics, 44(6–7), 876–889. doi:10.1016/ j.pragma.2012.03.014 Antaki, C., & Webb, J. (2019). When the larger objective matters more: Support workers’ epistemic and deontic authority over adult service-users. Sociology of Health & Illness, 41(8), 1549–1567. doi:10.1111/1467-9566.12964 Arminen, I. (1998). Sharing experiences: Doing therapy with the help of mutual references in meetings of alcoholics anonymous. The Sociological Quarterly, 39(3), 491–515. Retrieved from http://www.jstor.org/stable/4120804 Armstrong, D. (1983). The political anatomy of the body. Cambridge: Cambridge University Press. Armstrong, D. (2011). The invention of patient-centred medicine. Social Theory & Health, 9, 410–418. doi:10.1057/sth.2011.13 Armstrong, D. (2014). Actors, patients and agency: A recent history. Sociology of Health & Illness, 36(2), 163–174. doi:10.1111/1467-9566.12100 Balint, M. (1957/1964). The doctor, his patient and the illness. Edinburgh: Churchill Livingstone. Balint, E. (1969). The possibilities of patient-centred medicine. Royal College of General Practitioners, 17, 269–276. Retrieved from bgjp.org.content/17/82/269 Balint, M., Ball, D. H., & Hare, M. L. (1969). Training medical students in patient-centered medicine. Comprehensive Psychiatry, 10, 249–258. Barnes, R. K. (2018). Preliminaries to treatment recommendation in UK primary care: A vehicle for shared decision making? Health Communication, 33(11), 1366–1376. doi:10.1080/10410236.2017.1350915

130

References

Barnes, R. K. (2019). Conversation analysis of communication in medical care: Description and beyond. Research on Language and Social Interaction, 52, 300–315. doi:10.1080/08351813.2019.1631056 Barnett, N. (2018). Person centred over patient centred care: Not just semantics. Pharmaceutical Journal, 10(4). doi:10.1211/PJ.2018.20204578 Barnlund, D. C. (1976). The mystification of meaning: Doctor-patient encounters. Journal of Medical Education, 51, 716–727. Barry, M. J., & Edgman-Levitan, S. (2012). Shared decision making-pinnacle of patient-centered care. New England Journal of Medicine, 366(9), 780–781. doi: 10.1056/NEJMp1109283 Beach, W. A. (1996). Conversations about illness: Family preoccupations with bulimia. Mahwah, NJ: Lawrence Erlbaum. Beach, W. A. (2013). Introduction. In W. Beach (Ed.), Handbook of patient-provider interactions: Raising and responding to concerns about life, illness and disease. New York, NY: Hampton Press Inc. Bell and Mrs A vs Tavistock and Portman Trust. (2020). EWHC 3274. Berger, Z., Galasinski, D., Scalia, P., Dong, K., Blunt, H., & Elwyn, G. (2022). The submissive silence of others: Examining definitions of shared decision making. Patient Education and Counseling. doi:10.1016/j.pec.2021.10.026i Bloor, M. J., & Horobin, G. W. (1975). Conflict and conflict resolution in doctor/ patient interaction. In C. Cox & A. Mead (Eds.), A sociology of medical practice. London: Collier-MacMillan. Bochenski, J. M. (1974). An analysis of authority. In J. A. Frederick (Ed.), Authority (pp. 56–85). The Hague: Martinus Nijhoff. Bogoch, B. (1994). Power, distance and solidarity-models of professional client interaction in an Israeli legal aid setting. Discourse & Society, 5(1), 65–88. Bolden, G. (2006). Little words that matter: Discourse markers ‘So’ and ‘Oh’ and the doing of other-attentiveness in social interaction. Journal of Communication, 56, 661–688. doi:10.1111/j.1460-2466.2006.00314.x Bolden, G. B., Angell, B., & Hepburn, A. (2019). How clients solicit medication changes in psychiatry. Sociology of Health & Illness, 41(2), 411–426. doi:10.1111/ 1467-9566.12843 Borbasi, S., Jones, J., Lockwood, C., & Emden, C. (2006). Health professionals’ perspectives of providing care to people with dementia in the acute setting: Toward better practice. Geriatric Nursing, 27(5), 300–308. doi:10.1016/j.gerinurse.2006.08.013 Bosk, C. L. (1992). All god’s mistakes: Genetic counseling in a pediatric hospital. Chicago, IL: University of Chicago Press. Bosk, C. L. (2000). The sociological imagination and bioethics. In C. E. Bird, P. Conrad, & A. M. Fremont (Eds.), Handbook of medical sociology (5th ed.). Upper Saddle River, NJ: Prentice-Hall. Braddock, C., Fihn, S., Levinson, W., Jonsen, A., & Pearlman, R. (1997). How doctors and patients discuss routine clinical decisions: Informed decision-making in the outpatient setting. Journal of General Internal Medicine, 12(6), 339–345. doi: 10.1046/j.1525-1497.1997.00057.x Brown, R. F., Butow, P. N., Butt, D. G., Moore, A. R., & Tattersall, M. H. N. (2004). Developing ethical strategies to assist oncologists in seeking informed consent to cancer clinical trials. Social Science & Medicine, 58(2), 379–390. doi:10.1016/s02779536(03)00204-1

References

131

Buetow, S. (2016). Person-centred health care: Balancing the welfare of clinicians and patients. London: Routledge. Bury, M. (1997). Health and illness in a changing society. London: Routledge. Butler, C. W., Potter, J., Danby, S., Emmison, M., & Hepburn, A. (2010). Advice implicative interrogatives: Building ‘client-centred support’ in a children’s helpline. Social Psychology Quarterly, 73(3), 265–287. Byrne, P. S., & Long, B. E. L. (1976). Doctors talking to patients: A study of the verbal behaviours of doctors in the consultation. London: Her Majesty’s Stationery Office. Charles, C., Gafni, A., & Whelan, T. (1997). Shared decision-making in the medical encounter: What does it mean? (or it takes at least two to tango). Social Science & Medicine, 44(5), 681–692. doi:10.1016/s0277-9536(96)00221-3 Charles, C., Gafni, A., & Whelan, T. (1999a). Decision-making in the physician-patient encounter: Revisiting the shared treatment decision-making model. Social Science & Medicine, 49(5), 651–661. doi:10.1016/s0277-9536(99)00145-8 Charles, C., Gafni, A., & Whelan, T. (1999b). What do we mean by partnership in making decisions about treatment? British Medical Journal, 319, 780–782. Clarke, A. (2017). A rejoinder to ‘concepts of health, ethics, and communication in shared decision making’ by Lauris Kaldjian. Communication and Medicine, 14(3), 274–276. doi:10.1558/cam.36372 Clegg, J., Murphy, E., Almack, K., & Harvey, A. (2008). Tensions around inclusion: Reframing the moral horizon. Journal of Applied Research in Intellectual Disabilities, 21(1), 81–94. doi:10.1111/j.1468-3148.2007.00371.x Coleman, T., Murphy, E., & Cheater, F. (2000). Factors influencing discussion of smoking between general practitioners and patients who smoke: A qualitative study. Family Practice, 50, 207–210. Collins, S., Drew, P., Watt, I., & Entwistle, V. (2005). ‘Unilateral’ and ‘bilateral’ practitioner approaches in decision-making about treatment. Social Science & Medicine, 61(12), 2611–2627. Craven, A., & Potter, J. (2010). Directives: Entitlement and contingency in action. Discourse Studies, 12, 419–442. doi:10.1177/1461445610370126 Curl, T. S., & Drew, P. (2008). Contingency and action: A comparison of two forms of requesting. Research on Language and Social Interaction, 41, 129–153. doi:10.1080/ 083518108028613 Davis, F. (1960). Uncertainty in medical prognosis, clinical and functional. American Journal of Sociology, 66, 41–47. de Haes, H. (2006). Dilemmas in patient centredness and shared decision making: A case for vulnerability. Patient Education and Counseling, 62(3), 291–298. de Kok, B. C., Widdicombe, S., Pilnick, A., & Laurier, E. (2018). Doing patient-centredness versus achieving public health targets: A critical review of interactional dilemmas in ART adherence support. Social Science & Medicine, 205, 17–25. doi:10.1016/j.socscimed.2018.03.030 Department of Health. (2013). More care, less pathway: A review of the Liverpool care pathway. Retrieved from https://www.gov.uk/government/publications/ review-of-liverpool-care-pathway-for-dying-patients Department of Health (UK). (2001). Valuing people: A new strategy for learning disability for the 21st century Cm 5086. London: HMSO. Department of Health (UK). (2009). Valuing people now: A new three-year strategy for people with learning disabilities. London: HMSO.

