Introduction

Our Task

Our task is set out in the terms of reference and the commissioning letter from the Minister for Public Health (annex A). It consists of two parts. The first is to review the latest available information on health inequalities and "summarise the evidence of inequalities of health and the expectation of life in England and identify trends". This review would be based on data from the Office for National Statistics (ONS), the Department of Health (DH) and elsewhere.

The second is to identify, in the light of the review, "priority areas for future policy development . . . likely to offer opportunities for Government to develop beneficial, cost effective and affordable interventions to reduce health inequalities". These policy proposals are to be based on "scientific and expert evidence" and "within the broad framework of the Government's financial strategy".

Bearing in mind the commissioning letter and terms of reference, we have considered the work of the Inquiry to be scientific. We have limited our recommendations to those based on scientific and expert evidence.

The short timescale of the Inquiry, combined with the broad nature of inequalities in health and their determinants, prohibited a very detailed and comprehensive review. We acknowledge at the outset of this report that there are areas which, given a longer period of time for our work, we would have reviewed in more detail. Other areas of work were omitted because they were not included in our terms of reference.
So, although we recognise that the setting of targets concerned with reducing inequalities in health is an important area for policy development, we were advised that consideration of this issue was not within the Inquiry's remit. We do, however, welcome the setting up of the Chief Medical Officer's working group which will consider targets, including those which address inequalities in health, as part of the work on "Our Healthier Nation"1. In addition, we decided at an early stage not to consider recommendations for research and development, although the need for further research and development is implicit in many sections of the report.

A key objective of our report is to contribute to the development of the Government's strategy for health and an agenda for action on inequalities in the longer term. The publication on the 5 February 1998 of the consultation paper "Our Healthier Nation; a Contract for Health"1 was an important landmark. It identified the need "to improve the health of the worst off in society and to narrow the health gap" as an overriding principle. This principle also underpins consultation papers on public health from Scotland, Wales and Northern Ireland2-4.

Our report takes account of the main features of "Our Healthier Nation" as they affect inequalities. We discuss tackling inequalities in the settings of schools, the workplace and neighbourhoods. Our section on the NHS includes an element on the reduction of inequalities through local partnerships taking account of plans for Health Improvement Programmes and Health Action Zones. It also takes into account the changes outlined in the White Paper "The New NHS: Modern and Dependable"5.

Structure of the report

Our report is divided into two sections. Part 1 sets out the approach which we adopted in considering the causes of inequalities in health, and some of the principles which have guided our work. This is followed by a summary of our review of data on inequalities in health, "The Current Position". Part 2 is our review of the evidence from which we identified areas for future policy development, and contains our recommendations. This section also adds to, and amplifies, some of the data presented in Part 1. In each of the identified areas for future policy development, we have summarised the inequalities that exist, the evidence that indicates areas for policy development, and the benefit which might result from such development. A complete list of our recommendations, including cross references, is given at the end of Part 2.

Our approach

Historical context
Our report needs to be seen in its historical context, as an extension of scientific and policy development in this country over more than a century. There is a long tradition in Britain of analysing national statistics to shed light on the nature and causes of social inequalities in health6. This goes back at least to William Farr in 1837, when the General Register Office was set up. Farr, as the first Superintendent of Statistics, clearly believed that it was the responsibility of the national office not just to record deaths, but to uncover underlying linkages which might help to prevent disease and suffering in the future7.

Firm foundations were set at that time which have allowed the documentation and monitoring of health inequalities over the past 150 years to a much finer degree than in many other countries. Social and public health reformers since then - from Chadwick in the 1840s to Rowntree at the turn of the century and Titmuss and colleagues in the Depression and post-war period - have carried on the tradition, bringing the evidence into the light of day for public debate and action.

Evidence on social inequalities and of inadequate access to health care in Britain also played a key role in pressure to set up the welfare state in the post-war period, with the landmark Beveridge Report of 1942 setting out a national programme of policies and services to combat the "five giants of Want, Disease, Ignorance, Squalor and Idleness"8.

