Inside the bruising battle to purge race from a kidney disease calculator

There it was on James Cannon’s lab report, two tiny words: African American.

The words sat next to a number estimating how badly Cannon’s kidneys were failing. And they were definitely failing. His number read 37, less than half of what is considered normal for adults. Oddly, he thought, a different number labeled “if not African American” read even lower: 31.

When Cannon asked why results differ for African Americans, his doctor said it was because Black people have more muscle mass. The physician said Cannon had chronic kidney disease, but said he shouldn’t worry about it and just exercise more.

But the words nagged at Cannon, a Minnesota state judge and former Army lieutenant colonel who has never been particularly muscular. “Why,” he wanted to know, “is there a difference for African Americans and not any other racial group?”

It was 2017, three years before the widespread use of race in clinical algorithms — including the eGFR equation estimating Cannon’s kidney function — would become a flashpoint across medicine. Cannon, now 73 and retired, let the issue go. What he didn’t know at the time was that his kidneys — without attention or speciality care — were slowly but surely beginning to fail.

Seven years later — in a change remarkable for its sweep and speed — nearly all of the nation’s clinical labs have stopped using race in kidney function equations. Kidney specialists now acknowledge millions of Black patients like Cannon may have been undertreated and even denied transplants because the eGFR test — which assesses how well the kidneys filter a waste product of muscle breakdown — relied on an algorithm that echoes faulty assumptions about Black bodies that date back to the slave era. And the national transplant network has adopted a form of reparations, crediting many Black patients for time they should have spent on kidney waiting lists.

In dozens of interviews, kidney specialists told STAT that reaching this point was a messy and painful process. It featured a contentious and unprecedented meeting of nearly 100 Boston doctors, accusations of bullying, and surprising opposition from an eminent Black medical leader. In many ways, the fight was generational. On one side: young physicians in training with a more nuanced understanding of race than their elders. On the other: established physicians who fiercely resisted changing the race-based algorithms they’d either created or confidently used for decades. Tensions and resentments still run deep on both sides.

The years-long battle offers lessons on what other specialties may face as they begin to grapple with their own use of race adjustments. More broadly, it illustrates how difficult it can be to usher change into the culture and power structures of medicine.

It all began with a single question from a Harvard medical student.

In the spring of 2016, Melanie Hoenig, an associate professor at Harvard Medical School, was explaining the kidney function equation to her renal pathophysiology class when first-year medical student Cameron Nutt asked why lab results were adjusted to make Black kidneys look healthier.

It didn’t make sense: Black Americans, he noted, were at the greatest risk of kidney disease. They are nearly four times more likely to suffer kidney failure than white Americans and are more likely to have kidney disease that progresses rapidly. Yet they are far less likely to be placed on transplant lists.

Hoenig, who practices nephrology at Beth Israel Deaconess Medical Center, had never really liked the equation. She thought few clinicians understood its results were just estimates. But like most of her colleagues, she hadn’t ever questioned how the equation incorporated race. After hearing from Nutt and students Danika Barry and Leo Eisenstein, though, she started asking questions herself. Why was race involved? What exactly was the equation correcting for? And was it, as the students suspected, hurting Black patients? “I said, ‘You’re right. That’s nuts,’” she said. “Once you see it you can’t really unsee it.”

Hoenig and her students went to her hospital bosses. They asked the group to talk with the equations’ creators,  just a few miles away at Tufts Medical Center. That discussion, she said, didn’t go very far.

They pressed on. It was no small amount of work. The group talked to heads of the hospital lab, clinical computing, health care quality, and to primary care doctors who relied heavily on the tests. In March 2017, the Boston hospital became the first in the nation to address the issue — by reporting results as a range with no race notation.

Hoenig’s students were circulating to different hospitals for training and spreading word of the changes at Beth Israel Deaconess. Gradually, some hospitals began following its lead. Cannon, meanwhile, was getting weaker.

In 2020, the world changed. Covid-19 showed health disparities were rampant. In May, in Cannon’s hometown of Minneapolis, George Floyd was murdered by a police officer.

The ensuing protests against racial injustice reverberated across the nation and medicine — and eventually into the very tool that had kept Cannon off the transplant list even as his kidneys failed.

