Intended for healthcare professionals


“Variations” in health

BMJ 1995; 311 doi: (Published 04 November 1995) Cite this as: BMJ 1995;311:1177
  1. Richard G Wilkinson
  1. Senior research fellow Trafford Centre for Medical Research, University of Sussex, Brighton BN1 9RY

    The costs of government timidity

    From time to time researchers working on health inequalities listen with envy to stories from distant lands about open, all party talks organised by governments wanting to know what can be done to reduce inequalities in health. Last year the British government took its courage in both hands and unbarred the door to admit a small deputation of experts to talk about reducing the health divide. Political safety was ensured by limiting the discussion to what the NHS and the Department of Health could do. Poverty, housing, job insecurity, the inner cities, the rationing of health services, and other embarrassing problems were left outside, and the experts agreed to adopt their hosts' view of political correctness by referring to health “variations” rather than “inequalities.”

    Their report, published this week, is a welcome opening of negotiations.1 Although the evidence suggests that health services are not major contributors to health inequalities, the report's recommendations are worth while. As well as saying that health authorities should monitor health variations, target resources, ensure equal access, and evaluate interventions, the report also says a little (not enough) about the responsibilities of the NHS as the country's largest employer and--most crucially--emphasises the Department of Health's responsibility for informing the government of the impact of other aspects of policy on health.

    As well as influencing the content of this report, however, the political constraints risk starting the discussion off on the wrong foot. There are cheap and expensive ways of tackling health inequalities, and the expensive ones are unlikely to be the best. What is expensive is to leave the underlying causes intact while establishing new services for those “at risk” in an attempt to repair continuing damage. Take, for example, the high rates of suicide among adolescents in Japan. They might have been tackled by identifying those at risk and then providing expensive counselling services in every school and college; but in fact suicides fell when students' paths through the educational system became more predictable, leaving less room for illusions and disappointments. The same can be said of job stress: instead of paying for counselling for employees, companies might find that changes in office practice could reduce absence due to sickness and increase productivity.2

    The working group that produced this report drew heavily on a review of health service interventions intended to reduce health inequalities, commissioned by the working group from the Centre for Research and Dissemination.3 Ironically, most of the interventions involved people setting up local services to ameliorate the effects of national policies over which they had no control. Thus not only was the working group prevented from making recommendations to deal with causes of health inequalities lying outside the scope of health services but it drew heavily on policy models developed by people who could not hope to alter national policies.

    Whether the health gain for a given preventive investment is higher among healthy rich people or unhealthy poor people will reflect the underlying policies. There is a paradoxical impression that health gains are most expensive where excess morbidity and mortality are greatest; this probably means that we are using the wrong policies. The solution is to be more radical and to tackle problems at their root. This is now essential. Despite Britain's comparatively low social expenditure4 there is little prospect that any government is likely to pay for a whole new layer of health services. Researchers will have to spend less time identifying high risk groups in need of new services and more time identifying ways of developing the social and economic institutions of our society so that the initial damage to health is prevented.

    The confluence of poorer health and a range of other social problems associated with relative deprivation suggests that it should he possible to devise reforms that have multiple benefits. Indeed, in a broader context, new services may occasionally pay for themselves. For instance, the high/scope Perry randomised controlled trial of preschool education suggests that by the time children reach adulthood an initial investment can yield a sevenfold return by reducing crime rates and welfare dependency and improving educational performance and earnings.5 Clearly policy must be informed by a knowledge not only of health gains but of all the benefits likely to accrue from any proposed policy.

    Getting the research focus right is an urgent priority. The Economic and Social Research Council's pounds sterling4m programme of research on health variations is already under way, and the Department of Health is currently consulting on priorities for its own pounds sterling2.5m research programme in this area. Ideally, as well as putting together the policy interests of a number of different sectors, research would evaluate interventions across a broad range of outcomes. If the government is to give itself a chance of making an appreciable impact on inequalities in health or any associated social problems it must overcome its fears and encourage more imaginative intersectoral approaches to policymaking. The new research programmes should form part of an integrated attempt to develop the productive resources of our society, to equip people to take a full role in its economic life, and to reduce the burden of deprivation.


    View Abstract