Intended for healthcare professionals


Increasing inequalities in the health of the nation

BMJ 1994; 309 doi: (Published 03 December 1994) Cite this as: BMJ 1994;309:1453
  1. George Davey Smith,
  2. Jerry Morris

    Government action at last?

    Raising the subject of socioeconomic inequalities in health can produce the weary reaction that these are now so widely recognised that little purpose is served by flogging this dead horse again.1 Certainly the flat denial of such inequalities has almost become a thing of the past. The well known attempt by the government to stifle any reaction to the Black report on inequalities in health 14 years ago2 has given way to an apparently more reasoned approach. Indeed, it was announced at a conference organised by the BMA earlier this year that the government was establishing an interdepartmental working group to examine the links between social position and health3 as part of the continuing review of the Health of the Nation's targets.4

    The setting up of this working party is timely, given the accumulating evidence that socioeconomic differences in health have increased since the Black report was published.5 6 7 In this week's journal McLoone and Boddy show that between the early 1980s and the early 1990s differences in mortality between deprived and affluent small areas of Scotland have increased substantially,8 a phenomenon that has also been reported in the largest Scottish city, Glasgow (p 1482).9

    Much has been made of the poor health picture in Glasgow, which in the popular media has been awarded the title of “heart disease capital of the world.” But, as McLoone and Boddy show, inhabitants of the more affluent parts of Glasgow have the same favourable mortality as residents of equivalent areas in the rest of Scotland. There is no special Glasgow effect; the high concentration of particularly deprived areas in Glasgow translates into a high overall mortality there in the same way as it would elsewhere. The common sense view that poorer areas have worse health profiles because they contain a larger proportion of poor people receives further support from Sloggett and Joshi's analysis of the associations of socioeconomic indicators at individual and area levels with mortality in England and Wales (p 1470).10

    The data from Scotland complement a report of increasing socioeconomic differences in the Northern region of England, which the BMJ published earlier this year.11 Recent extension of this work has shown that the increasing differences in adult mortality have been accompanied by increasing disparities in infant mortality and birth weight.12 In the most deprived 20% of wards in the Northern region infant mortality did not fall between the early 1980s and early 1990s, while a sizeable fall was evident in the least deprived 20% of wards. A similar picture was seen with respect to the proportion of children weighing under 2500 g at birth. The consequences of these changes for differences in health in later life could take many years to emerge.

    These socioeconomic inequalities in health are, as Marmot has pointed out,13 a potentially serious barrier to the achievement of the Health of the Nation's targets. Thus mortality among young men in deprived areas is increasing rather than decreasing,*RF 8-10* partly because of an increase in suicide.8 This acts against the Health of the Nation's target to reduce the overall rate of suicide by at least 15% by 2000. Falls in deaths from coronary heart disease and lung cancer in affluent areas of Scotland are in line with targets from the Health of the Nation and Scotland's Health: A Challenge to Us All14; the much smaller falls in the most deprived areas will, however, severely limit what is achievable.8

    The generally frosty official response to discussions of inequalities in health in Britain contrasts with the experience in the Netherlands, described by Mackenbach (p 1487).15 Differences in life expectancy according to educational level stretch to 12 years among Dutch men, and their mortality and morbidity would fall by an estimated 25-50% if the whole population experienced the same rates as those experienced by the people with the most education. These differences in mortality, large as they seem, are less than those in Britain,16 in line with the smaller inequalities of income in the Netherlands.17

    Compared with Britain, the Netherlands has had little tradition of studying socioeconomic differences, which makes it more noticeable that while the British authorities continued to avoid the issue Dutch government agencies established a national research programme into inequalities in health in 1989. With a fair level of consensus it has been agreed by many agencies in the Netherlands that inequalities in health represent an important public health problem and that future research and health policies need to take them into account. As Mackenbach acknowledges, the approach built on consensus led to concentration on initiatives regarded as feasible “in the current political climate.” While researchers will understandably want to keep their jobs, it will be interesting to observe how the programme progresses over its next five years, when the logic of the research findings may test the boundaries of what is considered to be politically possible.

    Despite its limitations, the Dutch experience contrasts starkly with Britain's. The Health of the Nation devoted only one page to what are referred to as “variations” (not “inequalities”) in health among socioeconomic groups.4 These are said to be the result of “a complex interplay of genetic, biological, social, environmental, cultural and behavioural factors.” The same list of factors was given by Mrs Bottomley in her message to the BMA's conference on inequalities in health,3 with the rider that “there are no easy answers” in this area.

    Yet surely there is some obfuscation here: it is inconceivable, for example, that changes in the genetic make up of different socioeconomic groups have occurred over the past 15 years to produce the increases in differences in mortality in Britain discussed above. Although behaviour related to health is regarded as the appropriate target for health promotion activity, it must be recognised that eating, smoking, drinking, and participating in exercise do not occur in a social vacuum. In poorer areas less healthy food is available and is often more expensive than in affluent areas.18 Similarly, the reason why smoking breaks the rule that households with low incomes cope by decreasing the personal expenditure of adults cannot be reduced to personal failure. Thus for women caring for children in adverse socioeconomic circumstances smoking may be one of the few activities undertaken solely for themselves and one that provides some respite from the strain of coping with the consequences of material deprivation.19

    Policies that ignore the social and economic constraints on behavioural change may produce increases, rather than decreases, in inequalities in health. A more fundamental approach than oversimplifying the origins of behaviours related to health would be to recognise the close association between the size of income and differences in mortality between 1921 and 1981,7 20 with the recent widening of inequalities in health mirroring the gross upwards redistribution of wealth since 1979.21 If progress is to be made it is necessary at least to start addressing the fact that increasing inequalities in health are a consequence of our increasingly polarised society.

    The government's new working party will be judged by its efforts to move the national debate forward to remedying and preventing inequalities in health, with measures that can be started now. The NHS has its part to play—for example, in the allocation of resources to primary health care in inner cities.22 Research shows, however, that the main agenda lies elsewhere and should focus on child and family poverty, on housing (which will also create employment), on the provision of nursery education, and the like. The recommendations of the working party on such issues are awaited. “Interdepartmental” commitment, which is central to the Health of the Nation and its credibility, will be tested here. Any such measures will, of course, be costly. But then so is the waste of human capital in the premature death that the multiplying statistics represent—not to mention the burden of disability and mass of human misery that are entailed.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    13. 13.
    14. 14.
    15. 15.
    16. 16.
    17. 17.
    18. 18.
    19. 19.
    20. 20.
    21. 21.
    22. 22.
    View Abstract