Abstract

PIP: The idea that the health of individuals depends on the characteristics of the society in which they live and on their own characteristics is important. The aggregate level relation between income inequality and population mortality has been examined by empirical works. However, if the individual level relation between risk of mortality and income is curvilinear, at least part of any association between population mortality and income inequality is artifactual in the sense that it could arise even if individual risk was due only to individual income and not to its distribution. Wolfson et al attempted to estimate how much of the variation in cross-sectional US state-level mortality could be due to the curvature of the relation between individual level mortality and income interacting with differences in the distribution of income within states. They concluded that the artifact is not the main reason for the frequently documented correlations between population mortality and income distribution. However, in the absence of any detailed information on the regressions, it is difficult to determine if the difference between actual and hypothetical mortality is significantly related to income equality. The individual risk of mortality is also affected by other individual characteristics, like climate or public health infrastructure. The authors suggest that investigations of the determinants of individual health test the effect of societal factors and that such testing requires both individual level and aggregate data.