The world health report

Chapter 4


Risks to health and socioeconomic status

The greatest burden of health risks is very often borne by the disadvantaged in our societies. The vast majority of threats to health are more commonly found among poor people, in people with little formal education, and those with lowly occupations. These risks cluster and they accumulate over time. In attempting to reduce risks to health, the focus of WHO and many other international organizations and governments is on trying to redress this imbalance -- by directly tackling poverty, by concentrating on the risks to health amongst the impoverished, or by improving population health and hence overall economic growth (1). An important component of the strategy is first to assess how much more prevalent risks are among the disadvantaged. While this provides information relevant to the targeting of interventions, it should be borne in mind that poverty and socioeconomic status are also of themselves key determinants of health status. This report seeks to shed further light on the mechanisms through which poverty acts, by assessing the distribution of risk factors by poverty levels.

Unfortunately, data are particularly scanty where they are required most -- in the poorest countries of the world. Nonetheless, this report attempts to stratify global levels of selected risks by levels of absolute income poverty (US$ 2 per day), as well as by age, sex and region. These analyses were conducted using individual-level data, not just comparisons of regional characteristics. The mapping of risk factors by poverty was conducted for:

  • childhood protein--energy malnutrition;
  • water and sanitation;
  • lack of breastfeeding;
  • unsafe sex;
  • alcohol;
  • tobacco;
  • overweight;
  • indoor air pollution;
  • urban air pollution.

In addition, available research findings are summarized on the links between poverty and high blood pressure, cholesterol, physical inactivity, exposure to lead, and use of illicit drugs.

Rates of poverty across the world

Approximately one-fifth of the world's population live on less than US$ 1 per day and nearly a half live on less than US$ 2 per day. Of the 14 world subregions (derived by dividing the six WHO regions into mortality strata -- see the List of Member States by WHO Region and mortality stratum) three (EUR-A, AMR-A and WPR-A) had negligible levels of absolute poverty and were excluded from analyses. In the EMR-B subregion, 9% of people live on less than $2 per day (2% less than $1 per day), but the estimates for this subregion were based on sparse data. There were, however, more data supporting estimates for the remaining 10 subregions, where the corresponding percentages ranged from 18% (3%) for EUR-B to 85% (42%) for SEAR-D and 78% (56%) for AFR-D.

Relationships between risk factor levels and poverty

For all subregions, there was a strong gradient of increasing child underweight with increasing absolute poverty (see Figure 4.1). The strength of the association varies little across regions, people living on less than $1 per day generally being at two- to three-fold higher relative risk compared with people living on more than $2 per day.

Unsafe water and sanitation, and indoor air pollution are also strongly associated with absolute poverty. For unsafe water and sanitation, the relative risks for those in households with an income of less than $1 per day, as compared to households with an income greater than $2 per day ranged from 1.7 (WPR-B) to 15.1 (EMR-D), with considerable variation between regions. For the association between indoor air pollution and poverty, there is considerable variation between subregions in the average level and in the relative differences within subregions. In the subregions of Africa, there is both a high prevalence of exposure to indoor air pollution and little relative difference between the impoverished and non-impoverished.

The associations of poverty with tobacco and alcohol consumption, lack of breastfeeding, and unsafe sex (unprotected sex with non-marital partner) are weaker and more variable between subregions. There is considerable variation between subregions in tobacco consumption, and a relatively weak association, within subregions, of tobacco consumption with individual-level poverty. Similarly, there is a more marked variation in alcohol consumption between WHO regions than within WHO regions by individual-level absolute poverty. In none of the subregions analysed was there a suggestion of increased alcohol consumption among the more impoverished. But in two subregions, AFR-E (South Africa data only) and AMR-B (Panama data only), impoverished people had approximately half the alcohol consumption of non-impoverished people. However, these results were based on household survey data recording expenditure on alcohol (not consumption) that may not have fully captured individual consumption and consumption of non-manufactured sources, such as alcohol distilled locally. Findings were also consistent with the higher socioeconomic groups in the developing world having more adverse lipid profiles, high blood pressure and overweight than the poor. However, if the trends seen in the industrialized world are repeated, these patterns will reverse with increasing economic development. These cross-sectional analyses were consistent with differing stages of progression of tobacco, obesity and other key noncommunicable disease determinants in poorer regions of the world as they undergo economic development. For example, obesity and tobacco consumption are initially found among the non-impoverished within regions, and later these risks are given up by the non-impoverished but taken up among the impoverished. These findings were consistent with regions being at different stages of such a transition. In the absence of major public health initiatives, these risk factors are likely to become increasingly concentrated among poor people in the poorer regions of the world. Public health action is required now to prevent this progression.

Potential impact on risk factor levels of shifting poverty distributions

In addition to estimating the associations of risk factor prevalence with poverty, population impact fractions of poverty on the risk factors were estimated. If people living on less than $2 per day had the same risk factor prevalence as people living on more than $2 per day, then protein--energy malnutrition, indoor air pollution and unimproved water and sanitation would be reduced by approximately 37%, 50% and 51%, respectively (see Table 4.1). These total population impact fractions would be reduced to 23%, 21% and 36% if the impoverished had the same risk factor prevalence as people living on exactly $2 per day.

Other risks present a more variable pattern, although data gaps particularly limit certainty of conclusions. Nonetheless, these analyses suggest that the prevalence of alcohol consumption and being overweight would increase by approximately 20% to 60% in Africa overall if prevalence among the poor matched those amongst the better-off. The population impact fractions for breastfeeding, unsafe sex and tobacco were more moderate, and even varied in direction across subregions.