It was an assessment in the mid-1970s that Britain was slipping behind some other countries in health improvement, despite 30 years of the welfare state, and speculation that persisting health inequalities were to blame, that led to the setting up by the Government of the Research Working Group on Inequalities in Health in 1977, chaired by Sir Douglas Black. The resulting Black Report9 presented in 1980, shortly after a new Government took office, was a rare example, perhaps the first anywhere in the world, of an attempt authorised by Government to explain trends in inequalities in health and relate these to policies intended to promote as well as restore health10.
The thrust of the recommendations in that seminal report were concerned with improving the material conditions of life of poorer groups, especially children and people with disabilities, coupled with a re-orientation of health and personal social services. Although there was little sign that the report's recommendations were given any official priority in Britain throughout the 1980s, ripples from the report spread out far and wide, to be influential in research and public health debates in many countries. For example, the Black Report played a part in influencing the decision of the member states (including the UK) of the European Region of the World Health Organisation to agree a common health strategy in 1985, with equity in health as a theme running right through it, and reduction in inequities as the subject of the first of 38 targets to be achieved by the year 200011. This in itself has proved a significant development on the international front. In 1987, an update of the evidence in the Black Report was commissioned and published under the title of "The Health Divide"12. This stimulated widespread debate and led to renewed calls for greater priority to be given to the issue of inequalities in health10.

It was not until the 1990s, however, that significant movement on the issue was perceptible. The Chief Medical Officer for England set up a sub-group under the auspices of "The Health of the Nation" national health strategy, to look into what the Department of Health and the NHS could do to reduce variations in health13. The report of the sub-group was published in 1995, and in the same year, the King's Fund published an independent analysis of the wider policy options for tackling inequalities in health in relation to housing, family poverty, and smoking as well as the NHS14. These initiatives, together with a growing body of evidence from a great many in the public health field, were influential in convincing the new Government in 1997 of the need to set up the current Independent Inquiry.

Socioeconomic model of health
We have adopted a socioeconomic model of health and its inequalities. This is in line with the weight of scientific evidence. Figure 1 shows the main determinants of health as layers of influence, one over another15,16. At the centre are individuals, endowed with age, sex and constitutional factors which undoubtedly influence their health potential, but which are fixed. Surrounding the individuals are layers of influence that, in theory, could be modified. The innermost layer represents the personal behaviour and way of life adopted by individuals, containing factors such as smoking habits and physical activity, with the potential to promote or damage health. But individuals do not exist in a vacuum: they interact with friends, relatives and their immediate community, and come under the social and community influences represented in the next layer. Mutual support within a community can sustain the health of its members in otherwise unfavourable conditions. The wider influences on a person's ability to maintain health (shown in the third layer) include their living and working conditions, food supplies and access to essential goods and services. Overall there are the economic, cultural and environmental conditions prevalent in society as a whole, represented in the outermost layer.

The model emphasises interactions between these different layers. For example, individual lifestyles are embedded in social and community networks and in living and working conditions, which in turn are related to the wider cultural and socioeconomic environment.

Socioeconomic inequalities in health reflect differential exposure - from before birth and across the life span - to risks associated with socioeconomic position. These differential exposures are also important in explaining health inequalities which exist by ethnicity and gender. One model of how these risks interconnect is shown in figure 2.

This model has been used to guide research. The research task is to trace the paths from social structure, represented by socioeconomic status, through to inequalities in health. This can be done in stages, for example showing that work is related to pathophysiological changes such as raised blood pressure or biochemical disturbances which are in turn related to disease risk; or showing that the social environment in which people live is related to their health behaviour, such as patterns of eating, drinking, smoking and physical activity.

The model also illustrates various intervention points. Medical care, for example, might intervene at the level of morbidity to prevent progression to death, or earlier, at the level of pathophysiological changes to interrupt transition to morbidity. Preventive approaches might act at the level of attempting to change individual risk, by encouraging people to give up smoking or change diet. Interventions in the workplace or the social environment might encourage a climate which promotes healthy behaviour or improved psychological conditions. Interventions at the level of social structure would reduce social and economic inequalities.

Our approach is shared by the Government which, in "Our Healthier Nation", has expressed its determination to tackle "the root causes of health". The Prime Minister emphasised this approach in his answer to a Parliamentary Question on low income, inequality and health (11th June 1997).

". . . It is for that reason that the Secretary of State for Health has asked Sir Donald Acheson to conduct a further review into inequality and the link between health and wealth ............... These inequalities do matter and there is no doubt that the published statistics show a link between income, inequality and poor health. It is important to address that issue, and we are doing so. The purpose of the windfall tax is to address that matter on behalf of young people and the long-term unemployed. We are also addressing the issue by introducing the minimum wage, which will help those on low incomes, and with welfare measures, particularly those designed to get single parents back to work"17.

Need to intervene on a broad front
The socioeconomic model also dictates the breadth of our review. A broad front approach reflects scientific evidence that health inequalities are the outcome of causal chains which run back into and from the basic structure of society. Such an approach is also necessary because many of the factors are interrelated. It is likely to be less effective to focus solely on one point if complementary action is not in place which influences a linked factor in another policy area. Policies need to be both "upstream" and "downstream".

