EXECUTIVE SUMMARY

  1.  The term "health inequalities" describes a social gradient in health in which each step up the socioeconomic ladder, from the most disadvantaged to the most advantaged, is associated with progressively higher standards of health. The "health gap"—defined in England's targets as the difference between the health of disadvantaged groups and the population average—is one aspect of the health gradient.

  2.  Health gradients and health gaps are widely acknowledged to be unjust and unfair. International charters commit UK governments to pursuing policies which level up opportunities for everyone to enjoy the standards of health currently the preserve of the most advantaged groups.

  3.  Health inequalities are persisting despite record-breaking levels of wealth and health in the UK. Moreover, the evidence points to widening health inequalities. The trend reflects the slower rate of health gain in more disadvantaged groups. As a result, health gaps have been widening and health gradients have been steepening.

  4.  The major determinants of health lie beyond the NHS in people's living and working conditions and health behaviours. Inequalities in health are the outcome of the unequal distribution of these determinants. Advantaged groups enjoy greater access to the determinants of good health; disadvantaged groups experience greater exposure to the determinants of poor health, chronic disease and premature death.

  5.  Behind widening inequalities in health lie widening inequalities in health determinants. Over the last 30 years, inequalities in key determinants like living standards and cigarette smoking have widened sharply—and currently show little sign of narrowing.

  6.  Widening inequalities in health determinants reflect a mix of factors. These include a shift to less redistributive economic policies and the absence of strong regulatory policies with the potential to curb uptake and consumption of health-damaging products like cigarettes in disadvantaged groups. These factors are amenable to government policy.

1.  INEQUALITIES IN HEALTH TAKE THE FORM OF A SOCIAL GRADIENT

  1.1  The UK is scarred by deep socioeconomic inequalities in health.

  1.2  These inequalities take the form of a social gradient. Those in the most advantaged circumstances enjoy the highest standards of health: they are least at risk of disability in childhood and adulthood and of chronic disease and premature death. Standards of health are lower for those in less advantaged circumstances, and decline in stepwise fashion as levels of disadvantage increase.

  1.3  The health gradient is illustrated in Figure 1. It takes life expectancy as the measure of health, and social class based on occupation as the measure of people's socioeconomic circumstances. Occupational social class was the official measure of people's socioeconomic circumstances from 1911 to 2000 when it was replaced by a new schema, the National Statistics-Socioeconomic Classification (NS-SEC)[27]; occupational social class therefore provides the best measure of trends in health inequalities.

  1.4  Figure 1 also captures the health gap between the most disadvantaged group (social class V, unskilled manual) and both the most advantaged group (social class 1, professional) and the population average (all). These are indicated in darker shades.

  Source: ONS, 2007.[28]

  1.4  As Figure 1 indicates, health gradients exist for men and women. They are evident at all points in the life course and exist across both the majority white population and among minority ethnic groups.

2.  HEALTH GRADIENTS AND HEALTH GAPS ARE UNJUST AND UNFAIR

  2.1  Socioeconomic inequalities in health have long been acknowledged to be inequitable: to be unfair and unjust. In 1946, the World Health Organisation (WHO) was established under the UN Charter. It is founded on the principle that "every human being without distinction of race, religion, political belief, economic or social condition" has the fundamental right to achieve "the highest attainable standard of health"[29]. The principle has been repeatedly reasserted in international charters and in European strategies to which the UK Government has again been a signatory.

  2.2  The principle is commonly understood to mean that everyone in a society should have an equal chance of reaching the standards of health which are currently the preserve of the well-off. In policy terms, it means a commitment to "leveling up": to lifting levels of health across the socioeconomic hierarchy to those attained in the highest socioeconomic group.

3.  HEALTH INEQUALITIES ARE PERSISTING AND WIDENING DESPITE RECORD-BREAKING LEVELS OF WEALTH AND HEALTH IN THE UK

  3.1  Over the last 30 years, living standards have risen rapidly and life expectancy has improved steadily. But the policy mix which has made the UK population wealthier and healthier has failed to level up opportunities between socioeconomic groups to live a long and healthy life.

  3.2  Instead, inequalities in health have persisted. Moreover, the evidence points to widening inequalities in the risk of ill-health and premature death in recent decades. The Government's goal of reducing inequalities therefore requires first a halt, and then a reversal, of a long-running trend.

  3.3  The trend reflects the fact that health in more disadvantaged groups is not improving as rapidly as in more advantaged groups. There has been a differential rate of health gain, with rates of improvement in more advantaged groups outstripping rates in more disadvantaged groups. As a result, health gaps are widening and health gradients are steepening. This trend is captured for life expectancy in Figure 2.

