Listening to the views of people affected by cancer about cancer research: an example of participatory research in setting the cancer research agenda

Abstract

Aim  The study ‘Listening to the Views of People Affected by Cancer About Cancer Research’ is currently exploring the views people affected by cancer have about cancer research and identifying their research priorities. Integral to this is the broader aim of ensuring an effective, collaborative participation of patients and carers in the design and conduct of the study. On the basis of experiences with the study to date, the latter is explored in this paper.

Design  The study adopts a ‘participatory research’ approach entailing the formation of a ‘reference group’ and a subsequent patient and carer co‐researcher group. Patient and carer members of these groups were identified through the patient forums of UK cancer networks and by approaching ‘hard to reach’ representatives directly through community groups and participating study sites.

Findings  Experiences from this study illustrate that a ‘participatory research’ approach is appropriate in engaging patients and carers in the research process. Establishing a group of people affected by cancer in the study was found to be particularly effective in enhancing the design and conduct of the research.

Conclusions  ‘Participatory research’ offers an effective means of involving patients and carers throughout the research process, thus strengthening the relevance and appropriateness of research findings and methods.

Introduction

In recent years, the involvement of patients both in developing health services and in the conduct of research has steadily increased in importance. The government's commitment to patient involvement is evident with the launch of initiatives such as Patient Advice and Liaison Services and the Commission for Public and Patient Involvement. ¹ Similarly, the importance of involving patients in research has been recognized through the continuing work of INVOLVE and in publications such as ‘Research and Development for a First Class Service: R&D Funding in the New NHS’. ²

Within cancer research, whilst local initiatives have involved patients in the research process, it is clear that there is still a need for a national approach for patient involvement in cancer research. ³ This paper reports on our study, ‘Listening to the Views of People Affected by Cancer About Cancer Research’– an innovative national project that aims to facilitate patient and carer consultation on cancer research priorities. The paper details the collaborative nature of the study and the way in which a ‘participatory research’ approach has been adopted to ensure full patient and carer involvement in the project from its design through to dissemination. Through reference to our study, this paper will illustrate the appropriateness of ‘participatory research’ methods in ensuring patient involvement in cancer research.

Listening to people affected by cancer about cancer research: the context

Patient involvement: the UK context

The Department of Health has placed significant emphasis on supporting collaboration between patients and health professionals in the development, evaluation and monitoring of healthcare services. In 1998, the Secretary of State for Health called for a collaborative, ‘partnership’ dynamic between the state, health professionals and patients. ⁴ The principles of collaboration were reiterated in the Health and Social Care Act 2001, ⁵ which required strategic health authorities, Primary Care Trusts and NHS Trusts to make arrangements to involve and consult patients in the development of health services. Initiatives such as Patient Advice and Liaison Services, Patient Forums and the Commission for Patient and Public Involvement in Health have therefore been established to support such collaboration. ¹ More recently, the Department of Health has recognized the need to avoid tokenism and sustain a committed liaison with patients:

Real patient and public involvement is not about ticking boxes, it is about NHS organisations developing constructive relationships, building strong partnerships and communicating effectively. For patients’ experiences of health services to really improve, NHS staff will need to have ongoing and meaningful dialogue with them, their carers and the public about improving and developing services. ⁶

Sustaining a collaborative dynamic between patients, health professionals and academics is important in the conduct of health research. The development of ‘research governance’ in particular has called for a closer working relationship between patients, relatives and representative organizations, research, universities and health and social care professionals. ⁷ A commitment to patient involvement has also been detailed in the ‘Research and Development for a First Class Service: R&D Funding in the New NHS’. ²

The advantages that patient involvement can bring to research have been well documented. Involving patients can improve the ethical nature of research, the appropriateness of research methods, the dissemination of results, and the goodwill of participants. ⁸ , ⁹ , ¹⁰ Despite these advantages, it is surprising that relatively few studies or research organizations have actively involved patients. In a report published in 2001, it was found that only 42% of NHS providers receiving R&D Support Funding involved patients in their research activities. ¹¹

Patient involvement: the cancer research context

Within cancer research, a survey conducted by the US National Cancer Institute found that although patients and carers were supportive of cancer research, they felt that research often supported the interests of clinicians and academics more than the communities being researched. ¹² The National Cancer Research Institute (NCRI) has suggested that a key component in enhancing public confidence in cancer research is a greater involvement of people affected by the disease in research and research prioritization. ¹³

