Historically, there has been a large gap between African Americans and Whites in access to health care, but this gap was ostensibly lessened by the advent of Medicare and Medicaid for older adults in the mid 1960s. The extent to which older African Americans continue to receive less access to medical care as a result of economic inequalities, institutionalized forms of discrimination, and life-style factors remains a subject of policy debate. Empirical enquiry has produced inconsistent results. The purpose of this study is to test the same set of models of medical use using identically measured predictor variables in three nationally representative data sets of older Americans: 1984 Study of Aging (SOA); 1984 National Long-Term Care Survey (NLTC); and the 1987 National Medical Care Expenditure Survey (NMES). Multivariate logistic regression of use of physician and hospital services and Poisson regression of amount of service use identified inconsistent results in race differences across data sets, but consistent results in terms of the importance of health status and insurance as predictors of use and amount of use. The findings suggest that health status and financial resources may be more relevant areas for policy interventions than considerations related to race and ethnicity.