|Home | About | Journals | Submit | Contact Us | Français|
Evidence is needed on how best to reduce inequalities
A recent Unicef report ranked the wellbeing of children in 21 rich countries.1 The report aggregated national data on more than 40 indicators from credible sources in six dimensions—material wellbeing (related to income, poverty, material goods), health and safety, educational wellbeing, family and peer relationships, behaviours and risks, and subjective wellbeing (how the child sees his or her self). The press had a field day when the report was published,2 because the United States and the United Kingdom were in the bottom five countries for five of the dimensions. The UK ranked 12th in health and the US ranked 12th in education; questions were rightly asked about how this could happen and what the government was going to do about it.
In this week's BMJ, Pickett and Wilkinson3 attempt to explain the results of the Unicef report by combining the measures of wellbeing of children with national data on income. They selected three measures of income—income inequality (ratio of the top fifth of incomes to the lowest fifth); relative child poverty (the proportion of children living in households in which the income was less than 50% of the national median); and average income (gross national income per capita in 1999). Rather than taking national aggregates, with all types of children grouped together, the authors disaggregated the data; this approach revealed the profound impact of poverty—lower scores of wellbeing were seen right across the board for children in the lowest income groups.
Wellbeing—the state of being happy, healthy, and prosperous—comprises more than just health. Measuring health in the simplest sense, by measuring mortality and morbidity, falls short of what is needed to measure wellbeing. Measuring material poverty is simply not enough to measure inequality. As a result, measurements of both wellbeing and inequality have evolved rapidly in the past 10 years.
Several data sources are used to measure the health of children, but few take into account the many contexts in which children grow and develop, including their family and community environments. Databases that do this are relatively new—for example, the US national survey of children's health—which started in 2003.4 The Unicef report represents a welcome progression towards more complex tools that compile primary and secondary data to measure composite indexes of health and wellbeing. It adds depth and data to our understanding of what our children experience and confirms what we already know—that even the richest countries have poor children, and that these children do not fare so well on many counts.
Measures within populations or groups of people identify the differences and inequalities that occur, but they do not explain why they occur or recommend how they could be changed. Picket and Wilkinson's study goes some way to dealing with these matters by disaggregating data on wellbeing into their separate constituents and combining them with national data on incomes. The implications of the results on policy are clear—we must invest more money in children, especially those at the bottom of the pile. We have opportunities to affect development in childhood that will never occur again. The difficult question is—what kinds of investment work?
One suggested framework is a coordinated and integrated country-wide response that makes evidence based changes in social and economic policies; improves living and working conditions; and strengthens the health of communities and individuals, via social networks and effective healthcare interventions.5 A systematic review of health interventions that reduce inequalities recommended a framework comprising the systematic, intensive delivery of effective healthcare and improved access to health services, together with reminders to use these services. This should be achieved by a multidisciplinary approach, which ensures that needs are dealt with and peers are involved in the delivery of interventions.6
We may need to invest even before the child is born and to monitor outcomes for years. This makes practical sense, but it is a major challenge to prove that investment before birth is beneficial. Take the case of programmes of home visits by nurses to disadvantaged mothers during pregnancy and two years after birth. A large randomised controlled trial compared the effects of nurses visiting unmarried mothers of low socioeconomic status in New York State with standard care. In adolescence, the children in the intervention group had significantly less serious antisocial behaviour and use of drugs and alcohol.7 In contrast, a systematic review of similar interventions in the US (five studies) and Australia (one study) for mothers with alcohol or drug problems found no effect on meaningful health outcomes in the mother or child but the studies were limited in quality and in the outcomes measured.8 Systematic reviews of other home visiting programmes by a nurse or professionally supervised lay person that target disadvantaged teenage mothers have provided limited evidence of a positive effect on quality of parenting and child development outcomes9; other reviews, however, have found that parenting programmes targeted at teenage parents resulted in improved psychosocial outcomes for the parent and child.10 Another review found that parenting programmes delivered to disadvantaged adult mothers showed no evidence of benefit,11 but this lack of effect may have resulted from a failure to assess the mothers' needs and provide tailored interventions.
Regardless of the inconsistent evidence, we know enough to say that inequalities affect child wellbeing and that poverty kills as effectively as any disease. We need to get better at identifying the programmes that work and much better at getting governments to invest in the wellbeing of children. The debate will hot up in 2008, when the World Health Organization Commission on the Social Determinants of Health will report.12
This article was posted on bmj.com on 16 November 2007
Competing interests: None declared.
Provenance and peer review: Commissioned; not externally peer reviewed.