Service users in research and a ‘well ordered science’

Service user involvement in research is little over a decade old. Although there is no body of supporting evidence thus far, it is highly likely that research has much to gain by involving service users. Among the benefits are: increasing relevance of the research (for example, by enriching researchers' understanding of the illness, ensuring the questions being asked are meaningful, improving the design of a study, choosing appropriate outcome measures, and generally preserving a focus on the meanings of the research for those with the illness); better recruitment to studies and better, more open responses from research participants who are more likely to feel their interests are being addressed, and with less likelihood of dropouts; fresh insights in interpreting the results; service user support may assist dissemination and implementation of research findings; and more ethically sound research (for example, by defining acceptable limits of controversial research, and providing better explanations when seeking consent). (Staley & Minogue, 2006).

Different levels of user involvement in research have been defined. INVOLVE, an organisation funded by the Department of Health to support patient and public engagement in research, describes three levels: consultation, collaboration and user-controlled research. Successful examples of each have been described (see, for example, two recent texts on service user involvement in mental health research – Wallcraft, Schrank, & Amering, 2009 and Rose, Sweeney & Beresford 2009). It is clear that service user led research, in particular, can be powerfully emancipating for its participants.

However, serious obstacles to service user involvement in research are encountered (for example, a reluctance on the part of researchers who may fear losing control of the research, or a reduction in its quality; and, inadequate resources available to train and support service user researchers in necessary ways (MHRN Good Practice Guidelines).

As service user involvement in research is still relatively new, ideas about its scope and how it might be best realised are still evolving. I intend to take the opportunity presented by this editorial to suggest that current ideas of service user involvement in research may not be ambitious enough in their scope.

‘Well-ordered science’

I take as my starting point Philip Kitcher's analysis of ‘what science should be done?’, and his proposals for what he terms is a ‘well-ordered science’ (Kitcher, 2001). Kitcher, a philosopher of science at Columbia New York, presents a set of arguments that are very persuasive.

Kitcher defines research as acquiring truths worth knowing, for their practical significance, or because they are epistemically significant. He argues that the research agenda is not set by the ‘nature of things in the world’; it does not simply reflect a logical progression determined by the structure of the world and the research findings up to that time. The idea that what should be researched is somehow intrinsic to the process of research is rejected. What counts as ‘significant’ in the definition of research, above, depends on us, ‘on our cognitive capacities and on the interests we happen to have’. It cannot be divorced from human interests. Our multiplicity of sciences (here I include the ‘human’ and ‘social sciences’) reflects the fact that we have developed different lines of inquiry about phenomena that are salient for us and answer to our interests. The agenda is set by people; it depends on the ways our curiosity is aroused, on our history and culture. But people may have particular and sometimes conflicting, interests.

Having established that human interests determine what is significant, Kitcher goes on to elaborate his concept of a ‘well-ordered’ science. Its key aspect is how a research ‘agenda’ is therefore to be determined. Conventionally, what is to be researched, and how, has been determined by ‘elites’– mostly by communities of scientists engaged in a particular field, but increasingly by scientists in association with privileged groups of outsiders, predominantly the funders of research. Some funders, such as government departments or research councils are seen as representing society, though rarely is that representation wide; others, such as charities represent narrower interest groups, often in respect of a single disease or a narrow range of diseases. Another paymaster, industry, has a necessary interest in making a profit.

Kitcher argues that elitism is an unacceptable basis for setting a research agenda. If a community of scientists believes that the nature of the world determines what the next investigation should be, they are ignoring the critical role of interests. Science has the potential to improve the lives of many, and their voices need to be heard. However, a simple voting scheme cannot be supported. Such ‘vulgar democracy’ as a means for setting the agenda, Kitcher warns, would lead to a ‘tyranny of the ignorant’. The results will be research topics based on what it currently fashionable, or ‘hot topics’, usually with a short-term, and not infrequently, a political focus. Kitcher proposes that the proper basis for a ‘well-ordered’ science is ‘enlightened democracy’. Here decisions would be made on the basis of what he terms ‘tutored preferences’ (as distinct from the ‘raw preferences’ engaged in a ‘vulgar democracy’). Interest groups would be ‘tutored’ by scientists about what is known so far, how we know it, and the prospects for knowing more about a particular problem. They would be exposed to the conflicts within science (including the disagreements between scientists) as well as its successes. Non-scientist deliberators will then understand the goals, methods and limitations of a particular research approach as a means for solving a particular problem. They would thus become deliberators with ‘tutored preferences’ who can then engage in ‘ideal deliberations’ with the scientists, funders, government and each other about what research should be pursued for the common good.

But deliberations must involve a two-way ‘tutoring’ process - non-scientist deliberators also tutor the scientists about the personal meanings, social value and political implications that ‘their’ knowledge may carry. Tutored preferences will thus lead to a science which is ‘well-ordered’– that is with an agenda constructed by informed deliberators who will take account of competing interests, and who will agree on the problems to be researched, the kinds of science required to solve them, and the constraints that need to be imposed (such as ethical and budgetary ones). One would also expect that a ‘well ordered science’ is likely to produce high quality science – it embodies the full range of relevant expertise, set in a context that has been thoroughly explored, with methods whose validity has been challenged during the deliberation phase.

