One-third of U.S. labs have stopped using race-based equations to diagnose kidney disease

Even as a child, La’Tonzia Adams was interested in diagnosing disease. One day, when she noticed a bump on her chest, she decided to look up “chicken pox” in the Webster’s dictionary at her grandmother’s house to figure out if her symptoms matched the illness.

“I remember I scratched it, and it was watery,” said Adams. “When I looked up the chicken pox, I remember seeing ‘watery filled vesicle.’ I didn’t know what ‘vesicle’ was, so I looked up ‘vesicle’ in the book,” she added with a laugh.

“When my mom came to my grandmother’s, I said, ‘Mother, I believe I have chicken pox.’” Her mother argued with her, but Adams turned out to be right about her self-diagnosis.


Adams is now a staff pathologist and director at the Veterans Affairs Portland Health Care System in Oregon. But even though her job is to “have knowledge of disease from head to toe, inside and out,” she couldn’t diagnose her own father’s end-stage renal failure — partly because the metrics wouldn’t have allowed her to.

For decades, health care providers have diagnosed kidney disease with blood tests that use an equation for estimating glomerular filtration rate, or eGFR — a number that acts a proxy for how much blood the kidneys clean every minute. Until recently, the eGFR equation has included a coefficient for race to “correct” for different levels of creatinine (a waste product released from muscles) in African Americans. This adjustment was based on the incorrect assumption that Black people have higher muscle mass. Black people were therefore assumed to have higher baseline eGFR levels, which could in turn mark them at a less advanced stage of kidney disease even when their numbers are the same as a non-Black person’s.


But in 2021, an expert panel recommended a retooled eGFR equation that leaves out race and was recalculated using data from a diverse group of subjects. On Tuesday, the College of American Pathologists (CAP) published in the Journal of the American Medical Association the results of a survey that explores how many laboratories across the country have adopted the new equation, endorsed by the National Kidney Foundation and the American Society of Nephrology as well as nine pathology and laboratory societies.

The survey, sent to the 6,317 labs that participate in the CAP proficiency testing program, provided the rare chance to track how a recommendation is being adopted, said Jonathan Genzen, principal author of the study and chair of the College of American Pathologists Clinical Chemistry Committee. He was pleasantly surprised with the results. “Literally within six months of this new equation being announced, a third of laboratories surveyed had already adopted the equation,” he said.

Of the labs that hadn’t switched, 32% planned on switching to the new equation by the end of the year. The authors plan to send out a repeat survey next year to see if there’s more adoption going forward.

The survey covered about 11% of the eligible labs in the U.S., according to Genzen, and around 30% of U.S. hospital labs. However, as with any survey study, it’s unclear if this sample is representative, said John Arthur, chief of nephrology at the University of Arkansas for Medical Sciences, who was not involved with the survey. The people most likely to respond to the survey might be those who already knew of and already had switched to the new equation, thus making the true percentage of adoptees lower.

Arthur was also surprised at the finding that 58% of respondents who hadn’t yet switched to the new equation weren’t sure when their lab was going to adopt the new formula, given his own positive exchanges about the new equation with attendees at the recent annual American Society of Nephrology meeting. He questioned whether the survey respondents were the relevant people in each lab.

The eGFR metric is a key part of diagnosing chronic kidney disease. Directly measuring GFR is complicated and time-consuming, requiring many blood or urine samples over time. Instead, clinicians can estimate GFRs by taking a patient’s creatinine levels and running it through an equation that accounts for the different levels of creatinine produced by people of different ages, races, and sexes. (Finding an equation that does not rely on sex is the “next frontier,” said Adams, who was not involved with the survey.)

But a growing movement in medicine argues that race-based equations aren’t accurate — and they can be actively harmful. The race coefficient for eGFR has resulted in Black people having to wait an estimated extra two years before being eligible for a kidney transplant, or declining so much before treatment that they are no longer eligible for a transplant. Race-based equations are also complicated in that they can make the way that a bi- or multiracial person chooses to identify a key component in their care.

“Part of the problem with including race is that race is a social construct. It’s not really a biological variable,” said Arthur.

GFR estimates based on creatinine are also imperfect to begin with, said Arthur. A person’s creatinine levels depend not only on muscle mass but on meat consumption, how much one exercises, if the patient is underweight, and how the patient’s body clears creatinine through mechanisms other than glomerular filtration. To quantify this effect, Genzen pointed out a study showing that the person-to-person variability of eGFR is several times larger than the difference caused by using the race coefficient, which means that the “correction” for Black patients translates to a systemic bias from an eGFR number that says their condition is better than it is.

Revising the equation is “about not generalizing what somebody’s results would be expected to be based on their self-disclosed race, but rather based on what their individual results actually are, which is far more important,” said Genzen.

In a video for CAP, for which she is also a spokesperson, Adams recommends that patients check if the lab that processed their results is using the non-race-based equation. Which equation the lab used should be noted on the lab report — the most updated formula is called the “eGFR 2021 CKD EPI creatinine equation.” If there’s any confusion, patients can ask their health care provider, who can find out.

Such options weren’t available to Adams’ father. By the time physicians diagnosed her father, who had diabetes, with renal failure, he wasn’t strong enough to go through a transplant. Though Adams was prepared to donate a kidney, her father died in 2001.

“I’m sure they looked at [my dad’s] eGFR numbers,” said Adams. “But because they were falsely elevated, it didn’t probably raise it to the level that it should have been, as it would be today.”

Source: STAT