Income inequality and self-rated health in Stockholm, Sweden: A test of the 'income inequality hypothesis' on two levels of aggregation

The number of studies analysing income inequality and health are voluminous. However, when empirically testing the income inequality hypothesis, the level of aggregation could be crucial for whether we find an association or not and for the mechanisms we believe are active. This study hence investigates: 1) the two-year lagged effect by income inequality on health at two levels of aggregation; municipalities and neighbourhoods in Sweden; 2) whether spending on social goods accounts for the association between income inequality and health; 3) the effect by income inequality among the affluent and the disadvantaged in municipalities and neighbourhoods, respectively. The empirical data is based on a Swedish public health survey in 2002 and includes residents of Stockholm aged 18-84 years. The sample consists of 28,092 individuals nested within 22 municipalities and 709 neighbourhoods in the county of Stockholm with a non-response rate of 37 percent. A total population register (HSIA) is further used for the construction of contextual-level indicators. Primary method used is multi-level logistic regression. The findings indicate a moderate effect by high and very high income inequality on self-rated poor health at the municipality-level. The association, however, ceases after adjustment for spending on social goods. No detrimental effect by income inequality on self-rated health at the neighbourhood-level is found. The results further suggest that poor individuals residing in high inequality neighbourhoods do not have poorer health than those residing in low inequality contexts while high inequality is most deleterious for poor individuals at the municipality-level. In sum, the findings suggest that reduced spending on social goods could account for the association between income inequality and health at the municipality-level. The contrasting findings at the neighbourhood- and municipality-level indicate that it is important to consider the level of aggregation when studying health effects by income inequality. (C) 2012 Elsevier Ltd. All rights reserved.