On New Year’s Day 2020, Cannon was at home watching a college football bowl game with his family — his beloved Minnesota Golden Gophers were playing Auburn — when he started feeling dizzy. He fell out of his chair, and couldn’t get back up without his son’s help.

At the emergency room, his heart rate was perilously low. His eGFR had fallen to 14. He was hospitalized and spent a month on dialysis. It was official: Cannon needed a new kidney. Still, it would take almost a year for him to be added to the transplant list. By the time he became eligible that November, his transplant coordinator estimated he would be waiting seven years for a new kidney. Six if he was lucky.

The eGFR equation was seen as a huge advance. That made tampering with it harder, especially since many believed its greater accuracy benefitted Black patients.

Assessing kidney function directly is difficult. For decades, the gold standard was to inject a chemical into the blood and quantify its clearance from urine; other options include collecting and measuring urine over 24 hours. Neither is easy or quick. A method that could rapidly and cheaply estimate kidney function from a simple blood test was desperately needed.

The first commonly used equation, the Cockcroft-Gault formula published in 1976, didn’t account for Black patients at all. It was based on how well the kidneys of 249 hospitalized male Canadian veterans (presumably all white) processed creatinine, a waste product of muscle breakdown. Over time, studies showed Black people had higher average levels of creatinine in their blood. Some feared those higher levels could lead to overdiagnosis of kidney disease in Black patients and keep them from getting drugs known to harm kidneys, including certain antibiotics, chemotherapy agents, or the widely used diabetes drug metformin.

In 1999, a research group led by Tufts nephrologist Andrew Levey developed a new calculation that included race; doing so led to the most accurate estimates in all patients, Levey said. It was this equation that would evolve into the eGFR calculator Cannon’s doctor used.

In 2009, Levey’s team published a revised equation that reduced the racial adjustment — from 21% to 16% in Black patients — after analyzing data from a more diverse subject pool  (though national surveys showed less than half of all clinical labs were using the new equation as of 2019.)

Levey said in an interview with STAT that he feared that not including race could disadvantage Black patients who might lose access to medicines they needed, but added he was not thinking about race as thoughtfully as he would today. “We would do it differently now,” he said.

Even as Levey was rethinking the use of race, lines were being drawn in the larger nephrology community. At the nearby Mass General Brigham hospital system, nephrologist Mallika Mendu wondered how removing the racial correction would affect patients. So, with chief medical officer Tom Sequist and other colleagues, she used the system’s registry of chronic kidney disease patients to assess what would happen.

The group found one-third of Black patients would be shifted to a lower level of kidney function that justified more care. Not a single patient that would have been referred for transplant using the race-free equation had been referred under the old equation. “That set off alarm bells,” she said.

The data convinced the system’s leaders to drop the race modifier in June 2020. It was one of the first to remove the racial adjustment, and it was a high-profile Harvard-affiliated health system to boot.

That’s when the real pushback began.

Mass General Brigham’s decision caused widespread concern among nephrologists both inside and outside its walls. They worried about different hospital systems gauging kidney function in different ways. This led to an emergency Zoom meeting attended by nearly 100 Boston kidney specialists.

The eGFR equation’s proponents, according to some who joined the meeting, described it as the evidence-based way to assess kidney function and said making a change could cause unintended harm to patients. Those challenging it felt they were being labeled as woke, anti-scientific rebels. “We kept being told ‘We want to have a scientific discussion’ — implying that what we were saying wasn’t scientific,” said one person who attended. “It felt like an attack.”

Few senior nephrologists spoke in favor of making a change. One who did was Winfred Williams, associate chief of nephrology at Massachusetts General Hospital and the founding director of the hospital’s Center for Diversity and Inclusion. Although he’s a Black physician, like many of his generation, he had used the race-adjusted equation without question. “I was part of the unmindful masses,” he said in an interview with STAT.

“On that call, at one point I felt like someone is going to have to say something that is not comfortable,” he said. “And my younger colleagues were looking at me.” When it was his turn, he said nephrologists were thinking about race simplistically, dividing the world into Black and white without considering ancestry and population genetics. It was as if they were following the segregation-era “one drop rule” that a single Black ancestor, or drop of Black blood, means you’re Black.

No agreement was reached at the meeting. “Whenever you are changing a paradigm, there is resistance to that change,” Williams said, adding that the lack of diversity in nephrology — just 6% in the field are Black — contributed to the discussion moving so slowly.