Cancer research organizations outside the UK have been successful in involving people affected by the disease in research and more specifically in setting research priorities. The National Cancer Institute, for example, has involved advocacy groups in advising national research agendas through their representation on ‘Progress Review Groups’. Similarly, the National Breast Cancer Foundation of Australia has undertaken a national cancer research prioritization strategy through consultation with patients, members of the medical and scientific community and policy makers. ¹⁴

However, in comparison with these organizations, cancer research funding bodies in the UK have typically been less advanced in involving patients and carers. In 2002, the Macmillan Research Unit at the University of Southampton undertook a telephone/online survey of the then 15 organizations comprising the NCRI and 20 non‐cancer specific medical research bodies. ¹⁵ Both the NCRI member organizations and the non‐cancer research bodies composed of government and charitable bodies. This survey examined how priorities for funding cancer research were determined and identified the extent to which patients were involved.

Only five of the 15 NCRI member organizations consulted with patients or carers in setting research priorities. Those that did involve patients employed a range of strategies. The Scottish Executive Health Department, for example, consulted with patient representatives to develop their research strategy through ‘Portfolio Steering Groups’. By contrast, the Medical Research Council consulted with Consumer Liaison Groups that involved patients identified through national advertising in mainstream press and subsequent interviews. Consumer Liaison Group members have commented on patient information leaflets submitted as part of proposals for clinical trials as well as on corporate publications. They have also contributed to Medical Research Council responses to Government consultation documents and have attended MRC board strategy meetings.

In contrast, this study demonstrated that several of the non‐cancer funding bodies had developed sophisticated means of patient engagement. The Alzheimer's Society, for example, has established the Quality Research in Dementia Advisory Network, divided into 13 regions in England, Wales, Scotland and Northern Ireland, each involving a co‐ordinator and a maximum of 15 participants.

It was noticeable that two of the cancer research funding bodies involving patients were government bodies and therefore were required to involve patients. In contrast, within the non‐cancer funding bodies, the organizations leading in patient involvement were charitable bodies such as the Alzheimer's Society and the Multiple Sclerosis Society. ¹⁵

In recent years, however, there have been significant advances in the area of patient involvement in cancer research. These include the formation of the Consumer Liaison Group of the National Cancer Research Network and the development of Consumer Research Panels by the National Cancer Research Network and Macmillan Cancer Relief. The Consumer Research Panels are currently being piloted through the Cancer Networks in three regions and involve a collective of people affected by cancer working with academic institutions to strive towards greater involvement of patients in research. The Supportive Care and Psycho‐oncology Research Group based within the North Trent Cancer Research Network provides an excellent example of ways in which patients can be involved throughout the research process. ³

Beyond cancer research, advances have also been made in public engagement in research. For example, Telford et al. recently conducted a survey to assess the extent to which the patient engagement policy was being implemented in an NHS region. ¹⁶ Whilst researchers were found to be involving consumers in research in less than a third of trusts, they did detail examples where patient and public engagement has been successful. These include a Cochrane Collaboration group (where consumers participated in choosing topics for systematic reviews, as well as in writing protocols and reviews, and were consulted on the review before it was published to ensure the ‘user‐friendliness’ of the language), and a project on consumers’ knowledge, opinions and feelings concerning aspects of maternity care (where consumers identified the need for research, collaborated in refining the research question and methodology, and provided ongoing consultation).

Patients have been involved in generating wider priorities in non‐cancer settings. For example, Barnes and Bennett assessed the role of ‘user panels’ in identifying the needs and experiences of older service users recruited through service agencies and other local organisations. They found that user panels were particularly helpful in involving older service users as they were consultative and permitted participants to generate their own ideas. ¹⁷

However, certain difficulties have been noted in involving patients in cancer research. Often, it is difficult to ensure that patients who become involved are representative of a range of different backgrounds. Such difficulties are not unique to cancer research but exist in the wider research arena. There has been a tendency to identify patient representatives from existing networks who are therefore less likely to involve people from typically ‘hard to reach’ groups, such as patients from ethnic minoritized backgrounds, patients receiving palliative care, those with rarer cancers, and those living in rural locations. ³ In addition, attempts at involving patients in cancer research have often been undertaken in a local and piecemeal fashion with little co‐ordination regionally or nationally, or with little strategic overview. ³

Involving patients in cancer research clearly entails significant challenges both in the way that patients are consulted and in the way that research is undertaken. In the face of these challenges, Macmillan Cancer Relief has recently commissioned and funded a national study specifically aimed at exploring the views people affected by cancer have about cancer research and identifying their research priorities.

