Evolving the multiple roles of ‘patients’ in health‐care research: reflections after involvement in a trial of shared decision‐making

Abstract

Objective This paper offers ‘consumer‐led’ reflections by steering group members of a patient‐centred research study involving consumer advocates, patients’ associations and patients, throughout the whole study, from pre‐ to post‐study phases.

Original Study Design The study: ‘Shared decision making and risk communication in general practice’ incorporated systematic reviews, psychometric evaluation of outcome measures, and quantitative, qualitative and health economic analyses of a cluster randomized trial of professional skill development, all informed by consumer and patient engagement.

Setting and participants The work was produced by a wide collaboration led by researchers from the Department of General Practice, University of Wales College of Medicine, Cardiff, including a consumers’ advisory group and a patients’ association. The study participants were 20 general practitioners from Gwent, their practice staff, and almost 800 patients at these practices.

Discussion Consumers and patients contributed to several stages of the research from inception and design, securing of funding, implementation of the protocol, and interpretation and dissemination of the findings. ‘Patient involvement’ research initiatives that include an equally wide variety of ‘user’ participants as ‘health‐professional’ participants, accountable to a ‘Health in Partnership’ funded project, require a user‐led viewpoint to be presented and disseminated. This paper presents reflections on the processes of the research, the interpretations of study findings by the involved parties, and notes how this model is fundamental to effective research in the field of patient‐centred health care if future practice, policy and research are to change.

Introduction

The consultation –‘The clinical encounter – the focal point of patient – centred care’ ¹ is the pivotal experience in the provision of health care to citizens and patients (our understanding of these and various related terms used in this piece is given in Box 1). The National Health Service (NHS) generates around 250 million consultations every year. ² Determining exactly what is a ‘successful encounter and how it is to be accomplished’ is therefore a critical question. How should ‘success’ be considered and defined, and by whom? Who should establish the criteria? In the UK, the general practitioner (GP) is the gateway for citizens seeking health care: their experience in the consultation is crucial for their satisfaction, other process outcomes, general well‐being and future care. Understanding what expectations patients have, and the components that will enable health‐care professionals to negotiate a satisfactory outcome, are obviously important, not just to the two parties involved at the time, but to inform and direct the provision of optimal and equitable health‐care services generally, including the training of health professionals. An unsatisfactory consultation is a bad beginning: it has the potential to adversely affect ensuing encounters, continuity of satisfactory care there or elsewhere, or accomplishment of satisfactory decisions to deal with the individual patient's health problem and concerns in the context of their own circumstances, values and preferences.

Table Box 1  .

Our understanding of the terms used in this paper:

• Patients – people using health‐care services for health problems or conditions

• Consumers – people using or offered health‐care services,

• Citizens – the general public not immediately using health‐care services

• Patients’ associations – individuals coalescing to support patients with a or certain condition(s)

• Consumer representative (or groups) – individual consumers (or groups) working to enhance health‐care quality for patients or consumers or citizens

These questions arise in a broader context of ethical and conceptual developments. There are debates about the balance between personal autonomy and social responsibility ³ with attempts to define a new kind of relationship or even equality between the health‐care profession and consumers. These occur in a system that also promotes evidence‐based health care – which may benefit populations more than individuals, and therefore creates tensions. ⁴ ‘Shared decision making’ (SDM) has been proposed as a model that can address some of these competing tensions in real life, but it also means different things to different people – both professionals and consumers. It is also difficult to put into practice: weighing up the harms and benefits of different options of a healthcare decision is challenging, even if they are known and understood. Professionals first need help and training about risk perception, communication and management if they are to be useful in helping consumers come to satisfactory decisions.

