The Autopsy / Case File · Disparities
The Coefficient
Black Americans develop kidney failure at more than three times the rate of white Americans. For roughly two decades, the standard lab formula used to estimate how well a patient's kidneys were working quietly made that gap worse — by building in a number that made a Black patient's kidneys look better on paper than they actually were.
United States Renal Data System figures put Black Americans' end-stage renal disease incidence at more than three times the rate for white Americans, despite Black Americans making up roughly 13 percent of the U.S. population. Part of that gap traces to a specific, quantifiable mechanism: from 1999 through 2021, the two most widely used formulas for estimating glomerular filtration rate — the standard measure of kidney function, derived from a routine blood test — applied an automatic upward adjustment of roughly 16 to 18 percent to any patient marked as Black, an assumption rooted in an unverified claim about average muscle mass. Stanford researchers calculated that removing the adjustment would let 3.3 million more Black Americans cross the diagnostic threshold for chronic kidney disease stage 3, qualify roughly 300,000 more for a nephrology referral, and make roughly 31,000 more eligible for transplant evaluation. The equation was changed in 2021. A follow-up study two years later found no measurable change in nephrology referral or visit rates for Black patients.
Join the discussion →
Kidney function is measured almost entirely through inference rather than direct observation. A doctor doesn't typically look at a kidney to assess how well it's filtering; they order a blood test for creatinine, a waste product the kidneys clear, and plug the result into a formula alongside the patient's age, sex, and — for two decades — race, to produce an estimated glomerular filtration rate, or eGFR, the number that determines whether a patient gets flagged for chronic kidney disease, referred to a specialist, or placed on a transplant waiting list. That formula's race variable is the mechanism at the center of this story, and it is exactly the kind of quiet, upstream design choice this series keeps finding underneath a mortality disparity that looks, from a distance, like it must have a purely biological or purely economic explanation.
A Number Built on an Unverified Assumption
The race adjustment embedded in the MDRD and CKD-EPI equations — the two formulas that dominated American clinical practice for roughly two decades — increased a Black patient's calculated eGFR by approximately 18 percent under MDRD and 16 percent under CKD-EPI, relative to what the same creatinine value would produce for a patient not marked as Black. The justification, when the adjustment was introduced around 1999, rested on an assumption that Black patients have higher average muscle mass, and therefore generate more creatinine independent of actual kidney function — an assumption research has since found unsupported as a basis for a blanket, race-wide clinical adjustment. In practice, the effect of that adjustment was to make many Black patients' kidney function look better than it actually was, at exactly the threshold where a lower true eGFR would have triggered a specialist referral, a transplant workup, or a diagnosis of chronic kidney disease at an earlier, more treatable stage.
What the Coefficient Cost, Counted
Stanford researchers translated the equation's effect into concrete clinical-access numbers, and the scale is what makes the mechanism more than an academic footnote. Removing the race coefficient from the eGFR calculation, applied retrospectively to the population, would have moved an estimated 3.3 million Black Americans across the diagnostic line into chronic kidney disease stage 3 — meaning 3.3 million people whose kidney function, measured without the adjustment, already qualified for a CKD diagnosis they weren't given under the adjusted formula. Of that population, roughly 300,000 more would have qualified for a nephrology referral, the step that connects a patient to specialist monitoring and treatment planning before kidney failure becomes imminent, and roughly 31,000 more would have become eligible for transplant evaluation and waitlist placement — the step that determines whether a patient facing kidney failure has any realistic path to a transplant before dialysis becomes the only option. Every one of those numbers represents a gap between the care a patient's actual physiology warranted and the care the formula told their doctor they needed, compounding for as long as the adjusted number stayed on the chart.
The formula made a Black patient's kidneys look healthier on paper, at exactly the threshold where the truth would have triggered a referral.
Layered on Top of an Already-Documented Referral Gap
The eGFR coefficient didn't create racial disparity in kidney care from nothing — it compounded a referral gap researchers had already documented independent of the formula. Older research published in the American Journal of Kidney Diseases found that delayed referral to a kidney specialist was almost six times more likely among nonwhite patients than white patients, a disparity tied to differences in how symptoms were interpreted, how quickly primary care physicians escalated concerning lab results, and broader patterns of unequal access to specialist care. The race-adjusted eGFR formula sat directly on top of that pre-existing gap: a physician already less likely to refer a Black patient promptly was, for two decades, also working from a lab number that made that same patient's kidney function look artificially reassuring — two independent mechanisms pointing in the same direction, at the same patients, at the same decision point. Each mechanism alone would have been enough to delay the moment a patient reached a nephrologist's office; together, layered on the same visit, they moved that moment further out than either would have on its own, for a population already documented to start renal replacement therapy later and sicker than white patients with an equivalent underlying disease course.
