Racial differences in health: not just black and white, but shades of gray

Annu Rev Public Health. 1996:17:411-48. doi: 10.1146/annurev.pu.17.050196.002211.


Explanations for racial/ethnic disparities in health are varied and complex. This paper reviews the literature to assess the extent to which current disparities are a consequence of racial differences in the social class composition of the US population. We focus this review on African Americans and examine studies that provide information on the effect of race on four outcome measures: infant mortality, hypertension, substance use, and mortality from all-causes. Twenty-three studies were identified that met criteria for inclusion in this review. As expected, most studies provide evidence that socioeconomic conditions are a major factor explaining racial differences in health. Findings, however, vary for the different health indices. Research in the area of substance abuse provides the most consistent evidence that socioeconomic conditions account for observed racial differences. In contrast, studies on infant mortality and hypertension provide a compelling case that the effects of socioeconomic status are important, but not sufficient to explain racial differences. Evidence on mortality from all-causes is equally divided between studies showing no significant race effect and those in which racial differences persist after adjusting for social class. The paper offers possible explanations for the seemingly divergent results and identifies conceptual and methodologic issues for future research seeking to disentangle the complex relations between race, social class, and health.

Publication types

  • Review

MeSH terms

  • Black People*
  • Black or African American / statistics & numerical data*
  • Cause of Death
  • Health Status*
  • Humans
  • Hypertension / ethnology
  • Infant
  • Infant Mortality
  • Infant, Newborn
  • Likelihood Functions
  • Social Class
  • Socioeconomic Factors
  • Substance-Related Disorders / ethnology
  • United States / epidemiology
  • White People / statistics & numerical data*