  3.4  England's health inequalities targets focus on narrowing the gap between disadvantaged groups and the population as a whole and, specifically, on reducing infant mortality rates in lower socioeconomic groups and lifting life expectancy in disadvantaged areas closer to the national average by 2010. The latest data indicate that infant mortality rates have fallen and life expectancy has risen in the target groups from the baseline period of 1997-99. However, the rate of improvement has continued to be greater in the population as a whole. As a result, the long-term trend has been maintained rather than reversed, and the relative gap in both indicators has widened over the last decade[30], [31],[32]. This suggests that the Government is unlikely to meet its PSA targets in respect of health inequalities.

  Source: ONS, 2007.[33]

  3.5  It is important to note that England's health inequalities targets exclude groups at high risk of infant death and premature death. For example, the infant mortality target focuses on jointly-registered births where the father has an occupation in the NS-SEC "routine and manual" group. Babies born to other groups of parents, including those solely registered by the mother and jointly registered to parents where the father is long-term unemployed or has never worked, have appreciably higher infant mortality rates[34].

4.  INEQUALITIES IN HEALTH ARE THE OUTCOME OF INEQUALITIES IN HEALTH DETERMINANTS

  4.1  It is widely recognised that individual and population health is shaped by a web of social influences. This web is often referred to as the social determinants of health.

  4.2  While there is a range of models of the social determinants of health, they conform to a common format. In these models, the social determinants of health have their origins in:

—  The overarching structure of society, which includes its labour market, education system, tax and benefit systems etc. This structure shapes the conditions in which people live their lives, in childhood, through adulthood and into old age.

—  People's living and working conditions, as measured by occupation, household income, area deprivation etc. These conditions are shaped by wider societal factors and they mediate access and exposure to a set of intermediate risk factors which have more direct effects on health. Living and working conditions are therefore identified as the pivotal determinant of health.

—  Intermediate risk factors include the social and material environment of the home, neighbourhood and workplace, which both provide resources for health and contain risks for health. Intermediate factors include, too, behaviours which can be either health protecting and enhancing (like exercise) or health damaging (like cigarette smoking and energy dense/nutrient poor diets).

  4.3  Social determinants influence health from before birth and across life. How long an individual lives in advantaged or disadvantaged conditions therefore matters for their health. Figure 3 is based on a Scottish study which used information on social class at three time-points—at the time the study participant was born, when they first entered the labour market and their occupation in mid-life—to allocate men to a manual or non-manual social class across their lives[35]. A graded association was revealed between cumulative social class and all-cause mortality. Thus, men in a manual social class at all three time points—the largest group—had the highest mortality rate over the 21 years of follow-up. Mortality rates fell progressively for men who spent smaller proportions of their lives in the manual group.

  Source: Davey Smith, 1997.[36]

  4.4  Death rates from cardiovascular disease and cancer displayed a similar pattern (Figure 3). Men born into and remaining in the manual group experienced the highest rates of death from these diseases; rates were lower for those who spent longer periods in more advantaged class positions[37]. A similar pattern is also revealed in the 1958 British birth cohort study for physiological risk factors for chronic disease in mid-life, including BMI, cholesterol, chronic widespread pain and depressive symptoms[38]. As this suggests, "lifetime" socioeconomic position—measured from birth to adulthood—is a powerful predictor of adult disease and mortality risk[39].

  4.5  While people's circumstances across their lives matter for their health, childhood circumstances have been found to be particularly important. Figure 4 describes the patterns for premature mortality. Based on the 1946 British birth cohort study, it plots the survival of children from manual and non-manual backgrounds from the age of 26 (312 months), when almost all children were still alive, to age 54 (648 months). It captures the marked socioeconomic differentials in survival, with death rates among women and men born into manual households double those of women and men growing up in non-manual households. Their increased risk of death was little reduced after account was taken of socioeconomic circumstances in adulthood: in other words, the effect of poor circumstances in childhood was independent of whether adulthood brought continuing disadvantage or better circumstances[40]. What is evident for all-cause mortality is repeated for a number of specific causes of death. For example, childhood disadvantage has been found to make a significant contribution to the risk of cardiovascular disease, over and above the effects of adult social class[41].

  Source: Kuh et al, 2002.[42]

  4.6  The circumstances in which children are conceived, are born and grow up matter for their health because childhood conditions influence both socioeconomic circumstances and health in adult life. With respect to future circumstances, children born to professional parents are more likely to do well at school and progress to a secure, rewarding and well-paid career than those growing up in families lower down the social scale—and these inequalities in future prospects have been widening not narrowing. They are also less likely to follow domestic trajectories like early and lone parenthood which increase the risk of disadvantage in adult life—and increase the risk that their children will also face a lifetime of disadvantage.

