A direct regression approach to decomposing socioeconomic inequality of health.

This paper presents a new regression-based decomposition of socioeconomic inequality of health that is more direct than other approaches. The method can be applied to both rank-dependent and level-dependent indicators of inequality. The response variable of our regression model is a simple reformulation of the measure of overall performance of an individual in the health and socioeconomic domains. Regression results are described in terms of marginal effects of the explanatory variables, but also in terms of their logworths or importance values. We illustrate our method, and compare it with alternatives, using Australian health and income data.