Socioeconomic determinants of health: Children, inequalities, and healthBMJ 1997; 314 doi: https://doi.org/10.1136/bmj.314.7087.1122 (Published 12 April 1997) Cite this as: BMJ 1997;314:1122
- Helen Roberts, coordinator, research and developmenta
This article describes a growing body of evidence showing the adverse effects of the widening income gap on the health and welfare of children and young people. The effects of this go well beyond morbidity and mortality and can also be seen in the areas of crime, violence, and educational attainment. There is a need for evidence based policy in this area, but meanwhile there is scope for intervention in pregnancy and the early years, and good evidence that this is effective. A number of well evaluated interventions not necessarily directly related to health, such as early learning programmes and social support for parents, promise to have beneficial health effects.
In the early 1940s, on the publication of Richard Titmuss's Birth, Poverty and Wealth,1 newspapers reported “Poor folks' babies stand less chance” and “Babies beware of poor parents.” Titmuss's work showed that children's deaths were related to the occupations of their fathers and that the gap between the life chances of working class and middle class infants had increased since 1914. Some commentators found his conclusions unpalatable: an Evening Citizen reviewer wrote that the book ignored “the criminal ignorance and neglect of many mothers” who were inclined to give their babies “fish and chips, pickles, strong tea, lollipops, chocolate biscuits and toffee apples.”2
Half a century later, when Barnardo's published Unfair Shares: the Effects of Widening Income Differentials on the Welfare of the Young,3 favourable press coverage urged that it should inform evidence based policy,4 but the inequalities debate continues to attract other interpretations. A Daily Mail article in 1996 concluded: “Rich or poor, life is getting better … the vast majority are doing well and don't need welfare.”5 Now as in the 1940s, mothers–the main caretakers of children–continue to attract adverse press comment, with suggestions that the main need for change lies with them: their children suffer when they go out to work; their diets are not sensible; their discipline is lacking; their morality is in need of attention; and their family structures are suboptimal. This is despite evidence that the majority of mothers living in poverty successfully bring up their children and protect and promote their health under the most unpromising conditions.6 7
The health index least susceptible to interrater variation, or other kinds of reporting bias, is death. The postwar period has seen a decline in perinatal and infant mortality–indeed, in the United Kingdom mortality under the age of 20 years has fallen by over 90% during the 20th century.8 But this masks continuing (and in some cases increasing) problems facing young people. Before housing costs are taken into account, an estimated quarter of all children live in poverty9; when housing costs are allowed for, this rises to almost one in three.10 In Britain, as in the United States, patterns of poverty reflect and reinforce the wider inequalities between black and white communities. To a large extent, the health effects of poverty have been measured quantitatively and indirectly, but there is an important seam of qualitative work describing the texture of the lives of mothers and children living in poverty.11 Poverty involves social exclusion, which itself has adverse psychosocial effects. Emotional problems in childhood can cast a long shadow forward, affecting many aspects of health and behaviour in adult life.
Relative poverty has absolute effects. Data from the 1970 birth cohort show that hyperactivity and conduct disorder, and to a lesser extent anxiety, increase with decreasing social class. Moreover a child in the lowest social class is twice as likely to die before the age of 15 as a child in the highest social class.8 The social class gap in child deaths from accidents has widened over the past decade.12 If this trend persists, the Health of the Nation target on accidents is likely to be met for children in the non-manual classes but not for those from the manual classes.
What is it that links adverse social, psychological, and economic conditions in childhood with adult morbidity, mortality, and other undesirable outcomes? And given that children are not simply trainee adults, with childhood no more than a preparatory period for later outcomes, what is it like to be a child in a society where the income gap is widening? In 1994, it was observed3 that:
Total reported crime, including juvenile crime, increased by almost 80% and violent crime by 90% during 1981-9113
The number of drug offenders between the ages of 17 and 29 doubled between 1979 and 198914
Widening income differentials and relative poverty are not the whole story, but the statistical links between increasing relative deprivation and growing psychosocial problems among young people are compelling.
Early events and later outcomes
Probably the best sources of data on a link between early childhood events and later outcomes are the cohort studies which collect both health and social data from children at intervals, often from shortly after birth into adulthood, and studies that link poor intrauterine growth with later health.16 Substantial social class differences in birth weight may be expected to generate inequalities in health in the future.17 The new health variations programme of the Economic and Social Research Council will be exploring these influences.
From the cohort studies we know that risks of death and serious illness are greatest for people brought up in the lowest social classes, and so are the chances of relatively high blood pressure, poor respiratory health, obesity, and shortness of stature.18 19 20 Work derived from the Swedish level of living surveys similarly describes the effects of adverse childhood conditions on illness and mortality in adulthood. Childhood adversity in this study included family breakdown and–with a stronger impact–family dissent.21 This accords with British findings.22 23 Work based on the youngest cohort of the west of Scotland twenty-07 study looks at the ways in which different aspects of the family lives of young people relate to a range of outcomes chosen as broadly representative of lifestyle, health related behaviours, delinquency, and contact with the police.24 The authors considered both family structure (intact, reconstituted, or single parent), and reasons for family breakdown and distinguished between parental separation and death, and two aspects of family life: family centredness (a measure of time spent in joint family activities) and conflict (frequency of arguments between young people and their parents). Of the four aspects of family life, family centredness showed the strongest and most consistent relation with outcomes for both males and females. The distinction between family structure and family process is useful in helping to understand why simple policy solutions, such as penalising single mothers, are unlikely to be helpful. Outcomes for children in single parents families are, of course, heavily influenced by the fact that such families, usually headed by women, subsist on low incomes.
