Intended for healthcare professionals


Health checks in general practice

BMJ 1995; 310 doi: (Published 29 April 1995) Cite this as: BMJ 1995;310:1083
  1. Peter D Toon
  1. General practitioner London E8 3NB

    Time to review their role

    The NHS has changed a lot since the late Denis Burkitt (who linked high fibre diets with preventing bowel disease) compared illness to an overflowing bath and suggested that doctors and nurses might be better employed turning off taps than mopping the floor. Increasingly, general practice has been considered to be the right place for turning off taps,1 2 and in 1990 health promotion was made a contractual requirement for general practitioners.3 Does prevention of illness in primary care work? Three papers in this week's journal consider different aspects of that question.4 5 6 Unsurprisingly, the answer is not straightforward.

    One group based in Oxford, testing strategies to reduce the risks of cardiovascular disease and cancer in general practice populations, has greatly influenced British health policy. It is to the OXCHECK researchers' credit, therefore, that evangelistic fervour has not clouded their evaluative judgment. In their latest report they conclude that health checks for unselected middle aged people achieve little or no reduction in smoking and excessive drinking (p 1099).4 The most encouraging results were change in lifestyle—better diets, increased exercise—and falls in serum cholesterol concentrations. These changes were sustained over three years.

    Crucial issues in health promotion are knowing how best to intervene and whom to target. Lindholm and colleagues, reporting from Sweden on middle aged people with at least two cardiovascular risk factors, found intensive group education about the risks of cardiovascular disease to be more effective than standard advice in reducing total serum cholesterol concentrations (p 1105).5 The differences, however, are small, and many trends do not reach significance. This underlines the need to enrol large numbers of patients if such studies are to give unequivocal results.

    Perhaps most interesting for British general practice is Field and colleagues' modelling of the cost effectiveness of different strategies (p 1109),6 using data from the OXCHECK trial.4 They predict that the most cost effective way to reduce the risk of coronary heart disease in general practice populations would be to target high risk groups.

    One must beware of reading too much into these results. The cost effectiveness study had to make several assumptions, not necessarily all valid. In addition, the success of skilled procedures often varies considerably among operators. This may be as true for the interpersonal skills used in promoting behavioural change as for manual skills in surgery. Reluctant practices doing checks solely to receive payments for health promotion or to fulfil their terms of service may do far worse than the highly motivated and specially trained practices in Bedford that took part in the OXCHECK study. Populations also vary considerably. Prevention is a low priority in some groups, often for very good reasons.7 Generalising from Sweden to Britain, or even from Bedford to Hackney, may not be possible.

    Despite these caveats the papers have some important messages. They reinforce doubts about the current system of payments for health promotion in British general practice, which requires doctors to record risk factors but places little emphasis on effective intervention or on targeting. Policy on health promotion should be based not only on the recognition of risk factors but also on evidence that they can be changed and that changing them improves health.

    Better health promotion will depend on more research to determine which targeting and intervention strategies work best. To yield clear results, studies will have to be large, not least because the unit of analysis for intervention methods is the doctor or nurse, not the patient. Enough doctors and nurses have to see enough patients if researchers are to distinguish variation among techniques from variation among individuals. This will be expensive but much less costly than pursuing a flawed strategy.

    We also need to reconsider the role of targeting intervention at individuals. Other strategies may deserve more attention. These could include group education in schools and workplaces and through the media; and government action such as changing fiscal and agricultural policy, introducing stricter rules on tobacco advertising and food labelling, and tackling the social factors often associated with unhealthy lifestyles. Research techniques that allow us to compare these different approaches need developing.

    The most important aspect of this debate, however, is not research but priorities. The key tasks of general practice are helping patients to understand and cope with illness, relieving symptoms, and offering the occasional cure. They have been overshadowed recently by prevention and purchasing—activities that may improve or complement care for those who are or believe themselves to be ill8 but must never be allowed to displace it. This is important for patients as well as doctors. Failure of policymakers to recognise these priorities is an important cause of the present low state of morale among general practitioners.

    A prevention strategy that encourages interventions of proved efficacy among those most likely to benefit will be much more compatible with the core role of general practice than massive indiscriminate collection of data of no proved value. Without this focused approach prevention may lose credibility. The NHS may conclude that Burkitt's taps are stuck open irrevocably and again become preoccupied with developing better mops.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    View Abstract