Why did it take four years to debunk the black baby study?
A study published this week in a prestigious journal, Proceedings of the National Academy of Sciences, makes a claim that’s almost unheard of on the pages of leading medical journals: systemic racism and implicit bias are not the self-evident explanations for a pervasive racial disparity. To be precise: black newborns aren’t dying at higher rates when they’re treated by white doctors.
The study, conducted by a Harvard economist and a Manhattan Institute researcher, purports to debunk a widely circulated 2020 study, also published in PNAS, which concluded that black newborns attended by a white physician suffer a “mortality penalty” and are twice as likely to die. That study garnered incriminating headlines in USA Today, CNN, Science News, NPR, and The Washington Post. It was also so influential that it was cited by Supreme Court Justice Brown Jackson in the high court’s 2023 affirmative action case, in which the American Medical Association and 44 other parties declared in their amicus brief: “For high-risk Black newborns, having a Black physician is tantamount to a miracle drug.”
How could two teams of researchers look at the same data — 1.8 million childbirths in Florida between 1992 and 2015 — and reach diametrically opposite conclusions?
This time around, the researchers added one key variable that the 2020 researchers had overlooked — low birth weights — and the whole thing collapsed. The research design contained a fatal flaw, overlooking the fact that severely underweight babies, who have very high mortality rates to start with, tend to be treated by white doctors. Physicians who handle the most serious medical cases tend to see higher death rates.