Can anything be changed until everything is changed?
There are several competing explanations for health inequalities.25 Among these are the importance of psychosocial pathways in understanding the corrosive effect of the growing gap between the haves and the have nots.26 But some things can only be done by governments, and narrowing the income gap is one of these. There is evidence that this strategy works. A randomised controlled trial on income maintenance shows that a guaranteed a minimum income to pregnant women in low income families (by using negative income tax) was associated with a significant increase in birth weight in the intervention group.27 28
In the absence of political and policy change, is there anything which practitioners–health professionals, educationalists, and those delivering child welfare services–can do, or which people and communities can do for themselves? Are more studies of baboons and civil servants needed, or can we take forward, and use creatively, some of what is already known? Given the plausibility of psychosocial pathways, it would be ironic if all solutions were seen as beyond the grasp of ordinary people, who can only wait passively for the powerful to act.
The link between early events and later outcomes–and the recognition that interventions at crucial points may affect this–has long been understood: the Book of Daniel describes an experiment in which children were given pulses to eat and water to drink, rather than the king's wine and meat. While present day nutritionists might be surprised to know that after only 10 days the countenances of the experimental group “were fairer and fatter in flesh than all the children which did eat the portion of the king's meat,” good nutrition as a helpful early intervention policy was clearly established.29
More recently, data from the cohort studies have indicated what might be protective. Parental interest in, and enthusiasm for, education offers the best protection in the long term from the disadvantages of a start in poor socioeconomic circumstances.18 30 Children fortunate enough to have this help tended strongly to do better in cognitive tests and in educational attainment.31 In due course, such children as adults were more likely than were others to be enthusiastic about their own children's education.18 32 The importance of educational attainment is seen in all aspects of the findings on adult life. Those who gained qualifications at A level (or training equivalents) or above had much better chances in health33 34 as well as in occupation and income.18
Early interventions which show promising effects include Highscope, a preschool intervention that incorporates an active learning curriculum, trained staff, and parental participation.35 Highscope shares many of the elements of other good quality preschool interventions, but some aspects of the curriculum, concerned with encouraging the child to “plan, do, review” as part of a daily routine are seen as distinct. For the child, adult acknowledgement that she can make sensible decisions is important; the child is encouraged to be independent and seek solutions within the context of a secure and consistent daily routine.
A well conducted follow up study indicates that at age 27, children who had been randomised to the programme had higher monthly earnings, a higher proportion of home ownership, and fewer arrests including for crimes of drug taking or dealing than those not randomised to the programme (see box).36 What is as important as the later outcomes is the enhanced experiences in childhood and the enjoyment which children gain from these early encounters.
Later effects of Highscope
At age 27, graduates of the Highscope programme, a preschool intervention, had:
Significantly higher monthly earnings (29% v 7% earned $2000 or more per month)
Significantly higher percentage of home ownership (36% v 13%)
A significantly higher level of schooling completed (71% v 54% completed 12th grade or higher)
A significantly lower percentage receiving social services at some time in the last 10 years (59% v 80%)
Significantly fewer arrests (7% v 35% with five or more arrests)
In terms of health interventions, one promising social support intervention is the child development programme developed in Bristol; its fundamental goal is to help support and encourage parents in their task.37 Considerable emphasis is laid on the health and wellbeing of the mother as a woman with her own interests and future as well as being the mother of children. This programme offers monthly support visits to new parents, antenatally and for the first year of life. Most of the visits are undertaken by specially trained health visitors. Perhaps the most radical development of this programme is the community mothers intervention, in which mature mothers were recruited to provide support to younger mothers. A randomised controlled trial showed that children in the intervention group were more likely to have received all of their primary immunisations, to be read to, and to be read to daily. They were less likely to be given cows' milk before 26 weeks. Mothers in the intervention group had a better diet than the controls and at the end of the study, they were less likely to be tired or feel miserable.38
There is good evidence that early interventions effect change. While risks for poor later outcomes are cumulative, the benefits of early intervention cannot be underestimated, as an important paper by Power and Hertzman which pulls together evidence on the early years makes clear.19 These authors also make the point that such interventions cannot fully overcome socioeconomic disadvantage. Providing opportunities for improved cognitive and emotional functioning to socioeconomically disadvantaged children will improve their life chances, but this will not put them on an even playing field with their more advantaged young friends.
Where once the state's welfare apparatus stood as a clear statement of our mutual responsibilities to our fellow human beings, its decline now stands as a denial of that responsibility.3 Evidence based, redistributive social welfare policies would be the best option, not only for children living in poverty but for the rest of us, who live with the corrosive effects of a divided society.
Meanwhile, intervention at practice level, and in particular education in the early years, is clearly worthwhile in affording children and young people the opportunity to experience a good childhood. Just as early trauma may have long terms effects, early interventions enable children and young people to accrue some of the capital needed for good long term outcomes.
This article does not necessarily represent a Barnardo's view. I am grateful to Hilary Graham, Chris Power, Richard Wilkinson, and the BMJ reviewers for comments on an earlier draft.