Stats about Black infant mortality demand scrutiny
Last month, when the U.S. Supreme Court decided the case on affirmative action in college admissions, Justice Ketanji Brown Jackson issued a stinging dissent that included a litany of harms and injustices that Black citizens have had to endure from the age of slavery until the present.
I have no quarrel with anything in her dissent, save one accusation that she cited: “For high risk Black newborns, having a Black physician more than doubles the likelihood that the baby will live, and not die.”
The basis for Jackson’s statistic is a 2020 article, “Physician-patient racial concordance and disparities in birthing mortality for newborns,” in the Proceedings of the National Academy of Sciences, or PNAS, a prestigious scientific journal. The data appears reliable, although the report lacks medical details; the authors were not physicians.
But it is the interpretation of the data that is in question, and in my opinion, the justice put undue emphasis on this passage from the article, especially the second clause: “These results underscore the need for research into drivers of differences between high- and low-performing physicians, and why Black physicians systemically outperform their colleagues when caring for Black newborns.”
Frankly, from the standpoint of cause and effect, I doubt that the primary reason for this disparity in newborn deaths is due to the race of the physicians. I do not believe that Black physicians outperformed white physicians at a level so dramatic that mortality was halved. Not because I think white physicians couldn’t underperform their Black colleagues, but rather, I simply don’t think there is a measurable difference between the performance of Black and white physicians.
Along with this, a difference of cutting mortality in half owing to physician performance would have virtually no precedent in any other situation to my knowledge (especially when surgery, in which individual physician skill is a greater factor, is not an issue).
I am not making any judgment on the merits of the Supreme Court decision. Moreover, I fully support educating and training more Black physicians, which I did throughout my career. Some of those physicians are leaders in their communities today.
To return to the topic at hand, a doubling of survival is an incredibly huge difference for any medical situation, in this case an extra death for every 1,000 births. That is a very large number to attribute to any single intervention, let alone physician quality, and there are virtually no single factors that could account for such a large number. It is almost certainly primarily due to things other than the quality of the physicians. And probably several things — underlying diseases, prematurity, preexisting health of the newborns and their mothers, quality of the hospitals (nurses and other staff ), and other things we may not even be aware of. That would need a much deeper dive.
It is important to note the PNAS study was not randomized — that means many variables are not the same in cases with Black and with white physicians. In a randomized study, variables are assumed to be similar in different groups.
How exactly would such a doubling of mortality work? Is there something we can point to? Are white doctors denying Black newborns access to intensive care or ventilators or antibiotics? If such a huge effect was due to the quality of the physicians, it would be apparent to almost everyone who works with patients. Someone would be able to identify and explain how — and the underlying failures would be easy to see. If it were true that doctors were causing a doubling of mortality, people would be coming forward on a daily basis to shout it, such as the doctors did this or didn’t do that. My work with hospital mortality for 30 years strongly suggests that such an appalling physician effect would be obvious to nurses, hospitals, patients and families.
The implication that white physicians are biased — consciously or unconsciously — and are systematically giving substandard care to Black infants is just not credible, not to any extent in 2023. This belief runs completely counter to my experience and that of every physician I have asked about it. I won’t say this wasn’t true in 1923. I also won’t say you couldn’t pick out an occasional bigoted physician today. But most physicians, Black and white, are colorblind or relatively so when it comes to care. And this is probably more true of obstetricians, neonatologists and pediatricians, whom I believe are among the most sympathetic physicians in general.
Black people are unquestionably subject to higher health risks. Some of that may be due to racism and some not. Historically, they did suffer worse care. But broad claims of systemic racism compromising care to the extent the PNAS study and the justice are invoking now demand much higher levels of investigation, data analysis and ultimately proof than currently provided. Given such an overarching claim, medical and data scientists should now deliver that scrutiny.
To let such an accusation stand unchallenged does a profound disservice to doctors and to their patients, who have difficulty today knowing what’s true and what isn’t.
To do otherwise is unscientific and an insult to the profession.