The study design

Our study adopts an exploratory, qualitative, ‘participatory research’ approach and is currently recruiting participants from 11 sites across the UK located in Northern Ireland, Scotland, Wales, London, north of England, Midlands and the south of England. Focus groups are the principal method and are being conducted in a three‐phase process. In the pilot phase, two focus groups were conducted with participants over the age of 18. Twelve participants were sought for each focus group, although there was deliberate over‐recruitment here as we expected some non‐attendees and duly between six and eight participants subsequently took part in the focus groups. The pilot phase examined the feasibility of the focus group questions and explored whether any themes generated in discussion should be incorporated into subsequent phases of the study.

Phase 1 of the study involves ten focus groups with purposively sampled participants aged over 18 recruited from seven cancer centres distributed across the UK. This allows a comparison of general perspectives on research and research priorities to be made. One focus group is being held at each cancer centre involving a maximum of 12 participants. Recruitment is being undertaken primarily by research nurses, Clinical Nurse Specialists and other members of the healthcare team using a sampling matrix based on gender and stage of treatment. A limit of four breast cancer patients is set for all focus groups to ensure that a wide range of cancer types is represented. All eligible potential participants are approached regardless of previous participation in research, attitudes to research or perceived degree of articulacy. By approaching patients in this manner, potential biases of research ideas or interests through selecting one particularly vocal group are minimized.

Phase 2 involves a series of six focus groups with purposively selected population groups. This provides an exploration of the research views and priorities of participants with specific population characteristics often underrepresented in research studies, including people from diverse ethnic minoritized backgrounds (South Asian, Black), specific age groups [teenagers (13–19 years), over 75 years old], and patients in the palliative phase of illness.

Given the high interest in patient and carer views in this study, it was felt necessary to involve people affected by cancer throughout the research process. Consequently, patients and carers have been involved in the design and conduct of the study. Furthermore, patient and carer volunteers have received training and support to co‐facilitate the focus groups with the Macmillan Research Unit and to analyse the subsequent data. The success of the study is, in part, due to this collaborative approach and the careful adoption of the tenets of ‘participatory research’. The rest of the paper therefore explicates the principles of ‘participatory research’ and demonstrates its effectiveness in ensuring a sustained dynamic of patient involvement through specific references to our study.

Participatory research

‘Participatory research’ is seated in many different philosophical traditions but has at its root the principle of research as a democratic process in which participants occupy the role of an active citizen within the study. ¹⁸ ‘Participatory research’ strives to transform the ‘social relations of research production’, overturning the traditional perspective of a dominant researcher and a submissive research ‘subject’. ¹⁹ Instead, the roles of the researcher as an unobtrusive observer not influencing the behaviour of research subject and the participant being involved in the study only through the invitation of the researcher are blurred. ²⁰

Different terms have been used in relation to ‘participatory research’, including ‘participatory research’, ‘action research’ and ‘participatory action research’. Whilst many believe these terms to be synonymous due to their shared concern for community action and involvement, others draw distinctions between ‘participatory research and action’/‘participatory action research’. In particular, certain commentators perceive ‘participatory action research’ to be more ‘expert‐directed’, involving a lead researcher taking a major role in a study whilst participants take a back seat. ¹⁹ , ²¹ Hence certain researchers prefer to use the term ‘participatory research’ to refer to a process whereby participants work in a truly collaborative manner with the research team. ¹⁸ ‘Participatory research’ is not defined by any specific method but is characterized more as an approach to research. ²²

Applying participatory research to health care: learning from our study

Recruiting patients and carers into a participatory research group

A common facet of ‘participatory research’ is the formation of a ‘participatory research group’ to guide and inform the research process. Recruitment and membership of the ‘participatory research group’ can vary and Heron and Reason have outlined four modes of inquiry that influence the nature of the group: ²³

• • same role inquiry (where participants have the same background, e.g. all GPs);
• • reciprocal role inquiry (where participants have a similar social and cultural position, e.g. patients and spouses);
• • counterpartal role inquiry (where participants have different backgrounds but are linked by a particular process, e.g. GPs and their patients);
• • mixed role inquiry (where participants are from different backgrounds, e.g. patients, academics, health professionals).