132

References

Department of Health and Social Care. (2021). NHS constitution for England. Retrieved from https://www.gov.uk/government/publications/the-nhs-constitutionfor-england Dieppe, P., & Horn, R. (2002). Soundbites and patient centred care. British Medical Journal, 325, 605. doi:10.1136/bmj.325.7364.605 Dingwall, R. (2002). Bioethics. In A. Pilnick (Eds.), Genetics and society. Buckingham: Open University Press. Dingwall, R., & Pilnick, A. (2020). Shared decision making: Doctors have expertise that patients want or need. British Medical Journal, 368, m128. doi:10.1136/ bmj.m128 Drew, P. (1991). Asymmetries of knowledge in conversational interactions. In I. Markov´a & K. Foppa (Eds.), Asymmetries in dialogue (pp. 29–48). Hemel Hempstead: Harvester Wheatsheaf. Drew, P., Chatwin, J., & Collins, S. (2001). Conversation analysis: A method for research into interactions between patients and health-care professionals. Health Expectations, 4(1), 58–70. doi:10.1046/j.1369-6513.2001.00125.x Drew, P., & Heritage, J. (1992). Analyzing talk at work: An introduction. In P. Drew & J. Heritage (Eds.), Talk at work: Interaction in institutional settings (Studies in interactional sociolinguistics, 8, pp. 3–65). Cambridge: Cambridge University Press. Dwamena, F., Holmes-Rovner, M., Gaulden, C. M., Jorgenson, S., Sadigh, G., Sikorskii, A., … Olomu, A. (2012). Interventions for providers to promote a patient-centred approach in clinical consultations. Cochrane Database of Systematic Reviews, 12, CD003267. doi:10.1002/14651858.CD003267.pub2 Edwards, D., & Potter, J. (2017). Some uses of subject-side assessments. Discourse Studies, 19(5), 497–514. doi:10.1177/1461445617715171 Edwards-Leeper, L., Leibowitz, S., & Sangganjanavanich, V. F. (2016). Affirmative practice with transgender and gender nonconforming youth: Expanding the model. Psychology of Sexual Orientation and Gender Diversity, 3(2), 165–172. doi:10.1037/ sgd0000167 Elwyn, G., Frosch, D., Thomson, R., Joseph-Williams, N., Lloyd, A., Kinnersley, P., … Barry, M. (2012). Shared decision making: A model for clinical practice. Journal of General Internal Medicine, 27(10), 1361–1367. doi:10.1007/s11606-012-2077-6 Elwyn, G., Hutchings, H., Edwards, A., Rapport, F., Wensing, M., Cheung, W. Y., & Grol, R. (2005). The option scale: Measuring the extent that clinicians involve patients in decision-making tasks. Health Expectations, 8(1), 34–42. doi:10.1111/ j.1369-7625.2004.00311.x Elwyn, G., Stiel, M., Durand, M. A., & Boivin, J. (2011). The design of patient decision support interventions: Addressing the theory-practice gap. Journal of Evaluation in Clinical Practice, 17(4), 565–574. doi:10.1111/j.1365-2753.2010.01517.x Faden, J., & Gorton, G. (2018). The doorknob phenomenon in clinical practice. American Family Physician, 98(1), 52–53. Farsides, B., Williams, C., & Alderson, P. (2004). Aiming towards “moral equilibrium”: Health care professionals’ views on working within the morally contested field of antenatal screening. Journal of Medical Ethics, 30(5), 505–509. doi:10.1136/jme.2002.001438 Finlay, W., Antaki, C., & Walton, C. (2008). Saying no to the staff: An analysis of refusals in a home for people with severe communication difficulties. Sociology of Health & Illness, 30(1), 55–75.

References

133

Fisher, S. (1995). Nursing wounds: Nurse practitioners, doctors, women patients and the negotiation of meaning. New Brunswick, NJ: Rutgers University Press. Flyvbjerg, B. (2001). Making social science matter: Why social inquiry fails and how it can succeed again. Cambridge: Cambridge University Press. Ford, J., Hepburn, A., & Parry, R. (2019). What do displays of empathy do in palliative care consultations? Discourse Studies, 21(1), 22–37. Fotaki, M. (2014). Can consumer choice replace trust in the national health service in England? Towards developing an affective psychosocial conception of trust in health care. Sociology of Health & Illness, 36(8), 1276–1294. Fox, R. C. (1957). Training for uncertainty. In R. K. Merton, G. G. Reader, & P. L. Kendall (Eds.), The student physician: Introductory studies in the sociology of medical education (pp. 207–241). Cambridge, MA: Harvard University Press. Fox, R. C. (2000). Medical uncertainty revisited. In G. L. Albrecht, R. Fitzpatrick, & S. C. Scrimshaw (Eds.), The handbook of social studies in health and medicine (pp. 409–425). London: Sage. Fox, N., & Ward, K. (2006). Health identities: From expert patient to resisting consumer. Health, 10(4), 461–479. doi:10.1177/1363459306067314 Frankel, R. M. (1983). The laying on of hands: Aspects of the organisation of gaze, touch, and talk in a medical encounter. In S. Fisher & A. Dundas Todd (Eds.), The social organisation of doctor-patient communication. Washington, DC: Center for Applied Linguistics. Frankel, R. M., & Beckman, H. B. (1982). Impact: An interaction-based method for preserving and analyzing clinical transactions. In L. Pettegrew (Ed.), Explorations in provider and patient interaction. Louisville, KY: Humana Press. Frank, J. R., Snell, L. S., Cate, O. T., Holmboe, E. S., Carraccio, C., Swing, S. R., … Harris, K. A. (2010). Competency based medical education: Theory to practice. Medical Teacher, 32(8), 638–645. Freidson, E. (1970). Profession of medicine. New York, NY: Dodd Mead. Freidson, E. (1975). Doctoring together: A study of professional social control. New York, NY: Elsevier. Garfinkel, H. (1967). Studies in ethnomethodology. Englewood Cliffs, NJ: PrenticeHall. Garfinkel, H., & Bittner, E. (1967). Good organizational reasons for ’bad’ clinic records. In H. Garfinkel (Ed.), Studies in ethnomethodology (pp. 186–207). Englewood Cliffs, NJ: Prentice-Hall. Gill, V. T. (1998). Doing attributions in medical interaction: Patients’ explanations for illness and doctors’ responses. Social Psychology Quarterly, 61, 342–360. Gill, V. T. (2005). Patient “demand” for medical interventions: Exerting pressure for an offer in a primary care clinic visit. Research on Language and Social Interaction, 38(4), 451–479. Gill, V. T., & Maynard, D. W. (1995). On “labeling” in actual interaction: Delivering and receiving diagnoses of developmental disabilities. Social Problems, 42(1), 11–37. Gill, V. T., & Maynard, D. W. (2006). Explaining illness: Patients’ proposals and physicians’ responses. In J. Heritage & D. Maynard (Eds.), Communication in medical care: Interaction between primary care physicians and patients. Cambridge: Cambridge University Press.

134

References

Gill, V. T., Pomerantz, A., & Denvir, P. (2010). Pre-emptive resistance: Patients’ participation in diagnostic sense-making activities. Sociology of Health & Illness, 32(1), 1–20. Gill, V. T., & Roberts, F. (2013). Conversation analysis in medicine. In J. Sidnell & T. Stivers T (Eds.), The handbook of conversation analysis (pp. 575-592). Oxford: Wiley-Blackwell. Goffman, E. (1955). On face-work: An analysis of ritual elements in social interaction. Psychiatry, 18(3), 213–231. Goffman, E. (1967). On face-work. In Interaction ritual: Essays on face-to-face interaction (pp. 5–45). Chicago, IL: Aldine. Goffman, E. (1983). The interaction order: American sociological association, 1982 presidential address. American Sociological Review, 48(1), 1–17. Goldfarb, M. J., Bibas, L., Bartlett, V., Jones, H., & Khan, N. (2017). Outcomes of patient-and family-centered care interventions in the ICU: A systematic review and meta-analysis. Critical Care Medicine, 45(10), 1751–1761. doi:10.1097/ CCM.0000000000002624 Goodwin, C. (1994). Professional vision. American Anthropologist, 96(3), 606–633. Graham, H., & Der, G. (1999). Patterns and predictors of smoking cessation among British women. Health Promotion International, 14(3), 231–240. Greatbatch, D., Hanlon, G., Goode, J., O’Caithain, A., Strangleman, T., & Luff, D. (2005). Telephone triage, expert systems and clinical expertise. Sociology of Health & Illness, 27(6), 802–830. Griffiths, A., Knight, A., Harwood, R. H., & Gladman, J. R. F. (2014). Preparation to care for confused older patients in general hospitals: A study of UK health professionals. Age and Ageing, 43, 521–527. Guadagnoli, E., & Ward, P. (1998). Patient participation in decision-making. Social Science & Medicine, 47(3), 329–339. Gulland, A. (2011). Welcome to the century of the patient. British Medical Journal, 342. doi:10.1136/bmj.d2057 Habermas, J. (1981). Theory of communicative action. Vol. 2. Lifeworld and system: A critique of functionalist reason. Paris: Fayard. ¨ H˚akansson Eklund, J., Holmstrom, I. K., Kumlin, T., Kaminsky, E., Skoglund, K., ¨ Hoglander, J, … Summer Meranius, M. (2019). “Same same or different?” A review of reviews of person-centered and patient-centered care. Patient Education and Counselling, 102(1), 3–11. doi:10.1016/j.pec.2018.08.029 Halkowski, T., & Gill, V. T. (2010). Conversation analysis and ethnomethodology: The centrality of interaction. In I. Bourgeault, R. Dingwall, & R. De Vries (Eds.), The Sage handbook of qualitative methods in health research (pp. 212–228). London: Sage. Hall, S. (2011). The neo-liberal revolution. Cultural Studies, 25(6), 705–728. doi: 10.1080/09502386.2011.619886 Harvey, D. (2005). A brief history of neoliberalism. Oxford: Oxford University Press. Harwood, R. H., O’Brien, R., Goldberg, S. E., Allwood, R., Pilnick, A., Beeke, S., … Schneider, J. (2018). A staff training intervention to improve communication between people living with dementia and health-care professionals in hospital: The voice mixed-methods development and evaluation study. Health Services and Delivery Research, 6(41). doi:10.3310/hsdr06410

