I have spent years running personal experiments on variables most people ignore: sleep staging, glucose spikes, HRV trends. I am comfortable sitting with incomplete data and making a provisional call. So when I looked at the question of whether your baby's survival depends on your doctor's race, I expected to find a clean signal. I did not find one. What I found instead was a system so badly optimized that the signal is buried under noise we created on purpose.
Mississippi's infant mortality rate hit 9.7 deaths per 1,000 live births in 2024, the highest since 2013. That is not a rounding error. That is 3,527 babies dead before their first birthday since 2014. Rep. Zakiya Summers said it plainly on the House floor in February: Black women are dying disproportionately. Sen. Michael McLendon is asking whether it is the water or the air. Both are right to be alarmed. Neither is asking the sharper question.
The Variable Nobody Is Measuring
The specific claim, that your baby survives or dies based on the race of the attending physician, does not have clean national data behind it. The research brief I pulled for this piece found no recent studies directly linking infant survival rates to doctor race. That absence is itself a data point. We are not measuring it. A system that does not track a variable cannot optimize around it.
What we do have: a 2020 study from the University of Florida found that Black newborns cared for by Black physicians had significantly lower mortality rates than those cared for by white physicians, particularly in low-intervention births. That is a single study. N=1 is not proof, and neither is N=1 study. But the mechanistic argument is not crazy. Concordance between patient and provider correlates with better communication, more trust, and faster escalation when something goes wrong. Those are real outcome drivers.
I will grant the skeptics this: individual provider race is a weak lever compared to systemic ones. Concordance matters less than whether there is a NICU within 50 miles. A 25-week preterm infant has less than 50% survival odds, and that number moves with access to steroids and monitoring, not with who is in the room. Structural access is the dominant variable.
What Mississippi's Review Panel Actually Tells Us
House Bill 1637 creates a Fetal and Infant Mortality Review Panel reporting annually starting December 2026. It will track demographics, causes, and resource allocation. That is a decent data collection protocol. What it will not track is provider race concordance. So we will get another year of aggregate disparity data with no actionable variable attached to it.
The optimization play here is not to wait for the panel's first report. Mississippi, and every state running similar mortality numbers, should add provider concordance to the data stack now. Not because it is the biggest lever, but because it costs nothing to measure and the signal from that Florida study is worth replicating at scale. You do not ignore a biomarker because it is inconvenient to collect.
My HRV tells me things my annual physical misses. The reason is simple: I am measuring continuously, not once a year. Infant mortality reviews that meet annually and exclude key variables are the annual physical of public health. They catch the obvious. They miss the pattern.
Mississippi's babies deserve continuous monitoring of every variable that might move the needle. Right now, the state is not even asking the question. That is the real mortality risk.