  4.7  With respect to future health, childhood circumstances matter because early life is a period when children's body systems mould and adapt to the environments in which they are developing. While the development process is genetically regulated, how genes are expressed is determined by the child's physical and social environment[43]. The process is described as "embodiment" and "biological embedding" by epidemiologists[44]. Experiences which are physically and emotionally nurturing become written into bodily structures and functions in ways which promote and protect their future health. Conversely, environmental adversity in the early years of life has been found to induce long-term patterns of physical, cognitive and emotional development which leave children vulnerable to developmental delay and poor health.

  4.8  Health behaviours are also often set in childhood and then track into adulthood. For example, adolescence is the critical period for uptake of cigarette smoking. Young people growing up in disadvantaged circumstances are more likely than their advantaged peers to become regular smokers (Figure 5)—and to continue to smoke, and smoke heavily, through adulthood.

  Notes: (1) daily smoker reported smoking > 7 cigarettes a week and heavy smokers reported smoking > 35 cigarettes a week (2) parental social class based on occupation of head of household. Source: Sweeting and West, 2001.[45]

  4.9.  Taken together, the evidence on health determinants suggests that the NHS plays a secondary role in influencing people's health and reducing health inequalities. Services which screen and treat those at risk of early death are important but, on their own, they can not change the social conditions which generate risk in current and future generations. The DH emphasises the responsibility of local NHS services to meet England's health inequalities targets; there is a strong argument for shifting responsibility for delivery of the PSA health inequalities targets to Government departments with capacity to reduce inequalities in living conditions.

5.  WIDENING INEQUALITIES IN HEALTH ARE ASSOCIATED WITH AN INCREASINGLY UNEQUAL DISTRIBUTION OF HEALTH DETERMINANTS.

  5.1  It is widely accepted that inequalities in health are the outcome of inequalities in health determinants, operating along the causal chain from the overall structure of society, through people's living and working conditions to their health behaviours (see paras 4.1-4.2).

  5.2  Over the last 30 years, inequalities in key determinants of health have widened sharply—and currently show little sign of narrowing. Inequalities in living conditions (as indexed by living standards) and in health behaviours (cigarette smoking) provide two examples.

  5.3  Living standards are measured by disposable household income (income after direct taxation and cash transfers). Evidence for the 1940s and 1950s points to narrowing inequalities in household income, with richer households seeing their share of national income fall. Figure 6 picks up the story in the early 1960s, and suggests that, until the mid-1970s, income inequalities in Britain were marked but stable. From the late-1970s, real incomes at the upper end of the income distribution set off on a sharply upward trajectory; the incomes of the poor, meanwhile, stagnated. The trend has slowed since the early 2000s but has yet to be reversed.

  Source: FES, IFS analyses.[46]

  5.4  Because children are disproportionately represented among lower-income households, the trend towards wider income inequalities has disproportionately impacted on children. Rates of child poverty rose rapidly from the early 1980s (Figure 7). Because exposure to poverty in childhood has lifelong effects on future health (see paras 4.5-4.8), we should anticipate that rising rates of child poverty through the 1980s and 1990s will "cast long shadows forward" over the health of these cohorts as they grow older.

  Source: IFS, 2006, DWP, 2007.[47]

  5.5  Interventions like Sure Start and Health Action Zones are operating against the trend of widening inequalities in living standards. It should not surprise us that interventions which seek to reduce inequalities in health by tackling behavioural factors (eg parenting skills, infant feeding practices, cigarette smoking) are struggling to make a dent on inequalities in either the targeted risk factor or in health.

  5.6  Cigarette smoking provides a second example of widening inequalities in determinants. The downward trend in smoking prevalence has been associated with a steepening social gradient, with a later and slower rate of decline in prevalence in more disadvantaged groups. Figures 8 and 9 capture the widening of inequalities in smoking from 1958-2000 (based on social class) and continuing inequalities from 2001 (based on NS-SEC). In consequence, inequalities in cigarette smoking make a major contribution to health inequalities.

  Source: Wald and Nicolaides-Bouman, 1991, Table 5.2[48]; ONS, 2001, Table 8.8.[49]

  Source: ONS, 2006.[50]

6.  WIDENING INEQUALITIES IN HEALTH DETERMINANTS REFLECT A MIX OF FACTORS AMENABLE TO POLICY INTERVENTION

  6.1  Policies to tackle inequalities in health encompass those which moderate inequalities in the social structure (in the education system, labour market and the wider distribution of property and wealth), in people's living and working conditions and in environmental and behavioural risk factors. Governments exercise considerable influence over the magnitude of these inequalities. Inequalities in living standards and in cigarette smoking provide two examples.