This is clearly a description of an ideal and is regarded as such by Kitcher.

However, I believe that in the recent history of service user involvement in mental health research, we can see processes and practices that move us in that direction – a modest achievement, perhaps, in terms of the grand ideal, but not an insignificant one. Of course one can point to the obstacles – the difficulties in hearing or understanding each other's viewpoints – for service users the science can be hugely difficult to understand, as can be the convoluted and competitive workings of the academic world – while for the scientist, the lived experience of mental illness in all of its complexity and implications, can be difficult to grasp or even acknowledge, and it is easy to become complicit in the exclusion of the voices of a stigmatised group.

I would argue further, that progress towards a ‘well-ordered science’ in mental health is the most important in the medical or human sciences. This is because ‘values’ are so much more clearly at issue than, say, in cardiovascular disease or diabetes. So much is contested in mental health – even the idea of ‘illness’ itself, and thus of the value of entire domains of science as applied to the problems of mental health. If progress can be made in establishing ‘tutored preferences’ for all parties in this arena, and ideal deliberations, then it surely can be achieved in any area of health research. Most would also agree that mental health service users are, of all patient groups, the one with most to gain from the emancipatory implications of a ‘well-ordered science’.

I might mention at this point, although it is not directly relevant to the main argument, the loss from academia of a very significant group – people whose careers are brought to an end or seriously set back by being unfairly treated because they have had a mental illness. Higher Education Institutions need to review their policies on staff illness and to ensure that discrimination against those with a mental illness does not occur.

NIHR Mental Health Research Network and Service Users in Research

I believe the NIHR Mental Health Research Network (MHRN), of which I am an associate director, offers an important, practical opportunity to establish something approaching a ‘well ordered science’, albeit within the specific domain of mental health. Established in 2003, the role of the MHRN is to create an infrastructure for England that will facilitate high quality research, large enough so that findings are generalisable across communities in England. It is one of eight research networks set up by the Department of Health, and together with Cancer, the oldest. Through its Coordinating Centre, eight ‘hubs’, and close working relationships with the Comprehensive Clinical Research Network, it covers all of England. There are currently more than 160 projects on the network. Service user involvement in all aspects of the research enterprise – formulating priorities, research questions, design and methods, outcome measures, analysis and interpretation of results, dissemination – is an essential component of MHRN's stated mission. The Service Users in Research (formerly known as SURGE) is the medium through which MHRN seeks to achieve this goal. Service Users in Research comprises a central strategic and coordinating group and representatives from each of the ‘hubs’.

User-led or user-controlled research is difficult to achieve on the MHRN; the vast majority of projects are large and are carried out by multidisciplinary teams. But we aim for all to have significant service user involvement. In a well-ordered science, who should do the research is dependent on the conclusions of deliberations about who is best placed to do it. Furthermore, if the theoretical basis for a well-ordered science, as expounded by Kitcher, is accepted, then service user involvement is an essential ‘value’; an evidence base for its effectiveness is irrelevant.

The MHRN does not directly fund research; however it can significantly influence the research agenda. Through its structure and processes it can set examples and standards for service user (and carer) engagement in research. Through some of these means it can facilitate tutored preferences and work towards deliberations of the type envisaged in a well-ordered science. The MHRN has made a start in ascertaining service user priorities for research (MHRN 2007); requires clear and appropriate service user involvement plan for all studies to be adopted on the network; it facilitates the provision of a range of opportunities for service users on research steering groups and management committees; importantly, it supports service user engagement in Clinical Research Groups – these are groups of experts in a particular field of mental health research funded for two years by the Network to deliberate about projects that should be submitted for funding. There have been about 30 CRGs so far, and they have increasingly strong service user involvement. A recently established CRG has as its objective, the development of projects aimed at enhancing service user involvement in research. The MHRN has a Service Users in Research External Advisory Board comprising service users with relevant expertise (for example, in NHS management, user-led research, law, industry, the disability sector) drawn from the wider community. MHRN is also setting up a Service Users in Research Forum, which will offer opportunities for the MHRN and service user organisations to engage in discussion about all aspects of mental health research.

These initiatives have certainly had an impact on mental health research carried out on the network. I believe that some of the models of engagement adopted by the network will inform practice elsewhere. The research networks are co-ordinated for the Department of Health, National Institute for Health Research by the Clinical Research Network Co-ordinating Centre (CRN-CC). MHRN contributes substantially to the patient and public involvement mission of the CRN-CC, its influence thus extending beyond mental health research. MHRN is also often invited to contribute to decision-making in medical and healthcare research strategy nationally, where its models of service user involvement may be discussed.

Of course this is only a start. However, the direction of travel is one that is consistent with the idea of a well-ordered science. There is a long way to go and there are many tensions. But the MHRN has a structure in place, supported by modest but dedicated resources, which offers a unique opportunity for the realisation of an ambitious programme of service user involvement in research, with a potential influence extending far beyond anything previously achieved.

Appendix

MHRN: http://www.mhrn.info/index/ppi/service-user-involvement.html

INVOLVE: http://www.invo.org.uk/