References

135

Hayes, J., McCabe, R., Ford, T., & Russell, G. (2020). Drawing a line in the sand: Affect and testimony in autism assessment teams in the UK. Sociology of Health & Illness, 42(4), 825–843. Health Foundation. (2013). A portal for resources and learning on person-centred care. Retrieved from https://www.health.org.uk/newsletter-feature/a-portal-forresources-and-learning-on-person-centred-care Health Foundation. (2014). Helping measure person-centred care. Retrieved from https://www.health.org.uk/publications/helping-measure-person-centred-care Heath, C. (1981). The opening sequence in doctor-patient interaction. In P. Atkinson & C. Heath (Eds.), Medical work: Realities and routines. Farnborough: Gower. Heath, C. (1986). Body movement and speech in medical interaction. Cambridge: Cambridge University Press. Heath, C. (1992). Diagnosis and assessment in the medical consultation. In P. Drew & J. Heritage (Eds.), Talk at work: Interaction in institutional settings (pp. 235–267). Cambridge: Cambridge University Press. Heinemann, T. (2006). ‘Will you or can’t you?’ Displaying entitlement in interrogative requests. Journal of Pragmatics, 38(7), 1081–1104. Helyar, S., Jackson, L., Patrick, L., & Hill, A. (2022). Gender dysphoria in children and young people: The implications for clinical staff of the Bell vs Tavistock Judicial Review and appeal ruling. Journal of Clinical Nursing. doi:10.1111/ jocn.16164 Henwood, F., Harris, R., & Spoel, P. (2011). Informing health? Negotiating the logics of choice and care in everyday practices of ‘healthy living’. Social Science & Medicine, 72(12), 2026–2032. doi:10.1016/j.socscimed.2011.04.007 Henwood, F., Wyatt, S., Hart, A., & Smith, J. (2003). ‘Ignorance is bliss sometimes’: Constraints on the emergence of the ‘informed patient’ in the changing landscapes of health information. Sociology of Health & Illness, 25(6), 589–607. Hepburn, A., Wilkinson, S., & Butler, C. W. (2014). Intervening with conversation analysis in telephone helpline services: Strategies to improve effectiveness. Research on Language and Social Interaction, 47(3), 239–254. Heritage, J. (1984). Garfinkel and ethnomethodology. Cambridge: Polity Press. Heritage, J. (2009). Questioning in medicine. In A. F. Freed & S. Ehrlich (Eds.), “Why do you ask?”: The function of questions in institutional discourse (pp. 42–68). New York: Oxford University Press. Heritage, J. (2010). Questioning in medicine. In A. F. Freed & S. Ehrlich (Eds.), Why do you ask? The function of questions in institutional discourse (pp. 42–68). New York, NY: Oxford University Press. Heritage, J. (2011). The interaction order and clinical practice: Some observations on dysfunctions and action steps. Patient Education and Counseling, 84, 338–343. Heritage, J. (2012). Epistemics in action: Action formation and territories of knowledge. Research on Language & Social Interaction, 45(1), 1–29. Heritage, J. (2013). Action formation and its epistemic (and other) backgrounds. Discourse Studies, 15(5), 551–578. ¨ Heritage, J., & Lindstrom, A. (1998). Motherhood, medicine, and morality: Scenes from a medical encounter. Research on Language and Social Interaction, 31(3–4), 397–438.

136

References

Heritage, J., & Maynard, D. W. (2006a). Introduction: Analyzing interaction between doctors and patients in primary care encounters. In J. Heritage & D. Maynard (Eds.), Communication in medical care: Interaction between primary care physicians and patients. Cambridge: Cambridge University Press. Heritage, J., & Maynard, D. W. (2006b). Problems and prospects in the study of physician-patient interaction: 30 years of research. Annual Review of Sociology, 32, 351–374. Heritage, J., & Raymond, G. (2005). The terms of agreement: Indexing epistemic authority and subordination in talk-in-interaction. Social Psychology Quarterly, 68(1), 15–38. Heritage, J., & Robinson, J. (2011). ‘Some’ versus ‘any’ medical issues: Encouraging patients to reveal their unmet concerns. In C. Antaki (Ed.), Applied conversation analysis (pp. 15–31). London: Palgrave Macmillan. Heritage, J., Robinson, J. D., Elliott, M. N., Beckett, M., & Wilkes, M. (2007). Reducing patients’ unmet concerns in primary care: The difference one word can make. Journal of General Internal Medicine, 22(10), 1429–1433. Heritage, J., & Sefi, S. (1992). Dilemmas of advice: Aspects of the delivery and reception of advice in interactions between health visitors and first time mothers. In P. Drew & J. Heritage (Eds.), Talk at work. Cambridge: Cambridge University Press. Heritage, J., & Stivers, T. (1999). Online commentary in acute medical visits: A method of shaping patient expectations. Social Science & Medicine, 49(11), 1501–1517. Hindmarsh, J., & Pilnick, A. (2002). The tacit order of teamwork: Collaboration and embodied conduct in anaesthesia. The Sociological Quarterly, 43(2), 139–164. Hollin, G., & Pilnick, A. (2018). The categorisation of resistance: Interpreting failure to follow a proposed line of action in the diagnosis of autism amongst young adults. Sociology of Health & Illness, 40(7), 1215–1232. Holt, E. (1993). The structure of death announcements: Looking on the bright side of death. Text-Interdisciplinary Journal for the Study of Discourse, 13(2), 189–212. Honneth, A. (2014). Freedom’s right. New York, NY: Columbia University Press. Hughes, E. C. (1951). Mistakes at work. Canadian Journal of Economics and Political Science, 17(3), 320–327. Innes, M., Skelton, J., & Greenfield, S. (2006). A profile of communication in primary care physician telephone consultations: Application of the roter interaction analysis system. British Journal of General Practice, 56(526), 363–368. Ja´en, C. R., Ferrer, R. L., Miller, W. L., Palmer, R. F., Wood, R., Davila, M, … Stange, K. C. (2010). Patient outcomes at 26 months in the patient-centered medical home National Demonstration Project. The Annals of Family Medicine, 8(Suppl 1), S57–S67. doi:10.1370/afm.1121 Jefferson, G. (1984). On stepwise transition from talk about a trouble to inappropriately next-positioned matters. In J. M. Atkinson & J. Heritage (Eds.), Structures of social action (pp. 191–221). Cambridge: Cambridge University Press. Jefferson, G. (1988). On the sequential organization of troubles-talk in ordinary conversation. Social Problems, 35, 418–441. Jefferson, G. (2004). Glossary of transcript symbols with an introduction. In G. H. Lerner (Ed.), Conversation analysis: Studies from the first generation (pp. 13–23). Philadelphia, PA: John Benjamins.

References

137

Jefferson, G., & Lee, J. R. E. (1981). The rejection of advice: Managing the problematic convergence of a ‘troubles-telling’ and a ‘service encounter’. Journal of Pragmatics, 5(5), 399–422. Jefferson, G., & Lee, J. R. E. (1992). The rejection of advice: Managing the problematic convergence of a “troubles-telling” and a “service encounter”. In P. Drew & J. Heritage (Eds.), Talk at work: Interaction in institutional settings (pp. 521–548). Cambridge: Cambridge University Press. Jenkins, L. (2015). Negotiating pain: The joint construction of a child’s bodily sensation. Sociology of Health & Illness, 37(2), 298–311. Kaldjian, L. C. (2013). Communicating moral reasoning in medicine as an expression of respect for patients and integrity among professionals. Communication and Medicine, 10(2), 177–183. Kaldjian, L. C. (2017). Concepts of health, ethics, and communication in shared decision making. Communication and Medicine, 14(1), 83–95. Kant, I. (1785/1996). Groundwork of the metaphysics of morals. Cambridge: Cambridge University Press. ¨ Kasper, J., Hoffmann, F., Heesen, C., Kopke, S., & Geiger, F. (2012). MAPPIN’SDM–the multifocal approach to sharing in shared decision making. PLoS One, 7(4), e34849. Keevallik, L. (2011). The terms of not knowing. In T. Stivers, L. Mondada, & J. Steensig (Eds.), The morality of knowledge in conversation (pp. 184–206). Cambridge: Cambridge University Press. Keller, J. (1997). Autonomy, relationality, and feminist ethics. Hypatia, 12(2), 152–164. Kerr, A. (2004). Genetics and society: A sociology of disease. London: Routledge. Kessler, S. (1992). Psychological aspects of genetic counseling. VII. Thoughts on directiveness. Journal of Genetic Counseling, 1, 9–17. Kessler, S. (1997). Psychological aspects of genetic counseling. XI. Nondirectiveness revisited. American Journal of Medical Genetics, 72, 164–171. Kevles, D. (1995). In the name of eugenics: Genetics and the uses of human heredity. Cambridge, MA: Harvard University Press. Kevoe-Feldman, H., & Pomerantz, A. (2018). Critical timing of actions for transferring 911 calls in a wireless call center. Discourse Studies, 20(4), 488–505. King’s Fund. (2010). How to deliver high quality cost effective patient care how to deliver high-quality, patient-centred, cost-effective care. Retrieved from https:// www.kingsfund.org.uk/publications/articles/how-deliver-high-quality-patientcentred-cost-effective-care Kinmonth, A. L., Woodcock, A., Griffin, S., Spiegal, N., & Campbell, M. (1998). Randomised controlled trial of patient-centred care of diabetes in general practice: Impact on current well-being and future disease risk. British Medical Journal, 317, 1202–1208. Kitwood, T. (1997). Dementia reconsidered: The person comes first. Buckingham: Open University Press. Korsch, B. M., Gozzi, E. K., & Francis, V. (1968). Gaps in doctor-patient communication. Pediatrics, 42, 855–871. Korsch, B., & Negrete, V. F. (1972). Doctor-patient communication. Scientific American, 227, 66–74. Kurtz, S., Silverman, J., & Draper, J. (2004). Teaching and learning communication skills in medicine. Oxford: Radcliffe Medical Press.

