Does SES Explain More of the Black/White Health Gap than We Thought? Revisiting Our Approach towards Understanding Racial Disparities in Health

D. Phuong Do, University of South Carolina
Reanne Frank, Ohio State University
Brian K. Finch, San Diego State University

The inability of social factors to completely explain U.S. black/white health disparities has resulted in inferences of genetic predispositions. We contend that the persistent residual variation may be the result of methodological problems which include relying on cross-sectional data, limited set of SES factors, and inadequate covariate balance. Using data from the 1997-2007 Panel Study of Income Dynamics, we adjust for an extensive list of socioeconomic factors including wealth and long-term measures of family income and neighborhood poverty and apply a propensity score adjustment strategy to examine the black/white disparity in self-rated health. Compared to conventional regression estimates that resulted in a significant unexplained racial health gap, propensity score adjustment accounted for the entire black/white health disparity. Results suggest that previous studies may have inadequately adjusted for differences in SES across race and that social factors be returned to their place of explanatory dominance in racial health disparities studies.