A similar dynamic of most senior nephrologists downplaying the need to remove race played out across the country, when the University of Washington decided to stop using race in the kidney equation. Rajnish Mehrotra, a nephrologist and editor-in-chief of one of the field’s premier journals, supported the change and praised students, saying their activism “keeps us on our toes.” He was an exception.

Students found pushing for change risky. Naomi Nkinsi, a Black family medicine resident who led calls to remove race while a medical student at UW, worried about getting kicked out of her program. “I was very quickly labeled as a problem student — by faculty and fellow students,” she said. “It was one of the hardest times of my life.”

One of the loudest voices in favor of maintaining the status quo floored those pushing to change the algorithm. It came from Neil Powe, a Black professor of medicine at the University of California, San Francisco, chief of medicine at Zuckerberg San Francisco General Hospital, and an advocate for equity in kidney care.

Powe wrote editorials and gave interviews warning that individual hospitals were removing race from the equation too quickly, in disorganized and unscientific ways. “I think while the intention is good, the way that it’s being implemented may lead to more harm and in fact, could be, frankly, more racist than the equation itself,” Powe said in a JAMA podcast.

It drove Powe crazy that hospitals were assessing kidney function in different ways — some removing racial labels, some substituting muscle mass for race, and some averaging “Black” and “white” numbers. You could have kidney disease at one hospital and go around the corner to another one and not have kidney disease, he said.

Powe said he thought the race-based equation accounted for higher levels of creatinine seen in Black populations, produced more accurate results for all patients, and allowed more Black people access to important diabetes and chemotherapy medications. He said that a host of deeper issues caused disparities and that there was no rigorous evidence the equation harmed patients. “This just became a target, in some sense not quite fairly,” he told STAT.

Those wanting race removed thought Powe should have been on their side. “I was shocked I had to fight Neil on this,” said Vanessa Grubbs, a nephrologist who previously worked with Powe at San Francisco General and has argued against the use of race to assess kidney disease since 2007. “He hurt the effort because all these other folks could say, ‘This Black person says it’s OK so it must not be that bad,’’’ she said.

The UCSF clinical lab decided to remove race from the equation in October 2019. When Powe and other hospital leaders attempted to reverse the decision, it didn’t  go over well. A petition, charging that reinserting race affirmed that “embracing poor science and racial injustice is the best path forward,” gathered more than 700 signatures from doctors and other researchers. The decision stood.

A larger fight played out in news articles and clashing editorials in nephrology journals. It was a duel over accuracy, with one side saying keeping race in made kidney assessments more accurate, and the other side saying race was handled inaccurately from the start.

“To me, it was just stunning when you think about this specialty that calculates things out to three and four decimal points, but when it comes to race, they say ‘You think they’re Black, they think they’re Black, that’s good enough,’” Grubbs said in a recent interview. (Many opposed to the race-influenced equations note that in 2021, the authors of the original equations corrected — 15 years later — the description of how they determined race and ethnicity in their studies, citing “incorrect recall” as the reason; publishing a correction so long after the fact is extremely unusual.)

Amaka Eneanya, then an assistant professor at the University of Pennsylvania, had been pushing to revamp the equation since

co-writing an op-ed for JAMA in 2019, arguing that using race to assess kidney function was sloppy, arbitrary, and harmful. The nephrologist had given dozens of talks to raise awareness. It seemed illogical to her that an equation would deem all Black kidneys biologically different from white, Hispanic, or Asian ones. But she despaired that anything would change because the medical establishment seemed set on defending the equation, and on using its power to shut down voices and reject research studies that questioned it.

“I really saw the ugly side. Chiefs and chairs are very good friends with journal editors. People who were writing things saying race should be removed were getting their papers rejected without review while senior physicians saying don’t remove race were getting their papers accepted,” Eneanya said.

In August 2020, the National Kidney Foundation and American Society of Nephrology stepped in, announcing a task force would examine the issue. Powe would lead the task force with UCSF professor of medicine and nephrologist Cynthia Delgado — a move that raised eyebrows because both came from the same institution, and Powe was so outspoken against dropping race.

The task force heard from 97 experts and numerous patients, considered 26 different approaches to measuring kidney function, held more than 40 Zoom sessions, and had countless debates. After 10 months of work, in September 2021, the panel recommended immediately switching to a new equation that did not involve race adjustment. It also recommended increasing the use of an emerging blood test that measured the protein Cystatin-C, as a race-free measure of kidney function.