138

References

Landmark, A. M. D., Gulbrandsen, P., & Svennevig, J. (2015). Whose decision? Negotiating epistemic and deontic rights in medical treatment decisions. Journal of Pragmatics, 78, 54–69. Landmark, A. M. D., Ofstad, E. H., & Svennevig, J. (2017). Eliciting patient preferences in shared decision-making (SDM): Comparing conversation analysis and SDM measurements. Patient Education and Counseling, 100(11), 2081–2087. Le Grand, J. (2007). The other invisible hand: Delivering public service through competition and choice. Princeton, NJ: Princeton University Press. Lee, Y. Y., & Lin, J. L. (2010). Do patient autonomy preferences matter? Linking patient-centered care to patient–physician relationships and health outcomes. Social Science & Medicine, 71(10), 1811–1818. Lehtinen, E. (2005). Information, understanding and the benign order of everyday life in genetic counselling. Sociology of Health & Illness, 27(5), 575–601. Levenstein, J. H., McCracken, E. C., McWhinney, I. R., Stewart, M. A., & Brown, J. B. (1986). The patient-centred clinical method. 1. A model for the doctor-patient interaction in family medicine. Family Practice, 3(1), 24–30. Lewin, S. A., Skea, Z. C., Entwistle, V., Zwarenstein, M., & Dick, J. (2001). Interventions for providers to promote a patient-centred approach in clinical consultations. Cochrane Database of Systematic Reviews, 4, CD003267. Leydon, G. M., Ekberg, K., & Drew, P. (2013). “How can I help?” Nurse call openings on a cancer helpline and implications for call progressivity. Patient Education and Counseling, 92(1), 23–30. Lindholm, C. (2015). Parallel realities: The interactional management of confabulation in dementia care encounters. Research on Language and Social Interaction, 48(2), 176–199. ¨ Lindstrom, A. (2005). Language as social action. A study of how senior citizens request assistance with practical tasks in the Swedish home help service. In A. Hakulinen & M. Selting (Eds.), Syntax and lexis in conversation: Studies on the use of linguistic resources in talk-in interaction (pp. 209–230). Amsterdam: John Benjamins. Little, P., Everitt, H., Williamson, I., Warner, G., Moore, M., Gould, C., … Payne, S. (2001). Preferences of patients for patient centred approach to consultation in primary care: Observational study. British Medical Journal, 322(7284), 468. Lukes, S. (1978). Power and authority. In T. Bottomore & R. Nisbet (Eds.), A history of sociological analysis (pp. 633–676). London: Heinemann. Macintyre, S., & Oldman, D. (1977). Coping with migraine. In A. Davis & G. Horobin (Eds.), Medical encounters: The experience of illness and treatment. London: Croom Helm. Makoul, G., & Clayman, M. L. (2006). An integrative model of shared decision making in medical encounters. Patient Education and Counseling, 60(3), 301–312. Matthias, M., Salyers, M., & Frankel, R. (2013). Re-thinking shared decision-making: Context matters. Patient Education and Counseling, 91(2), 176–179. Maynard, D. W. (1991). Interaction and asymmetry in clinical discourse. American Journal of Sociology, 97(2), 448–495. Maynard, D. W. (2003). Bad news, good news: Conversational order in everyday talk and clinical settings. Chicago, IL: University of Chicago Press. Maynard, D. (2006). ‘Does it mean i’m gonna die?’: On meaning assessment in the delivery of diagnostic news. Social Science and Medicine, 62, 1902–1916.

References

139

Maynard, D. W., & Frankel, R. M. (2006). On diagnostic rationality: Bad news, good news, and the symptom residue. In J. Heritage & D. Maynard (Eds.), Communication in medical care: Interaction between primary care physicians and patients. Cambridge: Cambridge University Press. Maynard, D. W., & Heritage, J. (2005). Conversation analysis, doctor-patient interaction and medical communication. Medical Education, 39, 428–435. Mazeland, H., & ten Have, P. (1996). Essential tensions in (semi-)open research interviews. In I. Maso & F. Wester (Eds.), The deliberate dialogue: Qualitative perspectives on the interview (pp. 87–113). Brussels: VUB University Press. Mazumdar, P. (1992). Eugenics, human genetics and human failings: The eugenics society, its sources and its critics in Britain. London: Routledge. McCabe, R. (2021). When patients and clinician (dis)agree about the nature of the problem: The role of displays of shared understanding in acceptance of treatment. Social Science & Medicine, 290, 114208. McCabe, R., John, P., Dooley, J., Healey, P., Cushing, A., Kingdon, D., … Priebe, S. (2016). Training to enhance psychiatrist communication with patients with psychosis (TEMPO): Cluster randomised controlled trial. The British Journal of Psychiatry, 209(6), 517–524. McWhinney, I. (1989). The need for a transformed clinical method. In M. Stewart & D. Roter (Eds.), Communicating with medical patients. London: Sage. Mead, N., & Bower, P. (2000). Patient-centredness: A conceptual framework and review of the empirical literature. Social Science & Medicine, 51(7), 1087–1110. Mechanic, D. (1962). The concept of illness behaviour. Journal of Chronic Diseases, 15(2), 189–194. Meehan, A. J. (1981). Some conversational features of the use of medical terms by doctors and patients. In P. Atkinson & C. Heath (Eds.), Medical work: Realities and routines. Westmead: Gower Ltd. Michie, S., Smith, J. A., Senior, V., & Marteau, T. M. (2003). Understanding why negative genetic tests sometimes fail to reassure. American Journal of Medical Genetics, Part A, 119A(3), 340–347. Mid Staffordshire NHS Foundation Trust Public Inquiry. (2013). Report of the mid Staffordshire NHS foundation trust public inquiry: Executive summary. Retrieved from http://www.midstaffspublicinquiry.com/sites/default/files/report/Executive% 20summary.pdf Mishler, E. G. (1984). The discourse of medicine: Dialectics of medical interviews. Norwood, MA: Ablex. Mol, A. (2008). The logic of care: Health and the problem of patient choice. London: Routledge. Mondada, L. (2011). The organization of concurrent courses of action in surgical demonstrations. In J. Streeck, C. Goodwin, & C. LeBaron (Eds.), Learning in doing (pp. 207–226). New York, NY: Cambridge University Press (Embodied Interaction. Language and Body in the Material World). Mondada, L. (2014). Instructions in the operating room: How surgeons direct their assistant’s hands. Discourse Studies, 16(2), 131–161. doi:10.1177/ 1461445613515325 Montori, V. M., Gafni, A., & Charles, C. (2006). A shared treatment decision-making approach between patients with chronic conditions and their clinicians: The case of diabetes. Health Expectations, 9(1), 25–36. doi:10.1111/j.1369-7625.2006.00359.x

140

References

National Institute for Health and Care Excellence. (2006). Dementia: Supporting people with dementia and their careers in health and social care. Clinical guideline [CG42]. Retrieved from https://www.nice.org.uk/Guidance/CG42 National Institute for Health and Care Excellence. (2021). Patient experience in adult NHS services: Improving the experience of care for people using adult NHS services. Clinical guideline [CG138]. Retrieved from https://www.nice.org.uk/ Guidance/CG138 Nguyen, V. K. (2005). Antiretroviral globalism, biopolitics and therapeutic citizenship. In A. Ong & S. J. Collier (Eds.), Technology, politics and ethics as anthropological problems (pp. 124–144). Oxford: Blackwell. NHS Digital. (2021). Data on written complaints in the NHS data on written complaints in the NHS-2020–21 quarter 1 and quarter 2. Retrieved from https:// digital.nhs.uk/data-and-information/publications/statistical/data-on-writtencomplaints-in-the-nhs/2020-21-quarter-1-and-quarter-2 NHS England. (2022). The cass review: Independent review of gender identity services for children and young people. Retrieved from https://cass.independent-review.uk/ publications/interim-report Nishisaka, A. (2007). Hand touching hand: Referential practice at a Japanese midwife house. Human Studies, 30(3), 199–217. Oakley, A. (1980). Women confined: Toward sociology of childbirth. Oxford: Martin Robertson. O’Brien, R. (2019). The development of a communication skills training intervention for healthcare professionals caring for people with dementia in the acute hospital setting: An applied conversation analytic study. Doctoral dissertation, University of Nottingham. O’Brien, R., Beeke, S., Pilnick, A., Goldberg, S. E., & Harwood, R. H. (2020). When people living with dementia say ‘no’: Negotiating refusal in the acute hospital setting. Social Science & Medicine, 263, 113188. O’Brien, R., Goldberg, S. E., Pilnick, A., Beeke, S., Schneider, J., Sartain, K., … Harwood, R. H. (2018). The voice study-a before and after study of a dementia communication skills training course. PLoS One, 13, e0198567. doi:10.1371/ journal.pone.0198567 O’Doherty, K., & Suthers, G. K. (2007). Risky communication: Pitfalls in counseling about risk, and how to avoid them. Journal of Genetic Counseling, 16(4), 409–417. O’Hara, G. (2013). The complexities of ‘consumerism’: Choice, collectivism and participation within Britain’s national health service, c.1961–c.1979. Social History of Medicine, 26(2), 288–304. doi:10.1093/shm/hks062 Parry, R. (2009). Practitioners’ accounts for treatment actions and recommendations in physiotherapy: When do they occur, how are they structured, what do they do? Sociology of Health & Illness, 31(6), 835–853. Parry, N. C. A., & Parry, J. (1976). The rise of the medical profession: A study of collective social mobility. London: Croom Helm. Parsons, T. (1951). The social system. New York, NY: Free Press. Peek, M. E., Odoms-Young, A., Quinn, M. T., Gorawara-Bhat, R., Wilson, S. C., & Chin, M. H. (2010). Race and shared decision-making: Perspectives of African-Americans with diabetes. Social Science & Medicine, 71(1), 1–9. Pendleton, D. (1983). Doctor-patient communication: A review. In D. Pendleton & J. Hasler (Eds.), Doctor-patient communication (pp. 5–56). New York, NY: Academic Press.