Modern public health influences on the doctor–patient relationship

There is a wider context and influence. Throughout history until comparatively recently, the clinical encounter was initiated by the concerned individual with symptoms, who sought the help of a medical practitioner. But things have changed. Public health medicine and epidemiological research have gone beyond replacing a defective and contaminated Broad Street London water pump, to the establishment in the UK of extensive NHS prevention, vaccination and screening with the potential to benefit large sectors of the population. But when health practitioners are provided with incentives, or are set percentage targets to ensure viability of national programmes, in a reversal of the usual doctor–patient relationship, relationships sometimes become strained in this atmosphere of ‘coercive healthism’. ⁵ Medical screening has been described as ‘institutionalisation of risk’. ⁶ The ‘relationship‐reversal’ turns citizens into patients. Individuals who sought reassurance find themselves having to make very difficult risk assessments – many then regret having responded to persuasive invitations to present themselves for screening. This has had important ethical, psychosocial, epistemological, economic, political and policy implications, with consequences and repercussions. One of the repercussions has been to cause a shift in perceptions of the ‘usual’ doctor–patient relationship.

Need for research: Department of Health ‘Health in Partnership’ (HiP) Programme

The UK Department of Health recognized that a thorough research exploration of the primary encounter in the ‘usual doctor–patient relationship’ was needed. Such research might provide evidence that could be a basis for consultation approaches that accommodated more closely the desires of both parties in the process, and encompassed and met a wide variety of needs and aspirations in a multiplicity of contemporary settings. It might also provide evidence to improve training programmes. Crucially, developments in the way health care is delivered needed to take account of the range of views held by different people. The UK Department of Health commissioned a programme of research entitled ‘Health in Partnership’, and here we present an overview of one of these studies, drawing together particularly the findings about what patients in the study reported as important. We will first present a summary of the study and its findings, and then reflect on relevance to the conceptual developments noted above regarding patient involvement in research.

The ‘Health in Partnership’ study of SDM ⁷

This study was undertaken in Wales by a multidisciplinary group of researchers including GPs (including AE and GE), a research officer, a statistician, a research psychologist, a professor of health services research, health economists, five members of a school of care sciences (principally nursing), and an independent advocate for quality in research and healthcare (HT). The latter was founding chairman of the Consumers’ Advisory Group for Clinical Trials (CAG‐CT). The Study Steering Group thus comprised 18 members from a diversity of disciplines. In addition, there were nine other contributors to specific parts of the study, including sociologists, GPs, a health‐care scientist, a decision‐making research specialist and a representative from a patients’ association. The participants in the study itself were 20 GPs from Gwent, their practice staff, and almost 800 patients at these practices. The Gwent Local Research Ethics Committee approved the study. Regardless of amount of input, the contribution of every individual was viewed as equal in weight and a necessary component of the study. This paper later summarizes some of the research findings that are relevant to these reflections on patient and consumer participation in the research process. The full details and publications arising from the study are available at: http://www.healthinpartnership.org/studies/edwards.html. ⁷

Consumer contribution

Patient and consumer groups or representatives were engaged at several stages of the research process from inception to completion. These are listed in Table 1. The consumer representatives’ contribution came primarily from the CAG‐CT, which also provided the individual ‘advocate’ to contribute to the study Steering Group. This contribution will therefore be described in some detail.

Table 1.

The contribution of consumers to the different stages of the research

Stage of research process	Contribution of consumers to research
Setting research questions	Advocate engaged; consumer representatives and groups engaged in qualitative research 8
Protocol design	Advocate and patient association groups involved in formulation and drafting
Acquisition of funding	Advocate and patient association groups involved as co‐applicants
Implementing methodology	Advocate engaged as trial scrutineer
Interpretation of results	Advocate involved as Steering Group member for analysis and interpretation
Dissemination for policy and practice	Advocate involved in publicity, publications and presentations at press and scientific meetings