The official decision was unanimous. But behind the scenes, there had been acrimony, politics, and to some, strong-arm tactics to silence critics. The experience was traumatizing for Eneanya, a member of the task force who said she felt demeaned and dismissed as she brought up her concerns about race. “I was definitely bullied, it’s not a secret,” she said.

She almost quit, but was encouraged to stay. “I was told if you leave, this is not going to happen,” she said. “I still believe that.” Some task force members supported her only privately, saying they feared not having papers reviewed or grants funded if they made enemies of senior nephrologists.

“This really was about power in medicine,” said one task force member, who supported Eneanya’s account but asked for anonymity to discuss deliberations for fear of career reprisals.

Those who initially wanted to keep race in the equation reject these claims and say their arguments were unfairly characterized as being harmful to Black people. The trainees who pushed to remove race “were one-sided in how they thought about it,” said Powe.

Lesley Inker, who helped develop the original equation with Levey, said she thought the issue of individual patients not being able to access important drugs did not receive enough attention at the beginning of the debate. “I thought that there were some misunderstandings,” said Inker, who directs the Kidney Function and Evaluation Center at Tufts Medical Center. “Everyone acted on this because it came at an important time in our country, but I was really worried people would be harmed on an individual level.”

Task force co-chair Delgado, who is Puerto Rican and has wondered if she would be classified as Black or white if race was used in the equation, said conflicting voices on the task force began to compromise when they accepted that clinical accuracy and a responsibility to equity both mattered. “Like, ‘Aha, we are all on the same page,’” she said.

Also critical was letting all voices be heard. “I had to change some of my views and listen,” said Powe. He came to see that the racial adjustment was upsetting to some patients. “For that reason alone, I think it’s worthy to take it out,” he said.

Another major factor was the work of Inker, who was able to develop a new equation without using race as a factor; she found that while it was not as accurate as the original, it was still precise and more fairly distributed inaccuracies among different racial groups.

There are still holdouts who haven’t accepted the new way of estimating kidney function. And no one thinks changing one equation will end racial disparities in kidney disease. “Please stick with us,” Deidra Crews, president of the American Society of Nephrology and co-director of the Johns Hopkins O’Brien Center to Advance Kidney Health Equity, urges fellow nephrologists. “There really is more work to do.”

That work is benefiting from some new and unlikely allies. Levey is serving on an ASN task force making sure the new recommendations are implemented. He said it became clear to him his kidney equation had to be changed when he realized “the objections to use of race were not about accuracy; it had to do with whether people wanted to use race or not in medical decision-making.”

Working alongside him is Grubbs. “Now we’ve come completely 180,” she told STAT. “I say something and Andrew Levey says, ‘I agree with Vanessa.’”

It’s said history is written by the victors. In medicine, it’s often written by department chairs.

As medical systems implement task force recommendations, official narratives about who deserved credit for the change largely left out the students, trainees, and young physicians who took most of the risk by initially speaking out, and organizations like the Institute for Healing and Justice in Medicine that kept the issue in front of physicians. Honors flowed from academic societies and groups such as the National Kidney Foundation. But the names on those awards mainly belong to the established physicians who ran the task force.

“Some of the people now that are vocally for these changes … were actively against it and actively pushing against us at the time,” said Nkinsi, the former UW medical student. And now that it’s popular, everyone’s like, ‘Oh my gosh, we are standing with equity.’”

A history Powe published in Cell doesn’t mention by name people who called for change, but accuses many of them of focusing on rhetoric and misconceptions over evidence. “These fancied assertions clashed with the facts and muddled a constructive, evidence-based crusade for change,” he wrote.

That argument maddens the activists because it paints them as emotional and political, ignoring that they eventually were able to publish numerous studies and reviews in top journals demonstrating the racial adjustment caused harm. “To me, this white-knuckled grip on keeping the race correction in despite so much evidence to the contrary is the emotional and political part,” said Jenny Tsai, an emergency physician in Oakland, Calif. who has written extensively on the use of race in medicine and the harm it can cause.

“Now he’s taking credit and coming out like it was his idea to begin with,” Grubbs said of Powe.

“This is revisionist history, 100%,” said Eneanya.