References

141

Penn, C., Watermeyer, J., & Evans, M. (2011). Why don’t patients take their drugs? The role of communication, context and culture in patient adherence and the work of the pharmacist in HIV/AIDS. Patient Education and Counseling, 83(3), 310–318. Per¨akyl¨a, A. (1991). Hope work in the care of seriously ill patients. Qualitative Health Research, 1(4), 407–433. doi:10.1177/104973239100100402 Per¨akyl¨a, A. (1995). AIDS counselling: Institutional interaction and clinical practice. Cambridge: Cambridge University Press. Per¨akyl¨a, A. (1998). Authority and accountability: The delivery of diagnosis in primary healthcare. Social Psychology Quarterly, 61, 301–320. Per¨akyl¨a, A. (2006). Communicating and responding to diagnosis. In J. Heritage & D. Maynard (Eds.), Communication in medical care: Interactions between primary care physicians and patients (pp. 214–247). Cambridge: Cambridge University Press. Per¨akyl¨a, A. (2019). Conversation analysis and psychotherapy: Identifying transformative sequences. Research on Language and Social Interaction, 52(3), 257–280. doi:10.1080/08351813.2019.1631044 Per¨akyl¨a, A., Antaki, C., Vehvil¨ainen, S., & Leudar, I. (Eds.). (2008). Conversation analysis and psychotherapy. Cambridge: Cambridge University Press. Per¨akyl¨a, A., Ruusuvuori, J., & Lindfors, P. (2007). What is patient participation: Reflections arising from the study of general practice, homoeopathy and psychoanalysis. In S. Collins, N. Britton, J. Ruusuvuori, & A. Thompson (Eds.), Patient participation in health care consultations: Qualitative perspectives. Maidenhead: Open University Press. Pharmaceutical Society of Northern Ireland. (2016). Code of ethics and standards code of ethics and standards. Retrieved from https://www.psni.org.uk/psni/about/ code-of-ethics-and-standards/#:~:text5Principle%201%3A%20Always%20put% 20the%20patient%20first%20Principle,Act%20with%20professionalism%20and% 20integrity%20at%20all%20times Pilnick, A. (1998). ‘Why didn’t you just say that?’ Dealing with issues of asymmetry, knowledge and competence in the pharmacist⁄client encounter. Sociology of Health & Illness, 20(1), 29–51. Pilnick, A. (1999). “Patient counseling” by pharmacists: Advice, information, or instruction? The Sociological Quarterly, 40(4), 613–622. Pilnick, A. (2001). The interactional organization of pharmacist consultations in a hospital setting: A putative structure. Journal of Pragmatics, 33(12), 1927–1945. Pilnick, A. (2002a). What ‘most people’ do: Exploring the ethical implications of genetic counselling. New Genetics & Society, 21(3), 339–350. Pilnick, A. (2002b). ‘There are no rights and wrongs in these situations’: Identifying interactional difficulties in genetic counselling. Sociology of Health & Illness, 24(1), 66–88. Pilnick, A. (2002c). Genetics and society: An introduction. Buckingham: Open University Press. Pilnick, A. (2003). “Patient counselling” by pharmacists: Four approaches to the delivery of counselling sequences and their interactional reception. Social Science & Medicine, 56(4), 835–849. doi:10.1016/s0277-9536(02)00082-5 Pilnick, A. (2004). ‘It’s just one of the best tests that we’ve got at the moment’: The presentation of screening for fetal abnormality in pregnancy. Discourse & Society, 15(4), 451–465. Pilnick, A. (2008). ‘It’s something for you both to think about’: Choice and decision making in nuchal translucency screening for down’s syndrome. Sociology of Health & Illness, 30(4), 511–530. doi:10.1111/j.1467-9566.2007.01071.x

142

References

Pilnick, A. (2013). Sociology without frontiers? On the pleasures and perils of interdisciplinary research. Sociological Research Online, 18(3), 97–104. doi: 10.5153/sro.3108 Pilnick, A. (2022). Medicine and healthcare. In A. Carlin, A. Dennis, K. Jenkins, O. Lindwall, M. Mair (Eds.), Routledge international handbook of ethnomethodology and conversation analysis. London: Routledge. Pilnick, A., Clegg, J., Murphy, E., & Almack, K. (2010, March). Questioning the answer: Questioning style, choice and self-determination in interactions with young people with intellectual disabilities. Sociology of Health & Illness, 32(3), 415–436. doi:10.1111/j.1467-9566.2009.01223.x Pilnick, A., Clegg, J., Murphy, E., & Almack, K. (2011). ‘Just being selfish for my own sake…’: Balancing the views of young adults with intellectual disabilities and their careers in transition planning. The Sociological Review, 59(2), 303–323. Pilnick, A., & Coleman, T. (2003). “I’ll give up smoking when you get me better”: Patients’ resistance to attempts to problematise smoking in general practice (GP) consultations. Social Science & Medicine, 57(1), 135–145. doi:10.1016/s02779536(02)00336-2 Pilnick, A., & Coleman, T. (2006). Death, depression and ‘defensive expansion’: Closing down smoking as an issue for discussion in GP consultations. Social Science & Medicine, 62(10), 2500–2512. doi:10.1016/j.socscimed.2005.10.031 Pilnick, A., & Coleman, T. (2010). ‘Do your best for me’: The difficulties of finding a clinically effective endpoint in smoking cessation consultations in primary care. Health, 14(1), 57–74. doi:10.1177/1363459309347489 Pilnick, A., & Dingwall, R. (2001). On the remarkable persistence of asymmetry in doctor/patient interaction: A critical review. Social Science & Medicine, 72(8), 1374–1382. doi:10.1016/j.socscimed.2011.02.033 Pilnick, A., Fraser, D. M., & James, D. (2004). Presenting and discussing nuchal translucency screening for fetal abnormality in the UK. Midwifery, 20(1), 82–93. Pilnick, A., Hindmarsh, J., & Gill, V. T. (2009). Beyond ‘doctor and patient’: Developments in the study of healthcare interactions. Sociology of Health & Illness, 31(6), 787–802. Pilnick, A., Hindmarsh, J., & Gill, V. T. (2010). Communication in healthcare settings: Policy, participation and new technologies. Oxford: Blackwell. Pilnick, A., O’Brien, R., Beeke, S., Goldberg, S., & Harwood, R. (2021). Avoiding repair, maintaining face: Responding to hard-to-interpret talk from people living with dementia in the acute hospital. Social Science & Medicine, 282, 114156. doi: 10.1016/j.socscimed.2021.114156 Pilnick, A., & Zayts, O. (2011). ‘Let’s have it tested first’: Choice and circumstances in decision-making following positive antenatal screening in Hong Kong. Sociology of Health & Illness, 34(2), 266–282. doi:10.1111/j.1467-9566.2011.01425.x Pilnick, A., & Zayts, O. (2014). “It’s just a likelihood”: Uncertainty as topic and resource in conveying “positive” results in an antenatal screening clinic. Symbolic Interaction, 37(2), 187–208. doi:10.1002/symb.99 Pilnick, A., & Zayts, O. (2016). Advice, authority and autonomy in shared decision-making in antenatal screening: The importance of context. Sociology of Health & Illness, 38(3), 343–359. doi:10.1111/1467-9566.12346 Pilnick, A., & Zayts, O. (2019, July). The power of suggestion: Examining the impact of presence or absence of shared first language in the antenatal clinic. Sociology of Health & Illness, 41(6), 1120–1137. doi:10.1111/1467-9566.12888

