Primary care research: ends and means

In the UK, research in primary care has become a priority for support within the National Health Service (NHS) and other funding agencies, such as the Medical Research Council (MRC). This has resulted in the creation of a range of interventions aimed at contributing to research activity, capacity and utilization in primary care. It is now timely to reflect on the appropriateness and effectiveness of these interventions and to consider, for the future, what are the desired outcomes (ends) of primary care research and how we can best achieve them (means). Primary care consists of a sector of the health service which provides first-contact care for patients. Typically, primary care has a number of dimensions, including those of personal care, continuity or longitudinality and comprehensiveness.1 In general practice, this is reflected by the Stott and Davies'2 model of the consultation, in which as well as dealing with the presenting complaint, on-going medical, psychological and social problems are addressed, health care-seeking behaviour examined and opportunities for prevention and health promotion taken. We should now add to these facets of primary care its important public health function, in which responsibility for individuals is extended to responsibility, planning and service provision for populations and communities. The medical geography of primary care includes general practice, primary care dentistry and optometry, accident and emergency departments, community-based services, such as family planning and well-person clinics, and direct-access, personal and telephone advice services such as NHS Direct and walk-in clinics at railway termini. This landscape is populated by GPs and primary health care teams, dentists, optometrists, and a range of professions allied to medicine, including physical therapists of various kinds, some hospital staff and medical professionals providing telephone advice supported by computerized decision systems.3 The starting point of a consideration of the desired outputs and outcomes of research in primary care has to be the motivation to improve the quality, effectiveness and cost effectiveness of primary care, in all its manifestations. Where service provision is weak, research needs to find answers to questions about how to strengthen it. Where there is wide variation in standards, performance and professional behaviour, research questions need to address the reasons for these variations and ways of ameliorating them. Where the provision of primary care services lacks evidence, research is required to provide the evidence base. Research questions need, therefore, to spring from the realities of providing primary care services, and the outputs of primary care research need, in parallel, to make a contribution to the practice of high quality, clinically effective and cost-effective care. In the early days of research in primary care, individual GPs began to map the structure and processes of primary care and to describe the natural history of the common disorders seen in general practice, in contrast to those seen in the hospital setting.4,5 As the research infrastructure became more sophisticated, so the research questions became more complex and new methodologies developed to answer them. In the 1990s, a number of significant reports and developments within and outside the NHS have focused on the importance of primary care. The Culyer report recommended new mechanisms for providing support for research within the NHS6 and attempted to define a funding stream that would be applicable to primary care as well as to hospital-based research although the Culyer proposals fell short of an effective way of improving capacity within primary care research.7 The national NHS Research and Development (R&D) programme, established by Sir Michael Peckham,8 recognized the importance of community-based research and included a large funded programme on the primary–secondary care interface.9 The Mant Report on R&D in the NHS10 and the MRC Topic Review of Research in Primary Care11 made more specific recommendations on the content and conduct of primary care research, and have led to substantial (although in absolute financial terms relatively small) initiatives to provide research project and programme funding and training. The Royal College of General Practitioners (RCGP) is working hard to establish a method of accrediting research practices,12 and general practice research networks have sprung up in the UK and elsewhere,13–15 and offer one means of ensuring a degree of support for primary care researchers. The 1990s have also been important in terms of identifying the gaps between doing research and getting it into practice, and a large volume of literature has been generated about the problems of implementing the findings of research and evaluations of interventions aimed at putting research findings into everyday practice, as well as persuading practitioners to abandon outmoded and ineffective practice.16,17 In the Mant Report, careful consideration was given to the differences between research in primary care, research by primary care and research on primary care. In other words, some research was most appropriately undertaken by people working within the primary care sector, with other questions likely to be better answered by other researchers, such as health service research units and epidemiologists. These dimensions of primary care and their relationship to research topics and methodologies are laid out in Table 1. Thus, although GPs working within primary care groups (PCGs) are well placed to research factors affecting patients' health-seeking behaviour and clinical decision making in the consultations, the development of clinical guidelines at the primary care–secondary