  6.2  Because earnings are the major source of income for all households except those in lowest income deciles, a progressive direct tax system distributes income from rich to poor. Welfare benefits are the major source of income for poorest groups but represent a very small proportion of income of higher income groups, so welfare policies provide a lever through which to level up incomes. Through these two key policy instruments, inequalities in disposable income are less extreme than those in market income. In UK, benefits play the larger part in equalising incomes and living standards.

  6.3  Analyses by welfare economists suggest that the trend towards widening inequalities in disposable household incomes (Figure 6) reflects changes in Government policies. Compared with the period from 1940 to 1980, the tax and benefit systems do less to moderate inequalities in market incomes. In consequence, inequalities in disposable income have widened.

  6.4  The UK tax and benefit system is less effective in protecting the living standards of low-income households than those in other high-income countries. Figure 9 focuses on child poverty rates before and after direct taxation and cash benefits. It suggests not only are rates of poverty before income transfer appreciably lower in Nordic countries than in the UK, but also that national tax and benefit policies combine to offer a higher degree of protection from poverty. In Sweden, child poverty rates are over 70% lower after direct tax and benefits. In the UK, the proportion of children lifted out of poverty through these policy levers is appreciably lower: here, poverty rates are reduced by 46%.

  Source: Whiteford and Adema, 2006.[51]

  6.5  To date, redistributive income policies have not featured strongly in the UK's health inequalities strategy. However, they do in some other countries. For example, the first objective of the Norwegian's health inequalities strategy is to "reduce economic inequalities". It notes that "the Government is going to take steps to reduce economic inequalities in the population . . . The taxation system will be improved so it does more to ensure a fairer income distribution"[52]. Policies to achieve a fairer income distribution in the UK would provide a platform from which to tackle inequalities in other health determinants.

  6.6  Turning to cigarette smoking, evaluating the effects of policies is not straightforward. Widening socioeconomic gradients in smoking (Figure 8) suggests that tobacco control policies in the 1970s, 1980s and 1990s—which relied heavily on warning smokers of the dangers of smoking—had differential effects, achieving larger reductions in prevalence in better-off groups than in disadvantaged groups. However, it is important to note that smoking rates were already declining in advantaged groups before tobacco control policies were first introduces in the 1970s (Figure 8). As this suggests, changes in the social profile of smoking are likely to occur independently of tobacco control policies, making evaluation of their additional contribution difficult to assess.

  6.7  To address current socioeconomic differentials in smoking prevalence (Figure 9) requires tobacco control policies which are at least as effective in more disadvantaged groups as in better-off groups. There is suggestive evidence that these policies include population-wide policies which institute large and sustained increases in cigarette prices, comprehensive clean air laws, mandatory bans on cigarette advertising and promotion as well as well-funded public information campaigns[53],[54]. The UK has instituted such policies over the last decade, with cigarette prices now among the highest in Europe[55] and legislation prohibiting smoking in public places and most forms of tobacco advertising. In addition, while England's NHS smoking cessation services only reach a small minority of smokers who try to quit each year and disadvantaged smokers have lower quit rates[56], careful targeting means that they are helping proportionately more smokers in Spearhead areas than in other parts of the country[57].

  6.8  In combination, current tobacco control policies may break with the trend over the last 50 years and achieve a greater reduction in prevalence in lower socioeconomic groups—and thus narrow socioeconomic differentials in smoking. However, a policy programme which levels up life chances and living conditions across the population is likely to be a pre-condition for a wider reduction in inequalities in health determinants and a reduction in health inequalities.

Hilary Graham[58]

January 2008

28   Office for National Statistics. 2007, London: ONS. Back

29   World Health Organisation, Constitution of the World Health Organisation. 1946, London: WHO. Back

30   Department of Health, Tackling Health Inequalities: Status Report on the Programme for Action. 2005, Department of Health: London. Back

31   Department of Health, Tackling Health Inequalities: Status Report on the Programme for Action-2006 Update of Headline Indicators. 2006, London: Department of Health. Back

32   Department of Health, Tackling Health Inequalities: 2004-06 data and policy update for the 2010 National Target. 2007, London: Department of Health. Back

33   Office for National Statistics. 2007, London: ONS. Back

34   Department of Health, Tackling Health Inequalities: Status Report on the Programme for Action-2006 Update of Headline Indicators. 2006, London: Department of Health. Back