The CAG‐CT was involved from the pre‐trial phase, and contributed to protocol development and one of the early developmental focus groups. ⁸ This developmental work (with 47 consumer and patient focus group participants) explored the important outcomes for patients, when considering decision‐making approaches, information availability and involvement. Views of users and professionals are equally important in arriving at conclusions about shared decision‐making. It was important to establish their degree of congruence by using rigorous qualitative research methods that also took account of possible biases. ⁹ , ¹⁰ The focus group participants identified many affective outcomes that were consistent with current literature trends. However, many cognitive and behavioural outcomes that had been assessed in the literature were not thought to be important by the focus group participants. Importantly, a broader range of outcomes than was evident in the literature of the time was viewed as important to participants. These included feeling respected as a ‘player’ who has made a meaningful contribution to the discussion in the decision‐making process; the need for available and accessible extra information sources for better assessment of the information given; the opportunity for involving others; continuity of care, including clear arrangements for review of treatment decisions. ³ , ⁸ , ¹¹

The importance of pre‐trial qualitative research

This confirms the importance of undertaking qualitative research with consumers before completing a study outline and undertaking collaborative preliminary work with consumers in order to refine both the interventions and intended outcome measures. These strategies make the proposed research more relevant. ¹² , ¹³ , ²⁵ , ²⁶ The ethos of the CAG‐CT, a joint facilitative, working group of health professionals and consumers, rendered it the ideal consumer group to be involved in the preliminary exploration of a trial question. Not only was it familiar with engaging in iterative discussion for research processes as a small mixed group, for which it had received endorsement ¹⁴ , ¹⁵ but it could also provide consumer members for the focus group work who had had practical experience and enjoyed this type of iterative working within a research project. For example, the multi‐centre trial of hormone replacement therapy (HRT) in women with early breast cancer was informed by a research project and feasibility study planned and undertaken by the CAG‐CT in 1995. ¹⁶ In this project, the focus group work, facilitated by the King's Fund, London, identified and prioritized the desired outcomes of patients, researchers and clinicians, now included in the main multi‐centre HRT trial protocol. ¹⁷ It also identified the specific training needs for those who would run the trial and the information needs of the participants, patients and health professionals. The focus group process was one of clear ground rules and methodology for arriving at prioritization, enabling satisfactory and equitable input from a very mixed group of researchers, clinicians, patients and advocates. These principles were modelled again in the ‘Health in Partnership’ study of SDM. ⁷ After the initial inductive stages, patients and consumers were actively engaged at all further stages, including first the development of outcome measures. The further stages of the research itself at which patient or consumer data were evaluated are listed in Table 2, and will now be addressed in turn.

Table 2.

Methods used for different elements of the evaluation of this HiP Shared Decision Making study

Stage of study	Method
Assessment of relevant patient‐based outcomes: pre‐trial phase	Focus group study with patient and consumer representative groups
Piloting and evaluation of patient‐based outcome instrument (questionnaire)	Semi‐structured interviews and quantitative (factor) analysis of patient responses to questionnaire
Health economic analysis of patient preferences and utilities	Quantitative (multi‐level regression) analysis of patient responses to a further questionnaire
Qualitative methods of further enquiry	In‐depth interviews with patients immediately after study consultations
Micro‐analytic qualitative methods	Discourse analysis of interviews with further patient sample

Consumer and patient involvement in developing outcome measures in the Welsh Study

The objective of this stage was to develop and validate an instrument that addressed the range of outcomes identified by users as important, ⁸ some of which might be difficult to measure or to assign appropriate weight to. ¹⁸ Consumers and patients were involved in the piloting and revision of the instrument through focus groups, one‐to‐one interviews after completing questionnaires, and postal survey. After a series of iterations a finalized instrument was produced, known as ‘COMRADE’ (Combined Outcome Measure for Risk Communication and Decision‐making Effectiveness). ¹¹ The preliminary qualitative research with patient and consumer focus groups had thus not only informed the nature of the interventions for the SDM trial but had also informed the development of an outcome measure to evaluate risk communication (RC) and decision‐making (training) effectiveness in UK primary care.

This work offers a model for undertaking qualitative research in collaboration with ‘real patients and consumers’, both before setting up a randomized controlled trial, throughout, and post‐trial, including the development of evaluation instruments and their use. ¹¹ Improvements in undertaking satisfactory consultations depend on the availability of sensitive measurement instruments to measure what is being achieved. Crucially, the ‘success’ of this type of research should be interpreted and judged, not just against biomedical targets, but also against criteria that patients have identified. If delivery of patient‐centred health care is to become a reality, and satisfactory decisions achieved, it is vital to know what citizens value most when they become patients, and to be able to measure accurately and reliably.