The formation of the ‘participatory research group’ is dependent on the needs of the research project. For example, in a study exploring patient–nurse relationships with patients with life‐threatening or terminal illness, a same role inquiry was adopted due to the importance of providing a supportive, non‐threatening environment in which participants can share ideas. ²⁴ In contrast, Chenoweth and Kilstoff employed a counterpartal role inquiry, inviting clients, day care staff and family carers to ‘participatory group’ meetings in an effort to improve community dementia management. ²⁵

In our study, the impetus of the research came from the fact that a there is a paucity of literature specifically related to people affected by cancer determining their own research priorities. Consequently, it was decided to adopt a same role enquiry for the study where all participants in the ‘participatory research group’ (termed the ‘reference group’) were either patients, past patients, carers or people working specifically with patients in a consumer involvement capacity in Macmillan Cancer Relief. This decision is supported by the difficulties noted by others in establishing ‘participatory research groups’ with participants from mixed backgrounds where power differentials between, for example, health professionals and patients have been difficult to manage. ²⁶ In addition, it has been suggested that effective collaboration between participants may be limited if the experiences and backgrounds of group members are too diverse. ²⁷

There are concerns, however, that a patient group may become too homogenized. ³ Stevens et al. commented on the weaknesses inherent in recruiting participants from pre‐existing groups, often resulting in the exclusion of those who do not enjoy the same avenues of involvement, such as participants from diverse ethnic minoritized backgrounds, those in the palliative phase of their illness, those with rarer cancers, those who are elderly and those in rural locations. Consequently, in our study, a combined approach was used to ensure a range of participants in the ‘reference group’. All patient forums in each of the then 40 cancer networks in the UK were contacted asking for volunteers to take part in the research (the Cancer Network in Northern Ireland was not established at this time).

It was interesting that this exercise resulted in a more mixed group than was expected from the literature. At the initial meeting, of the 25 volunteers, 11 were male and 14 were female. Whilst the majority of representatives were patients, three were involved in a caring capacity. Furthermore, the volunteers represented a range of cancer types, rather than predominantly coming from the same background, e.g. breast cancer. However, it was evident that certain population characteristics were not represented in the ‘reference group’. There were no representatives from ethnic minoritized backgrounds, nor were there representatives from people in the palliative stage of their illness. As a consequence, certain individuals have been approached directly either through other groups (as is the case for non‐white representatives) or through participating centres (for example, patients in the palliative stage of their illness who were approached through a hospice).

The ‘reference group’ met at the beginning of the study and it was evident that the participants reflected a broad spectrum of research attitudes and awareness from the research novice to the relatively research informed, and from the research positive to the research sceptical. The meeting was co‐ordinated by the project researchers and ‘reference group’ members were given a summary of the preliminary design of the study, patient information sheets and a proposed list of focus group questions prior to the meeting. The meeting was successful in that patients and carers readily comprehended and responded to complex research issues resulting in changes made to the research design, the recruitment process, the study question schedule and patient related material, such as the ‘patient information sheet’. The precise input is as follows:

• • The ‘reference group’ approved of the overall design of the study and the use of focus groups, but stressed the importance of capturing individual responses, e.g. by the use of follow‐up questionnaires.
• • The ‘reference group’ assisted in the clarification of the ‘post‐treatment’ category of participants to be approached in the study.
• • The ‘group’ broadly agreed with the recruitment strategy, but emphasized the need not to select participants who were perceived to be research positive or experienced.
• • The ‘group’ changed the ‘patient information sheet’, rewording sections to make it more accessible and changing and increasing the font to make it more legible.
• • The ‘group’ examined the proposed questions for use in the focus group and rephrased some of these to make them clearer.

The nature of participatory research groups – the patient and carer co‐researcher group

The ‘reference group’ will continue to meet through the course of the study to comment on various aspects of the project. The group has already met to explore the focus of the study, plan methods for investigation and discuss recruitment procedures, as outlined above. They will meet again at the end of the study to have an input both in the interpretation of data and in the dissemination strategy. It is anticipated that the ‘reference group’ will be involved in disseminating the results of the study locally using their links with their own cancer research and support groups.

