Medical Students Lead Effort to Remove Race from Kidney Function Estimates

Calls for change

There has been growing skepticism of the use of black race in kidney function estimation among nephrologists. In a viewpoint published in JAMA in June 2019, Nwamaka Eneanya, MD, MPH, and colleagues argue that using race as a variable may restrict access to care for some patients and interfere with transparency in patient care (3). Eneanya is assistant professor of medicine and epidemiology at the University of Pennsylvania’s Perelman School of Medicine in Philadelphia. Another perspective in the Clinical Journal of the American Society of Nephrology by Vanessa Grubbs, MD, MPH, links this practice to the troubling history of racism in medicine (4). Grubbs is associate professor in the division of nephrology at the University of California, San Francisco.

“It all traces back to this legacy of trying to put biology into race,” Grubbs said. “Instead, race is a social construct that has been used to justify atrocities against certain groups of people. It is an insidious thing that has been going on since the beginning of the country.”

Eneanya and her coauthors state in their viewpoint that equations that use race to estimate kidney function such as the Chronic Kidney Disease Epidemiology Collaboration equation (CKD-EPI) and the older Modification of Diet in Renal Disease Study (MDRD) equation were created using large cohorts of patients who underwent measurements of their true glomerular filtration rate. Black race was associated with slightly higher GFR even when patients had the same blood creatinine levels, which was attributed to higher average muscle mass. As a result, the CKD-EPI with race adjustment increased the eGFR of black patients by about 16%.

There is evidence that using race increases the statistical accuracy of kidney function estimates (5), according to Peter Reese, MD, associate professor of medicine at the Perelman School of Medicine and co-author of the JAMA viewpoint. However, the decision to use the idea of black race as a proxy for muscle mass was never openly debated and raises ethical concerns about perpetuating the concept that race is a biological rather than a social construct.

“To build it right into an equation that gets used many millions of times a day just reinforces the bogus concept,” Reese said.

Additionally, the way race is assigned in medicine is often problematic. For example, a clinician may assign race based on a patient’s skin tone or hair or by the way the patient chooses to identify, notes Eneanya. But this does not take into account ancestry.

“It just lumps all black people together in a way that it doesn’t matter if you have two black parents, or you’re biracial, or just your great grandparent was biracial and everybody else was white,” Grubbs said.

In fact, nephrologist Malika Mendu, MD, assistant medical director for quality and safety at Brigham and Women’s Hospital, noted that the use of race adjustments in eGFR were based on studies in a few hundred patients who were classified as black by researchers, which raises questions about how representative these patients are of individuals who self-identify as black. When Mendu and a colleague analyzed the potential impact of the use of race in eGFR on patients in the hospital system, it raised concerns that fewer black patients were being referred for transplant care as a result of the adjustment.

Eneanya, Reese, and Grubbs share concerns that use of race may contribute to delayed referrals to kidney care or transplant for some black patients.

In her book Hundreds of Interlaced Fingers (6), Grubbs writes about a very muscular black male patient who over the course of his kidney disease experienced muscle wasting. She chose to use a non–race-adjusted eGFR and referred him for transplant evaluation, something that would have been delayed by 2 years based on his race-adjusted kidney function.

For patients with higher kidney function, the race-based adjustment is less likely to change their care, Eneanya noted. But when using the race-based adjustment would change their care, it is important for physicians to be transparent.

“I’m quite transparent with my patients about many things, not just race and eGFR, especially if they’re lying right on the border [of being referred for transplant or dialysis planning] and their race is making a difference,” she said. “I’ve talked to my patients about that and I usually do some confirmatory testing, either cystatin C or urinary creatinine clearance, to confirm where their kidney function is.”

Pursuing alternatives

The change at the University of Washington School of Medicine was the result of many years of advocacy by black students to address racism in the curriculum and in medicine, including a 2016 sit-in, noted Nkinsi. The protest gave rise to creation of the school’s Anti-Racism Action Committee (7), which is made up of medical students, faculty, and staff who work together to identify and eliminate racism.

“It’s our responsibility as students and faculty in academic medicine to take a second look at all of our curricula and make sure that we’re not using race in inappropriate and unconstructive ways,” said Elizabeth Stein, a medical student who co-chaired the committee.