References

143

Pino, M., & Parry, R. (2019). How and when do patients request life-expectancy estimates? Evidence from hospice medical consultations and insights for practice. Patient Education and Counseling, 102(2), 223–237. Pollner, M. (1987). Mundane reason. Cambridge: Cambridge University Press. Pomerantz, A. (1980). Telling my side: “Limited access” as a “fishing” device. Sociological Inquiry, 50(3), 186–198. Pomerantz, A. (1984). Agreeing and disagreeing with assessments: Some features of preferred/dispreferred turn shapes. In J. M. Atkinson & J. Heritage (Eds.), Structures of social action: Studies in conversation analysis (pp. 57–101). Cambridge: Cambridge University Press. Pomerantz, A. (1988). Offering a candidate answer: An information seeking strategy. Communication Monographs, 55(4), 360–373. doi:10.1080/03637758809376177 Pomerantz, A., & Heritage, J. (2013). Preference. In J. Sidnell & T. Stivers (Eds.), The handbook of conversation analysis (pp. 210–228). Oxford: Wiley-Blackwell. Press, N., & Browner, C. H. (1997). Why women say yes to prenatal diagnosis. Social Science & Medicine, 45(7), 979–989. Prior, L. (2003). Belief, knowledge and expertise: The emergence of the lay expert in medical sociology. Sociology of Health & Illness, 25(3), 41–57. Rapley, T. (2008). Distributed decision making: The anatomy of decisions-in-action. Sociology of Health & Illness, 30(3), 429–444. doi:10.1111/j.14679566.2007.01064.x Rawls, A. W., & Turowetz, J. (2021). Garfinkel’s politics: Collaborating with Parsons to document taken-for-granted practices for assembling cultural objects and their grounding in implicit social contract. The American Sociologist, 52, 131–158. Reader, T. W., Gillespie, A., Roberts, J. (2014). Patient complaints in healthcare systems: A systematic review and coding taxonomy. BMJ Quality and Safety, 23, 678–689. Roberts, F. D. (1999). Talking about treatment: Recommendations for breast cancer adjuvant therapy. Oxford: Oxford University Press on Demand. Robinson, J. D. (2001). Closing medical encounters: Two physician practices and their implications for the expression of patients’ unstated concerns. Social Science & Medicine, 53(5), 639–656. Robinson, J. D., & Heritage, J. (2005). The structure of patients’ presenting concerns: The completion relevance of current symptoms. Social Science & Medicine, 61(2), 481–493. Robinson, J. D., & Heritage, J. (2014). Intervening with conversation analysis: The case of medicine. Research on Language and Social Interaction, 47(3), 201–218. Robinson, J. D., Tate, A., & Heritage, J. (2016). Agenda-setting revisited: When and how do primary-care physicians solicit patients’ additional concerns? Patient Education and Counseling, 99(5), 718–723. Rogers, C. (1951). Client-centered therapy: Its current practice, implications and theory. London: Constable. Rogers, C. R. (1961). On becoming a person: A psychotherapists view of psychotherapy. Boston, MA: Houghton Mifflin. Rose, N. (1998). Inventing our selves: Psychology, power, and personhood. Cambridge: Cambridge University Press. Rose, N. (1999). Governing the soul: The shaping of the private self. London: Free Association Press.

144

References

Roter, D. (1977). Patient participation in the patient-provider interaction: The effects of patient question asking on the quality of interaction, satisfaction and compliance. Health Education Monographs, 5, 281–315. Roter, D., & Larson, S. (2002). The roter interaction analysis system (RIAS): Utility and flexibility for analysis of medical interactions. Patient Education and Counseling, 42, 243–251. Royal College of Speech and Language Therapists. (2014). Speech and language therapy provision for people with dementia. Retrieved from http://www.rcslt.org/ members/publications/publications2/dementia_position_paper2014 Royal Pharmaceutical Society of Great Britain. (2021). Medicines, ethics and practice (44th edition). London: RPSGB. Sacks, H. (1984). Notes on methodology. In J. M. Maxwell Atkinson & J. Heritage (Eds.), Structures of social action: Studies in conversation analysis. Cambridge: Cambridge University Press. Sacks, H. (1987). On the preferences for agreement and contiguity in sequences in conversation. In G. Button & J. R. E. Lee (Eds.), Talk and social organisation (pp. 54–69). Clevedon: Multilingual Matters. Sacks, H. (1995). Lecture 8: On measurement. In G. Jefferson., & E. A. Schegloff (Eds.), Lectures on conversation: Volumes I & II: Harvey Sacks. Oxford: Blackwell. Sacks, H., Schegloff, E., & Jefferson, G. (1974). A simplest systematics for the organization of turn-taking for conversation. Language, 50(4), 696–735. Sarangi, S. (2002). The language of likelihood in genetic counselling discourse. Journal of Language and Social Psychology, 21(1), 7–31. Schegloff, E. A. (2007). Sequence organization in interaction: A primer in conversation analysis. Cambridge: Cambridge University Press. Schegloff, E., & Sacks, H. (1973). Opening up closings. Semiotica, 8, 289–327. Schoeb, V., Staffoni, L., Parry, R., & Pilnick, A. (2014). “What do you expect from physiotherapy?” A detailed analysis of goal setting in physiotherapy. Disability & Rehabilitation, 36(20), 1679–1686. Seale, C., Chaplin, R., Lelliott, P., & Quirk, A. (2006). Sharing decisions in consultations involving anti-psychotic medication: A qualitative study of psychiatrists’ experiences. Social Science & Medicine, 62(11), 2861–2873. doi: 10.S0277-9536(05)00584-8[pii] Shakespeare, T. (1998). Choices and rights: Eugenics, genetics and disability equality. Disability & Society, 13(5), 655–681. Sharrock, W. (1979). Portraying the professional relationship. In D. C. Anderson (Ed.), Health education in practice. London: Croom Helm. Shay, L. A., & Lafata, J. E. (2015). Where is the evidence? A systematic review of shared decision making and patient outcomes. Medical Decision Making, 35(1), 114–131. doi:10.1177/0272989X14551638 Sidnell, J. (2013a). Conversation analysis. An introduction. Oxford: Wiley-Blackwell. Sidnell, J. (2013b). Basic conversation analytic methods. In J. Sidnell & T. Stivers (Eds.), The handbook of conversation analysis (pp. 77–99). Oxford: WileyBlackwell. Sidnell, J., & Stivers, T. (Eds.). (2013). The handbook of conversation analysis. Oxford: Wiley-Blackwell. Silverman, D. (1997). Discourses of counselling: HIV counselling a social interaction. London: Sage.

References

145

Silverman, D., & Bloor, M. (1997). Patient-centred medicine: Some sociological observations on its constitution, penetration, and cultural assonance. In M. Bloor (Ed.), Selected writings in medical sociological research. Aldershot: Ashgate. Silverman, J., Kurtz, S., & Draper, J. (1998). Skills for communicating with patients. Oxford: Radcliffe Medical Press. Sperlich, S., Maina, M. N., & Noeres, D. (2013). The effect of psychosocial stress on single mothers’ smoking. BMC Public Health, 13, 1125. doi:10.1186/1471-2458-131125 Stevanovic, M. (2013). Deontic rights in interaction: A conversation analytic study on authority and cooperation. PhD thesis, Department of Social Research, University of Helsinki. Retrieved from https://helda.helsinki.fi/bitstream/handle/10138/39270/ stevanovic_dissertation.pdf?sequence51 Stevanovic, M. (2018). Social deontics: A nano-level approach to human power play. Journal for the Theory of Social Behaviour, 48(3), 369–389. Stevanovic, M. (2021). Deontic authority and the maintenance of lay and expert identities during joint decision making: Balancing resistance and compliance. Discourse Studies, 23(5), 670–689. Stevanovic, M., & Per¨akyl¨a, A. (2012). Deontic authority in interaction: The right to announce, propose and decide. Research on Language and Social Interaction, 45(3), 297–321. Stevenson, F. A., Barry, C. A., Britten, N., Barber, N., & Bradley, P. (2000). Doctor-patient communication about drugs: The evidence for shared decision making. Social Science & Medicine, 50(6), 829–840. doi:10.1016/S0277-9536(99) 00376-7 Stevenson, F., Hall, L., Sequin, M., Atherton, H., Barnes, R., Leydon, G., … Ziebland, S. (2019). General practitioner’s use of online resources during medical visits: Managing the boundary between inside and outside the clinic. Sociology of Health & Illness, 41(1), 65–81. doi:10.1111/1467-9566.12833 Stevenson, F. A., Seguin, M., Leydon-Hudson, G., Barnes, R., Ziebland, S., Pope, C., … Atherton, H. (2021). Combining patient talk about internet use during primary care consultations with retrospective accounts. A qualitative analysis of interactional and interview data. Social Science & Medicine. doi:10.1016/ j.socscimed.2021.113703 Stewart, M. A. (1995, May 1). Effective physician-patient communication and health outcomes: A review. Canadian Medical Association Journal, 152(9), 1423–1433. Stewart, M. (2001). Towards a global definition of patient centred care. British Medical Journal, 24(322), 444–445. doi:10.1136/bmj.322.7284.444 Stewart, M., Brown, J. B., Donner, A., McWhinney, I. R., Oates, J., Weston, W. W., & Jordan, J. (2000). The impact of patient-centered care on health outcomes. Family Practice, 49(9), 796–804. Stewart, M., Brown, J. B., Weston, W. W., McWhinney, I. R., McWilliam, C. L., & Freeman, T. R. (1995). Patient-centered medicine: Transforming the clinical method. London: Sage. Stimson, G. V. (1974). Obeying doctor’s orders: The view from the other side. Social Science & Medicine, 8, 97–104. Stivers, T. (2002). Participating in decisions about treatment: Overt parent pressure for antibiotic medication in pediatric encounters. Social Science & Medicine, 54, 1111–1130.