For instance, a policy which reduces inequalities in income and improves the income of the less well off, and one which provides pre-school education for all four year olds are examples of "upstream" policies which are likely to have a wide range of consequences, including benefits to health. Policies such as providing nicotine replacement therapy on prescription, or making available better facilities for taking physical exercise, are "downstream" interventions which have a narrower range of benefits.

We have, therefore, recommended both "upstream" and "downstream" policies - those which deal with wider influences on health inequalities such as income distribution, education, public safety, housing, work environment, employment, social networks, transport and pollution, as well as those which have narrower impacts, such as on healthy behaviours. We describe the impact of these on health at the various stages of the life course, by ethnicity and by gender.

Absolute and relative inequalities
The health gap between socioeconomic groups can be considered in both relative and absolute terms. An example of a relative measure would be the ratio of the death rate in the lowest social class to that in the highest class. Death rates could be, for example, twice as high in the lowest as in the highest social class. The equivalent absolute measure would subtract the death rate in one group from that in another to give the rate difference. This could be expressed as, for example, the death rate in the lowest social class is 50 deaths per 100,000 population greater than the rate in the highest social class.

Both relative and absolute measures have important implications. However, it may be argued that absolute measures are the most critical, particularly with respect to identifying the major problems which need to be addressed. This is because an absolute measure is determined not only by how much more common the health problem is in one group than another, but also how common the underlying problem - for example the death rate in a particular population - actually is. A doubling in social class V of the rate of occurrence of a rare disease is not as significant as a doubling in the rate of occurrence of a common disease. Major gains in attacking health inequalities are most likely to derive from addressing those health problems which occur reasonably frequently, even if less common diseases may in relative terms demonstrate a steep gradient, occurring, say, ten or twenty times more often in social class V than I. Relative measures are particularly useful for assessing the relative importance of different causal factors, and are important tools in aetiological enquiry.

The penalties of inequalities in health affect the whole social hierarchy and usually increase from the top to the bottom. Thus, although the least well off may properly be given priority, if policies only address those at the bottom of the social hierarchy, inequalities will still exist. Accordingly, our approach addresses the socioeconomic determinants of health as they affect the whole social spectrum.

Social environment, social support and health
The economic and social benefits of greater equality seem to go hand in hand. The quality of the social environment is worst where financial deprivation is greatest, such as the inner cities. Recent research suggests that, in addition to the ill effects due to absolute poverty, societies in which there is a wide gap between the rich and the poor suffer additional social problems, for instance, through high rates of violence and crime, and truancy18. It has also been suggested that people with good social networks live longer, are at reduced risk of coronary heart disease, are less likely to report being depressed, or to suffer a recurrence of cancer and are less susceptible to infectious illness than those with poor networks19.

This work opens up a range of policy options. Policies to reduce social inequalities and to promote social networks are part of a strategy to reduce inequalities in health in just the same way as action on economic inequalities or improvements in the material environment of disadvantaged communities. These include, for instance, policies which reduce unemployment in areas of social need, those which improve the availability of social housing for families close to their social networks, and the provision of family support services which help parents protect their children from the effects of disadvantage. Freedom from prejudice or discrimination, a respect for individual worth and a sense of belonging to society will help to reduce the manifestations of exclusion, such as crime, violence, self-harm and isolation.

Priority for parents and children
While remediable risk factors affecting health occur throughout the life course, childhood is a critical and vulnerable stage where poor socioeconomic circumstances have lasting effects. Follow up through life of successive samples of births has pointed to the crucial influence of early life on subsequent mental and physical health and development20. The fact that the adverse outcomes, for example, mental illness, short stature, obesity, delinquency and unemployment, cover a wide range, carries an important message. It suggests that policies which reduce such early adverse influences may result in multiple benefits, not only throughout the life course of that child but to the next generation.

Another line of research, which concentrates on the effects of a mother's nutrition on her child's later health, has shown that small size or thinness at birth are associated with coronary heart disease, diabetes and hypertension in later life. As two principal determinants of a baby's weight at birth are the mother's pre-pregnant weight and her own birthweight, the need for policies to improve the health of (future) mothers and their children is obvious21. It also follows that, among migrants who move from a poorly nourished to a well nourished community, there will be implications for fetal growth and adult health for more than one generation.

Taking into account these findings and the view expressed in "Our Healthier Nation" that "good health is the supreme gift parents can give their children", we take the view that, while there are many potentially beneficial interventions to reduce inequalities in health in adults of working age and older people, many of those with the best chance of reducing future inequalities in mental and physical health relate to parents, particularly present and future mothers, and children.