Intended for healthcare professionals


Prescribing exercise in general practice

BMJ 1994; 309 doi: (Published 20 August 1994) Cite this as: BMJ 1994;309:494
  1. S Iliffe,
  2. S S Tai,
  3. M Gould,
  4. M Thorogood,
  5. M Hillsdon

    Exercise is good for us, especially as we get older.1, 2 People who are regularly active enjoy a lower risk of osteoporosis and a much reduced risk of coronary heart disease (the vigorously active have about half the risk of the inactive). Most people in Britain are not taking enough exercise to achieve these health benefits.3, 4

    If primary health care teams could increase their practice populations' physical activity then many benefits might follow, including lower rates of cardiovascular disease, fewer fractured hips, reduced depression and anxiety, and improved functional ability in elderly people. Although this list makes promotion of healthy exercise look tempting, primary health care teams, and the policymakers who determine allocation of resources, should look carefully before they leap.

    Very little is known about the effectiveness of exercise programmes and other initiatives designed to increase non-athletes' activity levels. The question of effectiveness should be answered by randomised controlled trials, although these are difficult to conduct in the community.5 After an extensive search we identified only 12 such trials,* none of them conducted in Britain, with few including those subjects with most to gain from increased exercise - middle aged and elderly people.

    Moreover, the results of these trials are not particularly encouraging. The increase in activity is not great and, in so far as a sustained increase is observed, subjects seemed to prefer home based, self monitored exercise programmes rather than the programmes carried out at sports facilities that are currently popular “on prescription.” In this discussion we should not forget that physical fitness and health status are interrelated but not synonymous, and improving fitness may not reduce the consequences of disease.6

    Thus before investing time and resources in prescribing exercise general practitioners need to answer four questions. Can people who do not exercise be recruited to an exercise programme by their doctors? Does a short term exercise programme tailored to individual needs initiate long term changes in exercise behaviour? If so, among whom will such changes be most effectively maintained? Does investing resources in the promotion of healthy exercise in this way affect morbidity or the use of health services? As with many other initiatives promoting health, there is a danger that effort and resources may be misspent in promoting exercise to those who would have taken it up anyway, the “worried well.” This group is likely to be younger and already more fit and active than average.

    Many barriers exist to increasing physical activity, particularly through formal exercise programmes, which may be perceived as competitive and intimidating.7 Other obstacles to participation include boredom with exercises, lack of knowledge of how to get involved in exercise programmes, difficulties with transport to exercise sessions, lack of self discipline, and concerns about exercise exacerbating existing medical problems.1

    Developing an exercise programme for those most likely to benefit requires further study of the factors that influence uptake and adherence and of how to tailor programmes to match people's changing needs. Work on brief interventions in medical settings suggests that practitioners should use a menu of strategies, selecting one that matches the patient's readiness to change. “Prescribing” behavioural change may lead to resistance; giving patients more freedom of choice is likely to lead to better outcomes.8 A leisure centre should not be the only choice on offer.

    Biddle and Mutrie have identified four problems that need to be addressed when exercise programmes are designed: getting started; keeping going with regular exercise; avoiding the start-stop syndrome; and improving knowledge about fitness.9 General practitioners or practice nurses may use these as starting points to construct and evaluate exercise programmes. We need to learn more about the successes and failures of pioneering projects; anecdotal data will not do. Any future prescription for exercise programmes should be carefully evaluated; the results will help in the design of a definitive multicentre trial. Unevaluated initiatives may be of no more value than prescribing coloured water.

    While we await the results of careful evaluation, primary health care teams should look closely before they leap into prescribing exercise. There may be many far more effective ways for them to use their resources to increase the fitness of their practice populations.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    View Abstract