Volunteers have been sought from the ‘reference group’ to establish a patient and carer co‐researcher group. Ten patients and carers volunteered and subsequently this has been increased to 15 co‐researchers to include ‘hard to reach’ groups (e.g. palliative care and ethnic minoritized backgrounds).

Heron and Reason identified two types of inquiry for ‘participatory research groups’: ²³

• • an ‘open boundary inquiry’ (using the group to support and direct research with participants outside the group);
• • a ‘closed boundary inquiry’ (concerning only what goes on within the group).

‘Open boundary participatory research’ has tended to be concerned with community‐based action research. Often this has involved inventive ways of engaging with the local community. The Bagamoyo College of Arts project involved participants working with local communities using popular theatre in their work with HIV and AIDS. ²⁸ However, in other projects, the establishment of a ‘participatory group’ is sufficient in itself to collect necessary information and data for a study. These ‘closed’ projects involve participants generating research information by allowing them to share experiences and discuss issues though a series of regular meetings. ²⁴ , ²⁹

For our study, it was decided that the patient and carer co‐researcher group should adopt an ‘open boundary’ approach. In this regard, the patient and carer co‐researcher group assisted in the design of the project and in the collection and subsequent analysis of patients and carers recruited throughout the UK. However, the group themselves did not provide any research data for the study.

Stages of involvement – co‐operative inquiry cycles

‘Participatory research’ often draws upon ‘co‐operative inquiry cycles’ to ensure a process of reflection and assessment is sustained during a project, and these cycles have been adapted to inform work with the ‘reference group’ and the patient and carer co‐researcher group. ²³ Typically, each cycle has four key phases. ²³

• • Phase 1: ‘participatory research group’ participants meet to explore the focus of the study, to develop questions, to plan methods for exploring the questions and to agree procedures for gathering data.
• • Phase 2: ‘participatory research group’ participants become co‐subjects, i.e. they take part in the inquiry that is being undertaken.
• • Phase 3: ‘participatory research group’ participants become immersed in the action and experience of the research, and subsequently develop new insights into the study.
• • Phase 4: ‘participatory research group’ participants meet to share and reflect on the data. This will inform the further direction of the study. This cycle of action and reflection can continue throughout the study to increase the validity of the findings.

These cycles have influenced the structure of the ‘reference group’ and the patient and carer co‐research meetings. In particular, the initial meeting of the ‘reference group’ explored the focus of the study and discussed the methods to be used. The ‘reference group’ served as the basis for identifying patient and carer co‐researchers and these co‐researchers have received training and support throughout the study to undertake data collection and analysis with the Macmillan Research Unit researchers. The initial training provided a background to general research methods, specific training on focus group approaches and a discussion on the focus group question schedule. Early meetings also involved ‘mock focus groups’ where patients and carer co‐researchers took it in turns to be moderators and observers. These meetings were designed to provide an essential background to the project and the methods used, as well as establishing confidence amongst the co‐researchers in co‐moderating the focus groups.

The patient and carers have continued to meet during the course of the study. This was important because, given the large number of co‐researchers and sites involved, not all co‐researchers could be involved in data collection at the same time. Hence meetings were organized such that those co‐researchers who had already undertaken data collection could feed back their experiences to those who had yet to moderate focus groups. These meetings included listening to recordings of focus groups and reading transcripts to discuss issues arising. This was important as it maintained the involvement and confidence of all co‐researchers, as well as providing an opportunity to reflect on the data collection already completed.

Patient and carer co‐researchers are subsequently analysing the data from the focus groups. Two principal data sets are being created from the focus groups: tables of research priorities arising from the focus groups and transcripts from the discussion. Both data sets are being subjected to thematic analysis and patient and carer co‐researchers have opted for which data set to analyse after receipt of an example from the lead researcher. Their analyses are then compared with those of the lead researcher.

The approach undertaken to involve patients and carers has been effective in allowing the co‐researchers to generate greater confidence and understanding about the study through learning from experience whilst at the same time ensuring the data collection is responsive to the findings of the study as it progresses. The patient and carer co‐researcher group also permits greater refinement of the focus group method and the question schedule used.