146

References

Stivers, T. (2005). Parent resistance to physicians’ treatment recommendations: One resource for initiating a negotiation of the treatment decision. Health Communication, 18(1), 41–74. doi:10.1207/s15327027hc1801_3 Stivers, T. (2006). Treatment decisions: Negotiations between doctors and patients in acute care encounters. In J. Heritage & D. Maynard (Eds.), Communication in medical care: Interaction between primary care physicians and patients (pp. 279–312). Cambridge: Cambridge University Press. Stivers, S. (2011). Prescribing under pressure: Parent-physician conversations and antibiotics. Oxford: Oxford University Press. Stivers, T., & Barnes, R. K. (2018). Treatment recommendation actions, contingencies, and responses: An introduction. Health Communication, 33(11), 1331–1334. doi:10.1080/10410236.2017.1350914 Stivers, T., Heritage, J., Barnes, R. K., McCabe, R., Thompson, L., & Toerien, M. (2018). Treatment recommendations as actions. Health Communication, 33(11), 1335–1344. Stivers, T., & McCabe, R. (2021). Dueling in the clinic: When patients and providers disagree about healthcare recommendations. Social Science & Medicine, 290, 114–140. Stivers, T., & Robinson, J. D. (2006). A preference for progressivity in interaction. Language in Society, 35(3), 367–392. Stivers, T., & Timmermans, S. (2017). Always look on the bright side of life: Making bad news bivalent. Research on Language and Social Interaction, 50(4), 404–418. Stivers, T., & Timmermans, S. (2020, March). Medical authority under siege: How clinicians transform patient resistance into acceptance. Journal of Health and Social Behaviour, 61(1), 60–78. doi:10.1177/0022146520902740 Stivers, T., & Timmermans, S. (2021). Arriving at no: Patient pressure to prescribe antibiotics and physicians’ responses. Social Science & Medicine, 27, 114007. Straus, R. (1957). The nature and status of medical sociology. American Sociological Review, 22, 200–204. Strong, P. (1979). The ceremonial order of the clinic parents, doctors and medical bureaucracies. Henley-on-Thames: Routledge and Kegan Paul. Strong, P. (1980). Doctors and dirty work—The case of alcoholism. Sociology of Health & Illness, 2(1), 24–47. Svensson, M. S., Luff, P., & Heath, C. (2009). Embedding instruction in practice: Contingency and collaboration during surgical training. Sociology of Health & Illness, 31(6), 889–906. Szasz, T. S., & Hollender, M. H. (1956). A contribution to the philosophy of medicine: The basic models of the doctor-patient relationship. Archives of Internal Medicine, 97, 589–592. Szasz, T. S., Knoff, W. F., & Hollender, M. H. (1958). The doctor-patient relationship and its historical context. American Journal of Psychiatry, 115(6), 522–528. doi: 10.1176/ajp.115.6.522 Takemura, Y., Sakurai, Y., Yokoya, S., Otaki, J., Matsuoka, T., Ban, N, … Tsuda, T. (2005). Open-ended questions: Are they really beneficial for gathering medical information from patients? Tohoku Journal of Experimental Medicine, 206(2), 151–154. doi:10.1620/tjem.206.151 Tanenbaum, S. J. (2015). What is patient-centered care? A typology of models and missions. Health Care Analysis, 23, 272–287. doi:10.1007/s10728-013-0257-0

References

147

ten Have, P. (1991). Talk and institution: A reconsideration of the ‘asymmetry’ of doctor-patient interaction. In D. Boden & D. Zimmerman (Eds.), Talk and social structure: Studies in ethnomethodology and conversation analysis. Cambridge: Polity Press. ten Have, P. (2007). Doing conversation analysis: A practical guide (2nd Ed.). London: SAGE. Thomas, G. M., & Rothman, B. K. (2016). Keeping the backdoor to Eugenics Ajar? Disability and the future of prenatal screening. American Medical Association Journal of Ethics, 18(4), 406–415. doi:10.1001/journalofethics.2016.18.4.stas1-1604 Timmermans, S. (2020). The engaged patient: The relevance of patient–physician communication for twenty-first-century health. Journal of Health and Social Behavior, 61(3), 259–273. Todd, A. D. (1989). Intimate adversaries: Cultural conflict between doctors and women patients. Philadelphia, PA: University of Pennsylvania Press. Toerien, M. (2021). When do patients exercise their right to refuse treatment? A conversation analytic study of decision-making trajectories in UK neurology outpatient consultations. Social Science & Medicine, 290, 114278. doi:10.1016/ j.socscimed.2021.114278 Toerien, M., Reuber, M., Shaw, R., & Duncan, R. (2018). Generating the perception of choice: The remarkable malleability of option-listing. Sociology of Health & Illness, 40(7), 1250–1267. doi:10.1111/1467-9566.12766 Toerien, M., Shaw, R., Duncan, R., & Reuber, M. (2011). Offering patients choices: A pilot study of interactions in the seizure clinic. Epilepsy and Behavior, 20, 312–320. Toerien, M., Shaw, R., & Reuber, M. (2013). Initiating decision-making in neurology consultations: ‘Recommending’ versus ‘option-listing’ and the implications for medical authority. Sociology of Health & Illness, 35(6), 873–890. US Institute of Medicine. (2001). Crossing the quality chasm: A new health system for the 21st century. Washington, DC: National Academy Press. Usher, C. (2007). Sociology of medicine. In G. Ritzer (Ed.), Blackwell encyclopaedia of sociology. Oxford: Blackwell. Versteeg, W. B. (2018). “How do you know?” Everyday negotiations of expert authority. PhD thesis, Department of Philosophy, University of Twente, The Netherlands. Retrieved from https://research.utwente.nl/en/publications/how-doyou-know-everyday-negotiations-of-expert-authority Voysey Paun, M. (1975/2006). A constant burden. Aldershot: Ashgate. Waitzkin, H. (1991). The politics of medical encounters. New Haven, CT: Yale University Press. Waitzkin, H., & Stoeckle, J. D. (1967). Information control and the micropolitics of health care: Summary of an ongoing research project. Social Science & Medicine, 10(6), 263–419. doi:10.1016/0037-7856(76)90070-6 Waitzkin, H., & Waterman, B. (1974). The exploitation of illness in capitalist society. Indianapolis, IN: Bobbs-Merill. Wang, N. C., & Liu, Y. (2021). Going shopping or consulting in medical visits: Caregivers’ roles in pediatric antibiotic prescribing in China. Social Science & Medicine, 290, 114075. doi:10.1016/j.socscimed.2021.114075 Watermeyer, J., & Penn, C. (2009). “Tell me so I know you understand”: Pharmacists’ verification of patients’ comprehension of antiretroviral dosage instructions in a

148

References

cross-cultural context. Patient Education and Counseling, 75(2), 205–213. doi: 10.1016/j.pec.2008.09.009 West, P. (1976). The physician and the management of childhood epilepsy. In M. Wadsworth & D. Robinson (Eds.), Studies in everyday medical life. London: Martin Robertson. West, C. (1983). ‘Ask me no questions’: An analysis of queries and replies in physician-patient dialogues. In S. Fisher & A. Todd (Eds.), The social organization of doctor-patient communication. Washington, DC: Center for Applied Linguistics. Whalen, J. (1995). A technology of order production: Computer-aided dispatch in public safety communication. In P. ten Have & G. Psathas (Eds.), Situated order: Studies in the social organization of talk and embodied activities. Washington, DC: University Press of America. Whalen, J., Zimmerman, D. H., & Whalen, M. (1988). When words fail: A single case analysis. Social Problems, 35, 335–362. Williams, R. (1976). Keywords: A vocabulary of culture and society. London: Croom Helm. Williams, C., Alderson, P., & Farsides, B. (2002). What constitutes ‘balanced information’ in the practitioners’ portrayals of down’s syndrome? Midwifery, 18, 230–237. Williams, V., Webb, J., Dowling, S., & Gall, M. (2019). Direct and indirect ways of managing epistemic asymmetries when eliciting memories. Discourse Studies, 21(2), 199–215. doi:10.1177/1461445618802657 World Health Organisation. (2007/2016). Framework on integrated people-centred health services. World Health Assembly adopts Framework on Integrated People-Centred Health Services. Retrieved from https://www.who.int/news/item/ 28-05-2016-world-health-assembly-adopts-framework-on-integrated-peoplecentred-health-services World Health Organisation. (2021). The end TB strategy. WHO/HTM/TB/2015.19. Geneva: WHO. Retrieved from https://www.who.int/news-room/fact-sheets/detail/ antimicrobial-resistance World Medical Association. (2020). Declaration on physician patient relationship (Declaration of Cordoba). Retrieved from https://www.who.int/news/item/28-052016-world-health-assembly-adopts-framework-on-integrated-people-centredhealth-services Young, I. M. (2001). Two concepts of self-determination. In A. Sarat & T. R. Kearns (Eds.), Human rights: Concepts contests, contingencies (pp. 25–44). Ann Arbor, MI: University of Michigan Press. Zayts, O., & Pilnick, A. (2014). Genetic counselling in multicultural and multilingual contexts. In H. Hamilton & W. Chou (Eds.), Routledge handbook of language and health communication. London: Routledge. Zottola, A., Jones, L., Pilnick, A., Mullany, L., Bouman, W., & Arcelus, J. (2021). Identifying coping strategies used by patients at a transgender health clinic through analysis of free-text autobiographical narratives. Health Expectations, 24(2), 719–727. doi:10.1111/hex.13222

Index Abandonment, 4, 23, 65–67, 85, 99–107, 115, 121 Activity contamination, 47 Activity-passivity model, 9 Advice giver category, 73–75, 84, 87 Advice giving, 33, 72–73, 75, 84 Affective neutrality, 86 Affirmative care, 107, 108, 114 Affordable Care Act, 13 Agency, human, 29 Agency, patient or client, 20–21, 29, 35, 79–81, 88–89, 116, 130 Agenda setting, patient or client led, 51–55, 84 Amniocentesis, 64, 69, 103 Antimicrobial resistance, 43–46, 61 Asymmetry, in doctor patient encounter, 8, 20, 24, 72, 89, 90–91, 98, 121, 123, 125 Auspicious interpretation, 69–70, 75, 104 Authority deontic, 85, 92–94, 97, 99, 101, 102, 105–106, 121, 125 epistemic, 85, 93, 94, 97, 99, 101, 104, 121, 125 Autonomy, 10, 99–107 and individual choice, 7, 21–22, 24–25, 37, 42, 46, 51, 85, 113–115 limits of, 115–117 and professionalism, 22–23 rise of, 20–21 Autonomy, patient, 7, 10, 20–23, 65, 67, 115–116, 118 Autonomy, relational, 22, 42, 59, 115–116

