Oaxaca-Blinder Health Decomposition

The Oaxaca-Blinder decomposition partitions the mean difference in a health outcome between two groups into a portion explained by differences in their measured characteristics and a residual, unexplained portion attributed to differences in how those characteristics translate into health. Developed independently by Ronald Oaxaca (1973) and Alan Blinder (1973) to study labor-market wage gaps, the method was imported into social epidemiology to quantify, for example, how much of a Black-White, urban-rural, or rich-poor gap in self-rated health, BMI, hypertension, or mortality is accounted for by differences in socioeconomic exposures versus differences in returns to those exposures. Group-specific regressions are estimated, the gap in fitted means is written as a function of mean covariates and coefficients, and that gap is algebraically split into an explained (composition) component and an unexplained (coefficient) component, each of which can be further decomposed variable by variable.