Where the Assumption Came From
The muscle-mass justification behind the original coefficient traces back to the study populations used to derive the MDRD and CKD-EPI formulas in the first place — cohorts in which Black participants, on average, measured higher serum creatinine at a given directly-measured GFR than white participants did, a real statistical pattern in those specific datasets. The equations' authors built a correction factor to account for it, reasoning that average muscle mass differences explained the gap. What that approach treated as a fixed, biological, race-wide constant was, critics later argued, more plausibly a marker of the specific study cohorts' composition — diet, body composition, and other measurable factors correlated with race in those samples without being caused by race itself — collapsed into a single multiplier and then applied to every Black patient nationally, regardless of that individual patient's actual muscle mass, diet, or any other trait the original correlation was standing in for. A statistical pattern observed in one set of research cohorts became, for two decades, a clinical rule applied to millions of people who were never part of the studies that produced it.
The Equation Changed. The Referrals Didn't.
In 2021, the National Kidney Foundation and the American Society of Nephrology's joint task force recommended abandoning the race-based coefficient in favor of a new, race-neutral CKD-EPI 2021 equation, and hospitals and laboratories across the country began adopting it over the following two years. That change is, on its face, exactly the kind of correction a documented measurement bias should receive. But a Stanford-affiliated follow-up study examining the two years after implementation found that removing the race adjustment did not produce a corresponding change in nephrology referral or visit rates for Black or African American patients — the diagnostic number moved, and the clinical behavior downstream of that number largely didn't. That finding doesn't undo the case for having removed the coefficient in the first place; a formula that systematically overstated kidney function for one racial group needed to change regardless of what happened next. It does mean the disparity in referral and specialist access that the coefficient contributed to wasn't simply a downstream artifact of a bad number waiting to be corrected — the referral gap researchers had already documented before the equation changed appears to have persisted on its own, through habits, workflows, and access patterns the formula never fully explained.
And Then You're Dead
A kidney doesn't know what race is typed into the lab software calculating how well it's working. For two decades, the formula did — and adjusted the answer accordingly, for millions of patients, at exactly the moment an accurate number would have gotten them to a specialist sooner. The formula has been fixed. The referral pattern the formula helped entrench largely hasn't, which is its own kind of answer about how much of this gap the coefficient was ever really responsible for — and about how much further behind it the rest of the system still is, three-and-a-half years after the number on the chart stopped being the excuse.
That's it. That's the whole thing.
Sources
- Black Americans developing end-stage renal disease at more than three times the rate of white Americans, despite being roughly 13 percent of the U.S. population — National Institute of Diabetes and Digestive and Kidney Diseases, United States Renal Data System (USRDS)
- The MDRD and CKD-EPI equations' race-based adjustment increasing calculated eGFR by approximately 18 percent and 16 percent respectively for patients marked as Black, based on an unverified assumption about average muscle mass — peer-reviewed research on race-based eGFR calculation, cited in New England Journal of Medicine and National Kidney Foundation analyses
- The estimate that removing the race coefficient would move 3.3 million Black Americans across the CKD stage 3 diagnostic threshold, qualify 300,000 more for nephrology referral, and make 31,000 more eligible for transplant evaluation and waitlisting — Stanford University Health Policy research on race adjustment in kidney function estimation
- Delayed referral to a kidney specialist being almost six times more likely among nonwhite patients than white patients — American Journal of Kidney Diseases, research on delayed referral of Black, Hispanic, and older patients with chronic renal failure
- The 2021 National Kidney Foundation/American Society of Nephrology Task Force recommendation adopting the race-neutral CKD-EPI 2021 eGFR equation, and its cited rationale — National Kidney Foundation and American Society of Nephrology, joint task force report, 2021
- The Stanford-affiliated follow-up study finding no significant change in nephrology referrals or visits for Black or African American patients two years after the race-free eGFR equation was implemented — Stanford University, follow-up analysis of CKD-EPI 2021 equation adoption
- Photo: "Nurse checks blood pressure," Linda Bartlett, National Cancer Institute, public domain, via Wikimedia Commons
Comments
Loading…