Given growing appreciation of how race/ethnicity is a social, not biological, construct, some epidemiologists are proposing that studies omit data on "race" and instead collect better socioeconomic data. This suggestion, however, ignores a growing body of evidence on how noneconomic as well as economic aspects of racial discrimination are embodied and harm health across the lifecourse. Developing a critical epidemiology of social inequalities in health will, at the very least, require incorporating thoughtful measures of race/ethnicity and social class in epidemiological studies and public health surveillance systems.