Melanie Hoenig, MD, a nephrologist and associate professor at Harvard Medical School and Beth Israel Deaconess Medical Center, remembers the first time she seriously questioned the use of a patient’s race in the common clinical algorithm that helps doctors determine kidney function.
It was the first week of her renal pathophysiology class in 2015, and she was teaching her first-year medical students about the estimated glomerular filtration rate (GFR). She explained it’s a common clinical formula that incorporates a patient’s sex, age, and race to approximate how well the kidneys function.
Cameron Nutt, a first-year student at the time, asked why the formula adjusted for race in a way that can make kidney function appear better than it actually is for Black people, who are at greatest risk for kidney disease of any racial group. He and two other students, Danika Barry and Leo Eisenstein, wondered about the use of a formula that made biological assumptions about race, which is widely recognized as a social construct.
These questions gave Hoenig pause. Why did they correct for race? What should they do if the patient is multiracial? And why were they using a formula that assumes better kidney function in a population group that disproportionately experiences kidney failure?
“I think it’s really important to take a step back from the way we always do things and listen to our students.”
Melanie Hoenig, MD, associate professor, Harvard Medical School, and a nephrologist at Beth Israel Deaconess Medical Center
These are questions that medical students, faculty members, and clinicians at academic medical centers across the country have increasingly been asking as systemic racism has gained attention from institutional leaders.
And even as some experts have spoken out against the use of race in estimated GFR and other clinical algorithms for years, medical students have proven to be a powerful force in bringing about change recently.
“I think it’s really important to take a step back from the way we always do things and listen to our students,” Hoenig says. “Having students ask probing questions makes me a better everything — better clinician, human, mother, teacher, and so on.”
Perpetuating inequity
The formula for estimating GFR now used by most laboratories was developed by a group of physicians and researchers in 1999 based on observations from a study that included 1,304 White people and 197 Black people. The researchers found that, on average, the measured GFR of study subjects who were identified as Black was higher than other groups’ GFR, which the researchers took to mean that the formula underestimated the level of kidney function in Black people. So, they added a “race correction” that assumes Black patients have a higher estimated GFR than the formula would suggest. The formula was updated a decade later in a project that looked at 10 studies that included 8,254 people, and it was further validated in 16 studies that included 3,896 people. The researchers were unable to explain the reason for the differences in kidney function between Black and non-Black people. Some researchers said the underestimation of GFR in Black people could be because they believed they have higher muscle mass.
Critics of the use of race in the formula at the University of California, San Francisco, School of Medicine argue that muscle mass can vary widely among individuals within the same race; the assumption that Black people are biologically different from people of other races is not backed by evidence and reinforces erroneous assumptions; to categorize people as either Black or not-Black fails to account for the diversity of the patient population; and the formula can lead to overestimation of kidney function for a Black patient, potentially delaying care.
“The implicit acceptance of GFR race correction reinforces antiquated colonial myths that there is something fundamentally different between races,” wrote four physicians in an opinion for the San Francisco Examiner. “These are the same fallacious narratives that have been invoked throughout history to justify horrendous acts such as slavery and indigenous genocide, on the ‘scientific’ basis that the persecuted race was biologically inferior to whites.”
But some nephrologists, including the lead researcher on the development of the formula, Andrew Levey, MD, caution against removing race from the equation before more extensive research is done because it could lead to unintended consequences.
“All of us should be asking what are the ultimate questions we all care about: ‘Why are there disparities in care? How can we do a better job to make sure that every patient that is in front of us gets the best care?’”
Lesley Inker, MD, MS, an associate professor of medicine at Tufts University School of Medicine
“Some have proposed eliminating the race coefficient, but this would induce a systematic underestimation of measured GFR in blacks, with potential unintended consequences at the individual and population levels,” Levey and other physicians wrote in an article published in the Clinical Journal of the American Society of Nephrology in August. “We propose a more cautious approach that maintains and improves accuracy of GFR estimates and avoids disadvantaging any racial group.”
These unintended consequences could include unnecessarily discontinuing medications such as metformin, an oral diabetes medication, which could lead to prescribing more expensive interventions with potential side effects, such as insulin, says Lesley Inker, MD, MS, an associate professor of medicine at Tufts University School of Medicine in Boston and an author of the article.
She and other researchers have been working for years on developing an alternative way of estimating GFR that does not include demographics and are analyzing the implications of eliminating race from the formula.
“All of us should be asking what are the ultimate questions we all care about: ‘Why are there disparities in care? How can we do a better job to make sure that every patient that is in front of us gets the best care?’” Inker says.
