Alternatives to Race-Based Kidney Function Calculations

Race has long been used as a biological variable in health research, under the mistaken belief that racial categories correlate with genetic traits that account for population-level biological differences. However, we now know that more genetic variation exists within race categories than between them, and that race correlates poorly with the spectrum of biological variability that exists among human beings.

Accordingly, NIDDK-supported research is leading to a change in the way kidney disease is diagnosed and monitored by removing race as a variable from the equations used to estimate glomerular filtration rate (GFR). Estimated GFR remains a primary tool to assess kidney function and to classify the severity of kidney disease. Estimated GFR also helps determine prognosis and treatment, such as when hemodialysis or a transplant may be needed and how to optimize the dosage of certain drugs.

Because measuring someone’s GFR directly is expensive, difficult, and burdensome on the person being tested, GFR is normally estimated by using an inexpensive blood test to determine the concentration of a compound called creatinine. Because creatinine is synthesized at a constant rate by one’s muscles and filtered out of the blood by the kidneys, its concentration in the blood is strongly linked to kidney function. For several reasons, however, creatinine is not a perfect biomarker of a person’s actual kidney function. For example, its synthesis rate is determined by how much muscle a person has, and its blood concentration is also affected to a degree by how much meat they eat. As a result, a person’s real GFR might be a bit higher or lower than their estimated GFR, but the estimates are generally close.

However, the original study data used to develop estimated GFR calculations came overwhelmingly from participants of European descent. Subsequent work showed that the relationship between creatinine level and real GFR, on average, was the same for people from most other groups. Researchers discovered, though, that for unknown reasons creatinine levels tend to be slightly higher in Black study participants than in participants from other populations, at any given directly measured GFR. As a result, for many years estimated GFR calculations have taken into consideration whether the person being tested is “Black” or “non-Black.”

This practice is problematic. Race was created for social and political reasons, and thus has no biological basis. Indeed, race categories do not align with the continuum of human genetic and biological variability. For example, many individuals who identify as Black do not, in fact, have a higher creatinine to GFR ratio than is found in other groups. Therefore, applying the “correction factor” for Black race sometimes leads to overestimation of GFR, potentially aggravating the significant health disparities that exist in kidney health outcomes. For example, a person who identifies as Black could be erroneously excluded from receiving a kidney transplant because the equation overestimates their GFR, making it appear that their kidneys are more functional than they are. Further, the physiological reason why some Black people have a higher creatinine to true GFR ratio remains unknown, so there is no way to test for it. Thus, it is unclear who, exactly, does have a higher creatinine to true GFR ratio and thus should receive a correction factor for determination of estimated GFR.

Recent NIDDK-supported research from the Chronic Renal Insufficiency Cohort Study and the Chronic Kidney Disease Epidemiology Collaboration has sought to address these issues by identifying new, better methods for assessing kidney function. For example, one group investigated whether genetic ancestry analysis might be useful for helping determine who the correction factor should apply to. While the use of ancestry did improve accuracy at the population level, it is both impractical and not always applicable at the individual level. Other approaches that considered body composition (e.g., how muscular a person is) or urinary creatinine excretion rates marginally improved accuracy. Another group tested an alternative formula for estimating GFR from creatinine that corrects somewhat for age and sex, but that does not use race as a modifier. On average, this approach slightly underestimated GFR for participants who identified as Black, and slightly overestimated GFR for people who considered themselves non-Black.

Encouragingly, both studies also found that estimating GFR based on blood levels of a compound called cystatin C, which does not vary by a person’s race, could help improve the accuracy of kidney function tests. One of the studies found that the most accurate, least biased results were obtained using equations that utilize both markers—creatinine and cystatin C. Thus, NIDDK-supported research has informed recent recommendations to use both serum creatinine and cystatin C to estimate GFR in adults, when cystatin C is available. Using the combined serum creatinine-cystatin C equation is particularly important when the estimated GFR value is close to a critical decision point, such as when determining drug dosing or kidney transplant eligibility.

At present, however, laboratory and reimbursement infrastructure are not yet adequate to support routine ordering of cystatin C tests in clinical settings for all people for whom GFR should be more accurately assessed. Therefore, two leading kidney health advocacy groups—the American Society of Nephrology and the National Kidney Foundation—have called for measures to improve the availability of cystatin C testing, as well as more research to find still better approaches for assessing kidney health. In the meantime, they have called for adoption of the improved creatinine-only based GFR estimation method that uses age and sex—but not race—as modifiers.

Thus, NIDDK-supported research is improving the equitable and accurate assessment of kidney function. NIDDK remains committed to research that builds on that improvement and to the overarching goal of reducing disparities in kidney disease.