Race, Quality of Care and Patient Outcomes: What Can We Learn from the Department of Veterans Affairs?

Abstract

The large and persistent differences in health outcomes between African Americans and whites have sometimes been attributed to differences in the quality of the facilities in which they receive care. A large body of literature documents systematically worse performance by facilities which serve a larger share of black patients. However, these estimates may be biased by unobserved variables which correlate with the racial profile of the clinic and also affect outcomes. This paper uses a unique dataset from the Department of Veterans’ Affairs system to examine the relationship between the quality of health care and the racial profile of the patient population in an equal-access health care system. An instrumental variable is used to overcome the omitted variable bias in the Ordinary Least Squares (OLS) estimates. Both the OLS and IV estimates show that increasing the proportion of visits by African–American patients results in improved outcomes for all patients treated at the clinic. The OLS results significantly underestimate the effect of increasing the proportion of black patients. Holding the overall clinic load constant and adding one hundred outpatient visits by African Americans in a year would result in a 0.5 percentage point decrease in 1-year mortality for the average patient in the clinic. Thus, contrary to results from the private health care sector, increasing clinic racial integration in an equal access system would benefit all patients.