Balint, Enid, 8 Balint, Michael, 7–8, 121–122 Balint’s approach, 8 Beneficence, 21, 76, 82, 84, 115 Bio-psychosocial perspective, 8, 12 Bioethical approach, 21 Bioethics, 17, 21, 76 Broad questions, use of, 10, 48–51 Capacity, for self-determination, 21, 24, 35, 38, 108, 116 Checklist-based approaches, 4, 18, 20, 27, 35, 51, 58, 67, 76, 111, 117–118, 124–125 Choice, as an ideological device, 118 Choice, consumer, 24, 113–115, 118 Choice, individual, 7, 21–22, 24–25, 37, 42, 46, 51, 85, 113–115 Choice, patient, 15, 22, 35, 45, 58, 61, 83, 92, 105, 113–115 Choice concept, 101 Choice in context, 117–119 Choice-centred conceptualisation of medicine, 85 Chronic illness, 9, 72, 86, 97–99 Chronic illness context, expertise in, 97–99 Client-led agenda setting, 51–55 Clinical judgement, 100–101 Clinical practice guidelines, 115–116 Co-design, of services and care pathways, 122 Code and count tradition, 11 Communication impairments, 76 problem of communication in healthcare, 41–42 skills training, 33, 49, 76, 112

150

Index

Complicity of social science, 119–122 Concordance, 10 Conflict limitations of training as solution to interactional conflict in healthcare, 83–84 problem of managing conflicting moral norms in interaction, 76–83 Constituting expertise in interaction, 94–97 Consumer choice, limitations as applied to healthcare, 113–115 Consumer, 14, 24, 112 Consumerism, 5, 14, 24, 108, 112, 113, 114, 122 Consumerist models of medicine, 114 Context, expertise in, 117–119 Control, of healthcare interactions, 12, 16, 21, 35–37, 44–46, 54, 61, 63, 67, 68, 83, 88–92 Conversation analysis (CA), 27, 101 as method for studying healthcare interactions, 28–33 research in primary care, 91 specific contribution of, 123–125 COVID-19 vaccination, 85 Cultural relativity, 120 Decision making, bilateral, 106 Decision making, distributed, 119 Decision making, shared, 3, 7, 15–20, 35, 46, 59, 67, 92, 101, 104, 106, 111, 118, 123 Decision making, unilateral, 106 Deontic authority, 85, 92–94, 97, 99, 101, 102, 105–106, 121, 125 Diagnosis/treatment-oriented interaction, 14 Doctor-as-person, 12 Doctor–patient interaction, 28, 31, 50, 88–92, 117 Doctor–patient relationship, 3, 8, 9, 12, 19, 31, 36, 57, 90, 118–120 Dominance, interactional, 89, 90–91

Doorknob phenomenon, 48 Double bind, for professionals, 89–90, 107 Double bind, for patients, 89–90, 106 Effectiveness of PCC, 14–16 Epistemic authority, 85, 93–94, 97, 99, 101, 104, 121, 125 Epistemic primacy, 24 Epistemic stance, 93, 104 Epistemic status, 93, 104 Epistemics, 92, 101 Epistemics of experience, 4, 85, 104, 106, 122 Epistemics of expertise, 104, 122 Essential tension, 70, 71 Ethnomethodology, 29, 31, 34, 86 Eugenics, 51, 63 Expert patient, 115, 119 Expertise, 85 autonomy, abandonment and, 99–106 constituting expertise in interaction, 94–97 in context, 117–119 in context of chronic illness, 97–99 Expertise by experience, 119 Face, loss of, 72 Face threat, 55, 73, 81, 108 Garfinkel, Harold, 29, 31, 34, 61 Genetic counselling, 3, 33, 51, 66 Genetic medicine, 51, 99 Goffman, Erving, 31, 108, 116 Guidance-co-operation model, 9 Habermas, Jurgen, 10 Hall, Stuart, 88 Health Education England, 1 Health Foundation, 12, 13, 21, 23, 87 Healthcare conversation analysis as method for studying healthcare interactions, 28–33

Index good organisational reasons for bad healthcare practice, 34–35 limitations of consumer choice as applied to, 113–115 problem of communication in, 27–28 talk, compared with ordinary talk, 46–48 Holism, 8 Imperatives, moral, 76, 82 Imperatives, organizational, 2, 4, 27, 34–35, 41, 42, 46 Interactional conflict in healthcare, 4, 83–84 Interactional difficulty of nondirectiveness, 62–68 Interactional dominance, 89, 91 Interactional dysfunctions, 47, 58–59, 71 Interactional norms, 4, 30, 50, 61, 70, 71, 73, 75, 83, 84, 124 Interactional submission, 91 International Alliance of Patient Organisations, 12 Jargon, use of, 97–98 Jefferson, Gail, 29, 73 Legitimacy, 21, 57, 99, 124 Logic of care, 21, 105, 123 Logic of choice, 21, 46, 66–67, 87, 105, 123 Meaning-making in healthcare interaction, 64, 66 Medicine, 97, 125–126 sociology in, 5, 120 sociology of, 5, 119 Mental Health Act, 87 Mishler, Elliott, 10, 11, 74–75, 103–104, 123 Modern Western psychiatry, 86 Moral dimensions of healthcare, 56, 72 Moral norms in interaction, 30, 61, 75–83

151

Moral principles, in policy making, 76, 84, 125 Mutual participation model, 9 National Institute for Health and Care Excellence (NICE), 16, 76, 116 Neo-liberalism, 21, 66, 88 Non-compliance, 10, 114 Non-directiveness, interactional difficulty of, 62–68 One-size-fits-all approach, 9, 27, 58, 84 problem with, 46–48, 117–119 Open-ended questions. See Broad questions Ordinary talk as compared with healthcare talk, 46–48 Orphan consultations, 118 Parsons, Talcott, 8, 20, 29, 31, 57, 86–87, 98, 111, 115 Passivity, patient, 9, 58 Paternalism, 11, 58, 62, 92, 97, 99, 117, 125 Patient troubles telling, 73–75 Patient affirmation, 85, 107–109 Patient autonomy, 7, 10, 20–23, 65, 67, 115–116 Patient centred care (PCC) and the complicity of social science, 119–122 conversation analysis as method for studying, 28–33 difficulties of distinguishing between good and bad practice, 58–59 evidence for effectiveness of, 14–16 and good organisational reasons for bad healthcare practice, 34–35 and limitations of consumer choice as applied to healthcare, 113–115 and limits of autonomy, 115–117 person centred care vs., 23–26

152

Index

pervasiveness of, 26 shared decision-making and, 16–20 Patient centred medicine (PCM), 2, 10–12, 16, 23, 78, 101, 111, 123 Patient Centred Outcomes Research Institute, 13 Patient dependency, 9 Patient engagement, 67, 118 Patient-as-person, 12 Patient-centredness, 11, 14, 125 Patient-led agenda setting, 51–55 Patient-oriented medicine, 8, 117, 119, 122 Person centred care, 1, 2, 7, 13, 23–26 Person-centred interaction, 62 Personal health, 43–46 Personhood, 24–26, 76 Professional autonomy, 115 Professionalism, 86–87, 106 autonomy and, 22–23 Progressivity, 89, 91 Psychoanalysis, 86 Psychotherapy, 24, 33, 62–63, 93, 109, 117 Public health, 42–47 Quality of shared decisions, 17–18 Question design, the impact of, 38, 47, 50 Refusal, of requests, 76–78, 82 Relational autonomy, 22, 42, 59, 116 Risk, communication of, 33, 64 Risk, to health, 97, 129, 136 Rogers, Carl, 24, 62 Role convergence, 22 Roter Interaction Analysis System (RIAS), 11

Sacks, Harvey, 29, 73, 75 Schegloff, Emmanuel, 29 Shared decision making (SDM), 3, 7, 15–20, 35, 46, 59, 67, 92, 101, 104, 106, 111, 118, 123 Sick role model, 8, 22, 46, 57, 98, 115 Smoking cessation, 55, 57, 61, 74 Social science, complicity of, 119–122 Sociology in medicine, 5, 120 Sociology of medicine, 5, 119 Sociology of professions, 87 Structural functionalism, 29 Struggle for control, 4, 16, 21, 85, 88–92, 108 Theory of communicative action, Habermas’, 10 Theory/practice gap, 111 Therapeutic alliance, 12 Therapeutic citizenship, 46 Treatment recommendations, 32, 43, 45, 47, 88, 92, 105 Troubles telling, 73–75 Trust, 51, 76, 87, 117, 121 UK Department of Health, 1, 33, 40, 71, 98, 114, 121 UK National Institute of Clinical Excellence, 97, 116 Uncertainty, 35, 68–71, 75, 95 US Patient Centered Outcomes Research Institute, 2 Values-based policy, 121, 125 Voice of medicine, the, 103 Voice of the lifeworld, the, 10, 76, 103 Waitzkin, Howard, 28, 32, 44, 123 World Health Organisation (WHO), 23, 43

This page intentionally left blank

This page intentionally left blank