Income inequality and population health: A review and explanation of the evidence
Introduction
Whether or not the extent of income inequality in a society is a determinant of population health remains a controversial issue despite a large body of research. Although the findings of a substantial majority of studies suggest that more egalitarian societies do have better health and longevity (Lynch, Smith, & Harper, 2004a; Subramanian & Kawachi, 2004), a minority conclude otherwise and several authorities remain skeptical as to whether inequality has any implications for population health (Deaton, 2003; Lynch et al., 2004a). To gain a clearer understanding of the evidence and the nature of the disagreement, we decided to review all the research reports published in peer reviewed journals, and then to see if we could arrive at an interpretation of them which made sense of both the supportive and unsupportive findings.
Section snippets
The review
We compiled a list of 155 published peer reviewed reports of research on the relation between income distribution and measures of population health. This is much the most comprehensive list of studies yet compiled: as well as containing all the eligible studies listed in three previous reviews of parts of the literature (Hsieh & Pugh, 1993; Lynch et al., 2004a; Subramanian & Kawachi, 2004), we also found 37 additional papers either by using electronic searches or through informal contacts.
Findings
Table 1 provides a summary of the 168 analyses according to classification and the type of area over which inequality was measured. Table 2 lists all the analyses included according to their classification. A tally of numbers showed 87 wholly supportive analyses, 44 partially supportive, and 37 unsupportive. Almost three-quarters of all analyses were classified as either wholly or partially supportive. Of all analyses classified as wholly supportive or unsupportive, 70 per cent were wholly
The size of area
Table 1 shows the per centage of analyses classified as either wholly supportive or unsupportive according to whether they were international analyses using data for whole countries, whether their data were for large subnational areas such as states, regions and metropolitan areas, or whether they were for smaller units such as counties, census tracts or parishes. The proportion of analyses classified as wholly supportive falls from 83 per cent (of all wholly supportive or unsupportive) in the
Discussion
Taking account of the size of the area and the use of control variables reveals a high degree of consistency in the research findings. Thus, if we confine our attention to the 128 analyses which use data for areas the size of metropolitan areas or larger, only 23 fail to find some support for the hypothesis. If we were to reclassify analyses on the basis of results before the use of potentially problematic control variables (including individual income in multilevel models), then only eight (6
Mechanism
Low social status and the quality of the social environment are both known to affect health (Berkman & Kawachi, 2000; Marmot & Wilkinson, 1999). Not only are more unequal societies likely to have a bigger problem of low social status, but there is now substantial evidence to suggest that inequality is socially corrosive, leading to more violence, lower levels of trust, and lower social capital (Wilkinson, 2005). Psychosocial factors, many of which are associated with low social status, are
Conclusions
Our interpretation of 168 analyses of the relationship between income inequality and health is that income distribution is related to health where it serves as a measure of the scale of social class differences in a society. In small areas, where income inequality is unlikely to reflect the degree of social stratification in the wider society, it is—as Table 1 shows—less likely to be related to health. The overwhelmingly positive evidence from studies of larger areas suggests that this
References (178)
- et al.
Metropolitan area income inequality and self-rated health
Social Science & Medicine
(2002) - et al.
The effects of regional characteristics on alcohol-related mortality
Social Science & Medicine
(2004) - et al.
Socioeconomic factors, material inequalities, and perceived control in self-rated health
Social Science & Medicine
(2000) - et al.
Mortality, inequality and race in American cities and states
Social Science & Medicine
(2003) - et al.
A multilevel analysis of income inequality and cardiovascular disease risk factors
Social Science & Medicine
(2000) Letting the Gini out of the bottle? Challenges facing the relative income hypothesis
Social Science & Medicine
(2002)- et al.
Contributions of social context to inequalities in years of life lost to heart disease in Texas, USA
Social Science & Medicine
(2003) - et al.
Income distribution and risk of fatal drug overdose in New York City neighborhoods
Drug and Alcohol Dependence
(2003) - et al.
Income inequality and pregnancy spacing
Social Science & Medicine
(2004) - et al.
Teen births, income inequality, and social capital
Health and Place
(2002)