Health economic methods to assess patient preferences

Extending the patient ‘level’ evaluations, a multi‐level analysis of data derived from questionnaires administered 6 months after a study consultation to 757 patients revealed the attributes within a consultation that patients valued most. ¹⁹ This data was derived by a discrete choice experiment with a design derived from Conjoint Analysis methods. It assesses strengths of preference, by quantitative methods, for various ‘attributes’– in this case, attributes of consultations such as time, information, involvement in decision‐making etc. The attribute that patients valued most was that the doctor listens, followed by ‘information easy to understand’, ‘who chooses’, ‘amount of information’ and ‘length of consultation’. In terms of involvement in decision‐making, patients most preferred a consultation in which they contributed but did not have sole responsibility for the decision taken (shared model). Least preferred were consultations where the doctor alone makes the decisions (paternalistic). Significantly, patient preferences were influenced by whether the patients’ doctors had received SDM or RC training interventions prior to the study consultation, suggesting that actual experience of involvement leads to people valuing it more.

This section of work thus identified patients’ priorities in consultations in a quantitative way, to complement other qualitative analyses. These data may be able to contribute to health economic analyses of further developments in health‐care provision (such as whether patient involvement is ‘worth it’, or whether training interventions are cost‐effective), and thus ensure that appropriate consumer outcomes are available for any policy decisions that may subsequently be made.

Qualitative methods of further enquiry

Broader qualitative work in this study also demonstrated that it is possible to enhance the content of consultations and its perceived value to both doctor and patient. One‐to‐one interviews with patients after study consultations found that RC interventions to doctors sharpened the focus of consultations, changed the content of the consultation, and resulted in a greater perception of decisions actually being made in the consultation. Doctors and patients often interpreted shared decision‐making differently: patients perceived few changes. In practice, doctors selectively applied SDM and RC approaches as appropriate in each consultation. These findings provided more detail on the process of what was happening within consultations, even when the outcomes were quite resistant to change: global ‘satisfaction’ with consultations was high before and after the training interventions to doctors, as measured quantitatively at group level. The findings provide empirical support for the theoretical and ethical justifications for professional training developments in communication skills. These skill developments should be incorporated into professional training, with emphasis on the benefit of their flexible use and selective application, so that GPs may be encouraged to persist in their selective use in the consulting room. Training is needed not only to improve doctors’ ability to engage in RC, but also to give them confidence to use it more widely, but selectively.

Micro‐analytic qualitative methods to examine patient perspectives

The last phase of the research was a post‐consultation detailed qualitative analysis of accounts of newly presenting patients who had engaged in consultations with GPs previously trained in the use of SDM and risk communicating skills. ¹⁰ Analysis of interviews conducted with this purposive sample was by a method combining Discourse Analysis and Membership Category Analysis. This method rests on the principle that people are what they are as a result of their activities. The consistent message emerging from an interesting variety of dynamics of consultations was that the doctor should listen; take concerns seriously; and communicate effectively. Ultimate decisions were shown to be dependent on a series of sequential decisions each of which rely on the doctor's involvement, at different levels. Patients continually expressed alignment with the doctor, but also actively retained their authority in the process. This rigorous analytic method, examining the ‘negotiation of meanings’, demonstrated how the understanding of the nature of patient involvement can be enhanced. It can feed into training and skill development programmes, probably mainly at post‐graduate level, so that doctors (and other health‐care professionals) grasp both the conceptual understanding, methods for practical application, and patient perspectives and desires from consultations.