care interface is more likely to involve secondary care colleagues, while the epidemiology and natural history of common disorders is likely to be better studied within larger groups of practitioners working in collaboration with epidemiologists/public health science. Table 1 also gives an indication of the range of research questions relevant to primary care and the corresponding range of methodologies and disciplines likely to be required to answer them; most research in primary care has to be multi-disciplinary if it is to tackle the relevant research questions effectively. It might be helpful to think of research output as a therapeutic agent which has a complex interaction with a range of potential research ‘receptors’. These receptors include individual clinicians and other practitioners, primary health care teams and practices, primary care groups and Trusts, research centres such as university departments of general practice, health authorities, and regional and national policy makers, including the NHS Executive, the National Institute for Clinical Excellence (NICE) and the National Electronic Library for Health. The delivery systems which take research to its receptors include peer-reviewed primary research studies, systematic reviews and meta-analyses, guidelines and protocols, continuing medical education, academic detailing and promotion, advertising and the media, informal contact with colleagues, formal continuing medical education and continuing professional development and audit. This issue of Family Practice is accompanied by a supplement dedicated to methodological issues in research into the implementation of guidelines. It follows that there are a range of opportunities to get different kinds of research to different potential research users, and that increasing research activity in primary care per se is by no means the only approach to strengthening the research base and improving clinical effectiveness in primary care. Indeed, an old adage from the RCGP—“every practice a research practice”—now seems outdated. In the course of setting up research networks in the north of England (NoReN), Hampshire (WREN) and London (STaRNet), around 10% of GPs declared an interest in research, although only 1 or 2% were research active. While there may be a case for building up research capacity from this level, it is also equally, and possibly more, important to consider other interventions likely to facilitate the uptake of research evidence and clinical effectiveness in general practice. Work on the implementation of research and guidelines/ guidance has indicated, time and time again, that single interventions are unlikely to change professional behaviour and that multiple interventions, tailored to the medical ‘community’ and the research topic, are most likely to succeed.17–20 Capacity-building interventions are, thankfully, now underway. The MRC/DH initiative has provided much needed project and programme support and encouraged collaboration. The training fellowships and clinical scientist fellowships will do much to increase research capacity, and regionally based health service research and primary care training fellowships add to this welcome development, particularly when there is an emphasis on multi-disciplinary work and the development of non-clinical research careers in primary care. Networks clearly provide a mechanism to bring researchers together, to share interests and expertise and to develop capacity further; their continued linkage to existing research centres, where they can access expertise, supervision and guidance on the conduct of research, is likely to be essential. Beyond this, however, the broader picture of research in primary care needs to take into account the challenges and opportunities for disseminating and implementing research, employing a combination of the methodologies mentioned above. This will be a particular challenge for primary care groups and Trusts in the future; the links between NICE, clinical governance and clinical practice within PCGs will have to be made with care, and further evaluative work is needed to determine the most fruitful interactions between them. A number of implications follow from this discussion, against the background of an intended doubling of primary care R&D spend within the NHS. The project/ programme support and training/capacity initiatives over the last 2 years are particularly welcome, and to some extent the importance of providing increased support for primary care research at regional level through research project grants seems to have been accepted and acted on. Research support funding, to cover infrastructural and health service costs, needs to be made more accessible to existing research centres, and consideration might even be given to the creation of a single R&D funding stream. Primary care networks are at a critical point in their development, and their efficacy in terms of meeting their stated aims and objectives about doing and implementing research are currently under review. Of more importance at present may be the imperative to ensure that research generated in, by and on primary care is digested, synthesized and, literally, put into practice where it is needed. We know enough about the implementation of research evidence and guidelines to make real progress in bringing evidence and practice closer together. This will require an imaginative use of resources currently earmarked separately for R&D and for educational purposes; intimate connections exist between research, dissemination and education which should be reflected in a funding system which recognizes the need to link research evidence with continuing professional development. Dimensions of primary care and primary care research Dimensions of primary care and primary care research