Race and Ethnicity A Part of the Equation for Personalized Clinical Decision Making?

Two 48-year-old men, Jake and Earl, walk into a physician's office. They both have a total cholesterol of 200 mg/dL, a high-density lipoprotein level of 40 mg/dL, systolic blood pressure of 130 mm (despite antihypertension treatment), and no history of diabetes mellitus. Their physician applies the same primary prevention guidelines, and-despite their identical cardiac risk factors-recommends an aspirin and a statin for Earl but not Jake. These patients might be surprised to find their discrepant treatment derives from the fact that Jake is white and Earl is African American. Thus, according to the pooled cohort equations, the calculated 10-year risk of atherosclerotic cardiovascular disease is 4.3% for Jake (below the 7.5% American College of Cardiology/American Heart Association guideline threshold(1)) and 8.7% for Earl. The inclusion of race in the pooled cohort equations as a predictor of 10-year atherosclerotic cardiovascular disease risk has generated this race-based treatment difference. In this commentary, we consider the use of race/ethnicity in clinical prediction models and discuss the similarities (and differences) between racial profiling in healthcare and in other settings, such as law enforcement and the insurance market.