The committee worked with faculty in family medicine, laboratory medicine, pathology, and nephrology to assess the use of race in eGFR and together determined that “use of race in the biomedical environment is an imprecise variable and does not meet the scientific rigor UW Medicine expects of diagnostic tools,” according to a statement (2). As a result, its laboratories have shifted from use of the MDRD equation to the CKD-EPI equation without race as a variable.

Stein acknowledged that the new equation is “still not perfect” and that more research on better alternatives is needed. “The move by University of Washington Medicine signals that we can move beyond race-based medicine and actually practice something closer to evidence-based medicine,” she said.

In 2017, medical students at Beth Israel Deaconess Medical Center in Boston lobbied for a similar change after students in its Racial Justice Coalition raised questions about the use of race in kidney function estimations, according to Leo Eisenstein, MD, who was a medical student and member of the coalition.

“[There] was a clear conflict in the curriculum in how race was described as a very unreliable proxy for genetics and the way race is used every day in medicine as a proxy for genetics,” said Eisenstein, who is now a resident at New York University Bellevue Hospital. Nephrologist Melanie Hoenig, MD, associate professor of medicine, helped champion the students' efforts and helped them work with faculty and multiple departments at the medical center. Eistenstein and his fellow coalition members also worked with faculty and multiple departments at the medical center. As a result, the center chose to report eGFR as a range between the race-adjusted and unadjusted values. This allows clinicians to take into account factors like an individual patient’s muscle mass or nutritional status in determining where patients likely fall on the range, he said. Physicians also receive a note explaining that patients with greater muscle mass or better nutritional status are likely to be near the higher end of the range.

“It takes away the need for this dubious assessment of each patient’s race,” Eisenstein said. “It restores attention to the relevant physiological differences such as muscle mass that are thought to bear on differences in serum creatinine.”

Eneanya said she hopes many more institutions will follow in the footsteps of the University of Washington and other schools that have made such changes. “I’m glad that this momentum is now turning into action,” she said. The decision to eliminate the use of the race multiplier at Mass General and Brigham and Women’s Hospital piggybacked off that work and was also a team effort involving multiple individuals and departments, said Michelle Morse, MD, MPH, assistant professor at Harvard Medical School. She presented a Grand Rounds on research that critically assessed the studies used to support the notion of higher muscle mass in black patients by Cameron Nutt, MD, who is was one of the medical students working on the effort at Beth Israel and is now a fellow at Brigham. Brigham’s Health Equity committee, which Morse co-chaired, funded Mendu’s research, which led to the decision to remove a notation about race adjustment in their medical record system and instead use the unadjusted number.

Grubbs also commended the move: “I’m really thrilled to see that the medical students were able to get the support they needed to make effective change.”

Grubbs recommended that hospitals eventually switch to using cystatin C, which she and her colleagues use at Zuckerberg San Francisco General, to calculate kidney function. She noted it provides a cleaner result than using creatinine. Eneanya agreed cystatin C is one alternative that should be considered, although she noted it is not widely available yet at many hospitals. Grubbs acknowledged that cystatin C can have a longer turnaround time than creatinine and is more expensive, costing about 25 cents per test compared to 5 cents for creatinine. Mendu has also recommended a shift to using cystatin C at Brigham.

“My hope is that if there is more demand for it, then people will be more familiar with it and be able to run tests faster onsite,” Grubbs said. She noted that several other academic medical centers across the country are considering removing the use of race in eGFR calculations.

Ultimately, many are hopeful these changes will improve patient care. “We expect earlier referral for black patients and better chronic kidney disease care and transplant outcomes,” said Morse.

Public efforts by hospitals to eliminate race-based medicine are an important way to begin to rebuild trust with black communities, who may lack confidence in clinicians because of historical breaches of trust, Nkinsi noted.

“The fact that programs are actively trying to rectify these issues is something that will help build that trust in the communities, and hopefully help repair those bridges that have been burned,” Nkinsi said. But, she noted, that is just the starting point; more community outreach is needed. She also said it is important to acknowledge that too often the voices of black students, faculty, and community members have been ignored or discounted.

Eneanya said physician advocacy is essential to these efforts.

“We have to take a stand on issues that have been overlooked and ignored previously to this point,” she said. “We’re moving beyond the path of just describing health inequities and it’s time to take action, whether it be developing equations that don’t use race or being more transparent with your patients when discussing these issues. It’s time for us to take a stand.”