35   Davey Smith, G., et al., Lifetime socioeconomic position and mortality: prospective observational study. BMJ, 1997. 314(7080): p. 547-52. Back

36   Davey Smith, G., et al., Lifetime socioeconomic position and mortality: prospective observational study. BMJ, 1997. 314(7080): p. 547-52. Back

37   Davey Smith, G., et al., Lifetime socioeconomic position and mortality: prospective observational study. BMJ, 1997. 314(7080): p. 547-52. Back

38   Power, C., et al., Life-course influences on health in British adults: effects of socio-economic position in childhood and adulthood. International Journal of Epidemiology, 2007. 36(3): p. 522-531. Back

39   Blane, D., et al., Association of cardiovascular disease risk factors with socioeconomic position during childhood and during adulthood. British Medical Journal, 1996. 313: p. 1434-8. Back

40   Kuh, D., et al., Mortality in adults aged 26-54 years related to socioeconomic conditions in childhood and adulthood: post war birth cohort study. British Medical Journal, 2002. 325(7372): p. 1076-80. Back

41   Lawlor, D.A., Y. Ben-Shlomo, and D.A. Leon, Pre-adult influences on cardiovascular disease, in A Life Course Approach to Chronic Disease Epidemiology, D.L. Kuh and Y. Ben-Shlomo, Editors. 2004, Oxford University Press: Oxford. Back

42   Kuh, D., et al., Mortality in adults aged 26-54 years related to socioeconomic conditions in childhood and adulthood: post war birth cohort study. British Medical Journal, 2002. 325(7372): p. 1076-80. Back

43   Halfon, N. and M. Hochstein, Life Course Health Development: An Integrated Framework for Developing Health, Policy, and Research. The Milbank Quarterly, 2002. 80(3): p. 433-479. Back

44   Kuh, D., et al., Life course epidemiology. Journal of Epidemiology and Community Health, 2003. 57(10): p. 778-783. Back

45   Sweeting, H. and P. West, Social class and smoking at age 15: the effect of different definitions of smoking. Addiction, 2001. 96(9): p. 1357-1359. Back

46   Institute for Fiscal Studies web site, Inequality, poverty and well-being spreadsheet. Accessed December 2008,www.ifs.org.uk/projects_research.php?heading_id=8_. Back

47   Institute for Fiscal Studies web site, Inequality, poverty and well-being spreadsheet. Accessed December 2008,www.ifs.org.uk/projects_research.php?heading_id=8_. Back

48   Wald, N. and A. Nicolaides-Bouman, UK Tobacco Statistics. 1991, Oxford: Oxford University Press. Back

49   Office for National Statistics, Living in Britain: Results from the 2000/01 General Household Survey. 2001, London: The Stationery Office. Back

50   Office for National Statistics, Smoking and Drinking among Adults, 2005. 2006, London: ONS. Back

51   Whiteford, P. and W. Adema, Combating Child Poverty in OECD Countries: Is Work the Answer? DELSA/ELSA/WD/SEM(2006)7. 2006, Paris: OECD. Back

52   Norwegian Ministry of Health and Care Services, National strategy to reduce social inequalities in health, Report No 20 (2006-2007) to the Storting. 2007, Oslo: Norwegian Ministry of Health and Care Services. Back

53   Levy, D.T., F. Chaloupka, and J. Gitchell, The effects of tobacco control policies on smoking rates: a tobacco control scorecard. J Public Health Manag Pract, 2004. 10(4): p. 338-353. Back

54   Joossens, L. and M. Raw, The Tobacco Control Scale: a new scale to measure country activity. Tobacco Control, 2006. 15: p. 247-253. Back

55   Joossens, L. and M. Raw, The Tobacco Control Scale: a new scale to measure country activity. Tobacco Control, 2006. 15: p. 247-253. Back

56   Ferguson, J., et al., The English smoking treatment services: one-year outcomes. Addiction, 2005. 100(s 2): p. 59-69. Back

57   Bauld, L., K. Judge, and S. Platt, Assessing the impact of smoking cessation services on reducing health inequalities in England: observational study. Tobacco Control, 2007. 16: p. 400-404. Back

58   Hilary Graham is Professor of Health Sciences, University of York. She has a background in social and public health research, was a member of the Acheson Inquiry into Inequalities in Health and is currently Director of the DH Public Health Research Consortium. Her submission is made in a personal capacity and draws on her book, Unequal Lives: Health and Socioeconomic Inequalities, published in October 2007 by Open University Press (http://mcgraw-hill.co.uk/html/0335213693.html). Back