Powe told STAT that he credits the younger generation with speaking out against what they saw as a moral wrong but says they went about it the wrong way, stoking controversy and division. He said he and Delgado, as task force leaders, were successful in bringing people together and creating recommendations that have been taken up rapidly across medicine.

For others, the process was too slow, at times humiliating, and even led some to leave academic medicine. A group of trainees published a history highlighting their contributions, contemplating “how much more readily progress could have been made if medical hierarchies were more willing to listen.” After not being included in a paper that UW faculty members published, Nkinsi wrote her own history of how the change occurred there. “In a sense, I do feel erased,” she said.

Run-ins with Powe — he once told Grubbs to leave the issue of kidney assessment to the experts, she said, when she is a nephrologist and he is not — are one reason she left academic medicine; she now provides care at a facility serving a predominantly Black population in West Oakland, Calif.

Eneanya left academic research, and her position at Penn, after seeing how political — and not evidence-based — academic medical publishing could be. “I fell out of love,” said Eneanya, who now serves as chief transformation officer for Emory Healthcare.

Meanwhile in May, for their work on removing race from kidney assessments, Powe and his task force co-chair Delgado were placed on Time’s list of the 100 most prestigious people in medicine.

Black Americans are nearly four times more likely to have end-stage kidney disease than white Americans, but are less likely to receive the optimal treatment: a kidney transplant. Black Americans are less likely to receive kidneys from living donors and less likely to be placed on waiting lists for deceased-donor kidneys, even after accounting for factors like comorbidities and insurance status.

There are many reasons for the racial disparities in access to transplants, but one of them, according to the nation’s organ transplant network, is that the racial adjustment in the kidney function equation “led to a systemic underestimation of kidney disease severity for many Black patients.”

In 2022, the Organ Procurement and Transplant Network required all hospitals involved in its transplant network to use the race-free calculation. It was a big stick: Hospitals that refused could face sanctions. Last year, the group stunned even health equity advocates when it announced it would adjust waiting times for Black patients whose care had been delayed because of the race-based equation.

For many, the decision was a form of medical reparation. “To be able to see the actual compensatory action and restorative justice, the redress happening in front of our eyes for kidney function and for transplant, is so exciting,” said Michelle Morse, who led a U.S. House investigation of clinical algorithms and now serves as chief medical officer of the New York City Department of Health and Mental Hygiene.

For others, the change came too late. La’Tonzia Adams, who works with the College of American Pathologists to ensure race-free kidney assessments become adopted, lost her father to severe kidney disease that went undiagnosed for years. By the time his kidney failure was diagnosed, Chief Charles C. Adams, a firefighter and Vietnam veteran, was too sick to receive the kidney his daughter wanted to give him. He died in 2001, at the age of 54.

“He could definitely have had earlier interventions,” said Adams, a physician in Portland, Ore. But her dad is also the reason she works so hard on the issue now. “We can fix this,” she said. “It would make him proud.”

Patients who have benefited include Jazmin Evans, who was diagnosed with kidney disease at age 17 and placed on the transplant list in 2019 after years on dialysis. At the time, she was told it might be a decade before she’d get a new kidney.

In April of last year, the Temple University doctoral student received a letter “literally out of the blue” saying she should have been placed on the list in 2015. Evans, now 30, received a kidney a year ago. Without the correction, she told STAT, “I 100% believe I’d still be on a waiting list.”

It wasn’t as easy for Cannon. He was told he’d earn waitlist time only if he could produce lab reports showing he should have qualified earlier. He knew those reports existed, but was told by his health system any readings labeled African American had been scrubbed from patient records because of the new policy. They’d thought that was the right thing to do.

Cannon was flabbergasted. “What good does that do for Black patients? How can they get a lab report that meets the criteria?” asked Cannon. “Don’t erase it like you’re erasing history.”

After numerous calls to several labs and hospitals, one dredged up the reports he needed, helping him cut his wait in half. He now expects to get a new kidney within two years.

The transplant network told STAT that more than 16,000 Black patients have received a waiting-time modification of an average of 2.3 years. As of mid-July, nearly 4,000 of them had received new kidneys.

“When I saw those numbers,” said Eneanya. “I was tearing up.”

STAT’s coverage of  health inequities is supported by a grant from the Commonwealth Fund. Our financial supporters are not involved in any decisions about our journalism.