Synthesis

There are significant potential shifts in how health care could be provided. We are in a transition period between the old‐style paternalistic mode of consultation and prescribing to the patient‐centred shared decision‐making model and the wider use of RC tools in a sensitively negotiated exchange of information. Consumer or patient involvement is increasing, although arguably from a very low base. However, data from patients in this Welsh ‘Health in Partnership’ study ⁷ indicate that once patients have experienced involvement they place a higher value on it – they will increasingly value it in the future. Once involvement is desired, the next (almost automatic) desire is for information to support the decision‐making. Qualitative data here indicated that after interventions to enhance RC, the consultations were more likely to result in active decisions about treatment or care. However, this is not yet easy to deliver in practice. Professionals need more training and skill development. Information provision (for both professionals and consumers) needs substantially more resources directed towards it. But even with more information, the public's understanding of risk also leaves much to be desired. ²⁰ , ²¹ Understanding must be facilitated by balanced use of information already available, and must be explored sensitively in discussions between professionals and patients. ²² The caring effect of consultations must not be neglected.

There are also risks for health‐care systems, which are at risk of becoming an industry rather than a service. Achieving behavioural change in the population is a major pre‐occupation, frequently undertaken with little understanding of people's motivations and needs. ²³ , sometimes with commercial and political rather than patient‐orientated objectives in mind. What is needed now is not so much public understanding of science as understanding of the public by scientists. ²⁴

This brief description of a research project attempts to show how the public can be engaged successfully and in a responsible and respectful partnership so that important consumer and patient perspectives and outcomes are identified, and used in a responsible and ethical manner. This model of partnership was achieved by both clinicians and researchers in this study exploring shared decision‐making and the use of risk communicating tools in general practice. ⁷ There are other recent examples of similar research partnership activity, employing entirely different methodologies. ¹⁷ , ²⁵ , ²⁶ By addressing consumer and patient concerns and identifying workable solutions, it is possible to explore how the conceptual developments (SDM) might be translated into the reality of clinical practice.

To date, health‐care developments have largely been led by policy‐makers, professionals and academics. It is not certain that they will have always addressed the processes of care and outcomes that are most important to citizens, consumers or patients. Early qualitative work in this study suggests the opposite. There needs to be a fundamental realignment of research development towards the values and needs of the public. This research project attempted to do this, with a view to findings being available for future policy debates and developments. There are a variety of findings from a variety of methods in this project but each is important. At a more immediate level, the findings are important for consumers and patients who want to be involved and feel respected for their participation in health‐care decision‐making. ⁸ At a more global level it is also possible to envisage how this may influence how risks are discussed, interpreted, managed and perhaps therefore ‘governed’ in our society.

Conclusion

If we are to derive maximum benefit from this research work, with its multi‐faceted consumer and patient evaluations, it is important to appreciate that responsibility for developing satisfactory and satisfying consulting styles does not rest solely with professionals. The experience of user involvement was valued by all study participants in this work. There is a responsibility on both consumers and consumer groups, and patients and patient groups, not only to encourage professionals to practice in this way, but to ensure that there is no misuse of these approaches by those charged with the wider governance of risk. ²⁴ This needs active, intelligent promotion by patients and consumers (both individual and group) if maximum return and sound ethical practices are to be achieved from the outlay for this research work. In turn, recognition by policy‐makers will also be important, with active support to enable consumer organizations to disseminate these findings and initiate action.

The main requirement for incorporating users’ input into research is researchers who believe that the validity of the work is enhanced by ensuring that the voice of the patient has an equal opportunity to be heard from inception through to dissemination. ²⁵ , ²⁶ It is essential that all collaborators are able to engage in well‐facilitated respectful debate and iteration within sound methodologies so that no voice, including the consumer's, has greater weight than another's in deciding, for example, on methods of outcome evaluation. ²⁷ More importantly, the development and implementation of these findings requires attitudinal shifts in patients, professionals, policy makers and purchasers, not in isolation, but in acute awareness of the need to identify and shoulder responsibilities together, in respectful iteration and collaboration, to forge new and better relationships. The evidence generated and subsequent decision‐making is then likely to be more balanced through not having neglected to include important stakeholders in the process. The goal of improving processes at the focal point of care – the consultation – may be much more attainable in this light.