Using Race to Predict Risk: How Accurate Are Hospital Data?

eAlert e8a1508e-7050-4993-9be6-0be839c6fe83

Medical professionals often take into account patients' race or ethnic background when diagnosing conditions or determining treatment options. For the condition known as hyperbilirubinemia — an acute and potentially devastating form of neonatal jaundice — hospitals use the race of the newborn's mother to predict risk, with blacks being at lowest risk for developing the condition.

A new study published in the journal Pediatrics, however, finds significant disparities between the race assigned to mothers of newborns by hospital staff and mothers' self-described race, potentially undermining efforts to identify and treat the condition.

The research team, led by the Commonwealth Fund's Anne C. Beal, M.D., reports that of the mothers documented as black in the medical record, 23 percent described themselves as being of two or more races, while only 70 percent described themselves as black. Of those mothers documented as white, only 64 percent actually identified themselves as white.

Such disparities, the authors say, point to a pressing need to establish rigorous standards for measuring race in health care settings.