Involving patient and carer co‐researchers as collaborators with the research team at the Macmillan Research Unit is a highly innovative and complicated aspect of the study. Running focus groups successfully is a particularly challenging task, requiring significant levels of training and thus sufficient resources to support this. However, the collaborative nature of the data collection is such that patient and carer co‐researchers are not left on their own to moderate focus groups, but are supported by researchers from the Macmillan Research Unit during the data collection. This has resulted to date in successful focus groups for the following reasons:

• • The patient and carer co‐researchers undermine the distinction between researcher and patient. Consequently, participants can feel more at ease, generating better discussion. For example, participants have expressed relief and greater comfort when they find other patients involved in the group on entering the room in which the focus groups are held.
• • The patient and carer co‐researchers often draw upon their own experiences to be particularly responsive in prompting participants and following up issues. In particular, the patient and carer co‐researchers have been able to draw upon their own experiences of the disease, the treatment and side effects to stimulate and guide the discussion effectively.
• • The patient and carer co‐researchers with the Macmillan research team can use their experience to remain aware of the needs of participants (e.g. if a participant becomes fatigued), thus enhancing the ethical nature of the study.
• • Given that the groups are co‐moderated by patients and carers, participants can feel more at ease in discussing a range of research interests rather than focusing on what they perceive to be the interests of ‘experts’.
• • Participants have commented on the effectiveness of the focus group design, the nature of the questions and the means by which research issues are generated and prioritized. These comments are supported by questionnaires given to participants after each focus group which has generated feedback that also supports the effectiveness with which the co‐researchers have moderated the discussion.

Issues relating to the formation of a reference group and a patient and carer co‐researcher group

The establishment of the reference and co‐researcher groups has generated challenges. In particular, both groups require significant resources both in time and money. The logistics of organizing patients and carers across the UK to attend meetings are considerable. Furthermore, attention has to be paid not just to the financial costs of travel but also to hotel accommodation for those who have to travel far.

There were also ethical considerations in forming the groups. Initial discussions with the Central Office for Research Ethics Committees (COREC) suggested that ethics approval was required to establish the groups. This was subsequently rejected when it became apparent that members of the groups were volunteers and that their input was to advise on the study and not to provide data for analysis. However, ethical principles were adopted in the study (for example, co‐researchers must keep information shared during the focus groups confidential, both co‐researchers and ‘reference group’ members are able to leave the study at any time without having to give a reason).

Further difficulties arose with research governance, especially with some sites requiring patient and carer co‐researchers to have honorary contracts in order to co‐moderate or observe the focus groups with participants recruited from that site. This was a particular challenge as the honorary contract process in some sites required occupational health clearance.

In addition, it was recognized that patient and carer co‐researchers may find issues raised in the focus groups emotionally challenging. Consequently, a meeting was arranged independently from the Macmillan research team to provide peer support and supervision. One‐to‐one peer supervision was then offered to participants to discuss personal issues arising from participating in the study as co‐researchers. This peer support is not specific to patient and carer co‐researchers but is seen as general good research practice.

Despite these issues, the effort required to establish the groups has been worthwhile in facilitating the design and conduct of our study.

Discussion and conclusion

This paper details the relatively limited levels of patient involvement that currently exist in cancer research, particularly in terms of patients and carers determining research priorities. Our study is seeking to rectify this through a national research prioritization exercise. The study has adopted the principles of ‘participatory research’ to ensure that the patients and carers are collaborators in the research. ²³

Involving patients and carers in this way has been criticized, principally because it is often felt that patients are difficult to identify and involve, that patient involvement groups are unrepresentative, and that patients become ‘professionalized’ through their involvement in research. ⁸ Experiences from our study reject these assertions. Identifying participants through the patient forums of Cancer Networks has proved to be relatively straightforward and the high response rate suggests that patients and carers are interested in being involved. Careful and appropriate strategies for identifying participants can be used to ensure that certain ‘hard to reach’ groups are involved. ³ Finally, whilst it is true that the patient and carer co‐researchers in the study become ‘professionalized’ to a certain extent through their training and fieldwork experience, this does not negate their current or previous patient or carer status, particularly in relation to academic teams with comparatively less personal experience.

Our study has found that ‘participatory research’ approaches offer an effective means of involving patients and carers in cancer research. Through the establishment of a ‘reference group’ and a subsequent patient and carer co‐researcher group, it has been possible to design and conduct the study in collaboration with patients and carers, drawing upon their own unique experiences and expertise. Adopting a ‘participatory approach’ enhances the appropriateness of research findings and methods, as well as responding to the imperative of patient involvement in health research.