Professor of Primary Care and Co-Director, WA Centre for Cancer and Palliative Care, Curtin University of Technology, Perth, Australia
Those who are the architects of future primary care, namely researchers, are often funded to work with policy makers to support the treasury’s desire to contain the cost of healthcare. Bearing in mind that the bulk of health and research expenditure is in secondary care we should be asking more questions. Barbara Starfield has long championed the case for increasing investment in primary care, and to an extent this message seems to have been heeded in the UK in what is now touted as a primary care-led NHS. However, published research, even when clearly led in primary care, offers evidence for the policy tail wag-ging the academic dog.
Mant and colleagues writing in the BMJ last year reported that hypertensive patients attending general practitioners (GPs) are older than those recruited to clinical trials, and proposed, by implication, that this is the reason why stroke reduction guidelines were not implemented in clinical practice. By corollary the publication of yet more evidence-based guidelines would be more enthusiastically implemented. If the paper was interesting then the rapid response was fascinating. The majority of commentary was sub-mitted by hospital specialists. One pointed out that:
treatment is decided following discussion between patient and doctor, with the aid of the information provided by randomised-controlled trials.
Another suggested that
we should not be ageist, but rather take a more belligerent approach in trying to tackle the major public health challenge.
And the last respondent who declared a conflict of interest as someone involved in marketing anti-hypertensive drugs promoted the need for ‘a pooled analysis’ of data from other clinical trials. Therefore the reason why the guidelines were not implemented it seems was either a failure to publish the appropriate guidelines, a failure to inform the patient with refer-ence to the trials, an ageist attitude, or a lack of statistical effort. One might conclude that the solution to reducing the incidence of stroke in the community is something to do with clinical trials, guidelines and the failure to impress practitioners with the need to promote the some-would-say flawed research – in other words yet more investment in more of the same.
Analysis of practitioner compliance with guidelines and other benchmarks is a routine, one that is fre-quently published in quality of care journals, each prescribing yet another ‘fix-it’, often without any Reference to those to whom the statistics apply. Edu-cate, incentivise or penalise are offered as ways to steer those who are the first, and, in most cases, the only doctors the patient will consult with a healthcare issue. In reality, patients choose to present their problems or symptoms amidst a plethora of other needs and wants. The most appropriate research in most cases is not the randomised controlled trial because most sample frames also trawl a raft of confounding variables. Primary care is not a corner shop, failing only for want of customer service. Many patients enter healthcare bewildered, confused, upset, angry, despondent, fright-ened, needy, ignorant or misled. The port of entry is almost always primary care, and the truth is that we don’t fully understand patients’ health-seeking behav-iour nor why primary care succeeds in one case but not another.
Shedding light on the issues, Glasziou and Haynes offer a model that affords more sophistication in the attempt to understand and influence the implemen-tation of guidelines or uptake of research findings. They identify the multitude of factors impeding the route from publication to practice. They conclude that the delivery of evidence-based medicine should be concerned with clinical content but also with the processes of changing care and the systems of care. These issues have profound implications for research and publication in primary care. Studies of patients in a defined context, examining relevant regional issues, are often best published in the local language and in local journals, where the message has regional import-ance. This is entirely appropriate, however it sits uncomfortably in academic institutions bound to reward staff who publish in ‘high-impact’ journals, if only because that is how they themselves can expect to flourish.
That leaves us defending work that seeks answers to the difficult questions. The reasons why primary care might fail its patients is not simply a function of research awareness, the quality of published data or the lack of randomised controlled trials. It probably reflects the messy business that is inherent in the role. The patient journey in any healthcare system begins with undifferentiated symptoms. In primary care, consultations can be subject to noise and red herrings, interruptions and flawed processes. These need to be factored into data analysis, and more often than not are context specific. The output of the best of our research, if it is to stimulate real improvements in patient care, should be interpreted with a profound understanding of the paradigm and take account of the medicine man’s role in that society. This is a good time to be a researcher in primary care, a time when policy makers are wedded to an exploration of clinical performance, equity, access and patient safety. Re-searchers are in the business of designing healthcare for the future. An exciting journey lies ahead for those with the good fortune to be working in an enlightened and enabling environment.