The National Kidney Foundation (NKF) and the American Society of Nephrology (ASN) announced in July that they would form a joint task force to study and make recommendations on the use of race in GFR estimation, with initial recommendations expected later this year.
Black people are three times more likely than non-Hispanic white people to experience kidney failure and are less likely to be identified as kidney-transplant candidates, according to the NKF and ASN. They noted that the current formula is widely accepted and “provides reliable and accurate information on kidney function” but also that race is a social — rather than a biological — construct, and the inclusion of race in the formula ignores diversity within racial groups.
“Because of the complexity, it’s a discussion that’s been around for quite some time,” says Tod Ibrahim, executive vice president of the ASN. “In 2020, with the twin challenges of the COVID-19 pandemic and the recognition or willingness as a country to really confront systemic racism, it became clear we needed to accelerate that discussion.”
Ibrahim says the task force will rely on expert testimony and will seek to learn from academic institutions that have already made the move to eliminate race from GFR estimation.
Challenging the status quo
Beth Israel Deaconess Medical Center in Boston officially removed the use of race in estimation of GFR in 2017. Around this time, students at several other universities started the many-months process of lobbying for the change.
At the University of Washington (UW), questions from medical students inspired the creation of a working group that included the students and nephrologists as well as perspectives from social scientists and other interested groups. After a process that stretched over two years, the university dropped race from its estimated GFR equation on June 1, 2020.
“One of the joys of working at an academic medical center is seeing generations of people that interface with the academic medical center and bring in fresh ideas — bring in a fresh set of eyes to look at old problems,” says Rajnish Mehrotra, MD, MS, interim head of the Division of Nephrology at UW and the editor-in-chief of the Clinical Journal of the American Society of Nephrology. “It keeps us on our toes.”
While many medical schools have historically been hierarchical, a culture shift seems to be occurring that has inspired medical students to challenge the status quo, according to Oluwaferanmi Okanlami, MD, MS, an assistant professor of family medicine, physical medicine and rehabilitation, and urology at the University of Michigan Medical School.
“The culture and the climate of when I was a student didn’t really give room for questioning,” says Okanlami, who has been an advocate for diversity, equity, and inclusion. “We were … meant to assume what we were being taught is true… [now,] people feel more empowered to assert their own truth and question others.”
“One of the joys of working at an academic medical center is seeing generations of people that interface with the academic medical center and bring in fresh ideas — bring in a fresh set of eyes to look at old problems. It keeps us on our toes.”
Rajnish Mehrotra, MD, MS, interim head of the Division of Nephrology at UW and the editor-in-chief of the Clinical Journal of the American Society of Nephrology
Karampreet “Peety” Kaur, now a fourth-year medical student at Vanderbilt University School of Medicine in Nashville, was one of the students who took on the initiative to create change. Kaur’s effort began with doing her research. She and other students reached out to the students at Harvard who had already successfully removed race from the formula at Beth Israel Deaconess Medical Center and read up on the related literature.
“We wanted to make sure we were thinking about race in the right way,” Kaur says. "The scientific literature shows that race is a social construct rather than a biologic variable that reflects genetic differences ... and that fueled our questioning of the use of race.”
Throughout the nearly two-year process, Kaur leaned on her own position as a student to pose questions and respectfully challenge the conventional thinking within the various departments.
The result: Vanderbilt removed race from the estimation this summer.
“As medical students, our job is to be curious and to learn,” Kaur says. “By using that role to our advantage, we were able to have meaningful conversations with people who certainly know more about the kidney than we do.”
Rethinking race in medicine
While the use of race in the estimated GFR has gained national attention, it is not the only clinical algorithm that incorporates race.
Darshali Vyas, MD, a second-year resident at Massachusetts General Hospital in Boston, worked with two other physicians to assess more than a dozen examples of algorithms that incorporate the patient’s race and potentially direct health care resources away from people of color.
Vyas and the others identified potential inequities in algorithms that estimate risk for heart failure, of complications for vaginal birth after a cesarean section, and of complications and death in cardiac surgeries, among others. In each of these cases, Vyas found that the algorithms had the potential to steer people of color away from care, whether it be because their risk for heart failure was underestimated or the risk of complications from surgery were overestimated.
The article, published in the New England Journal of Medicine, garnered national attention and prompted the chairman of the U.S. House of Representatives Ways and Means Committee to call on medical professional associations to issue new guidance that corrects misuse of race in clinical algorithms. Already, the NIH Maternal Fetal Medicine Units Network, which provides the calculator to estimate the risks of vaginal birth after a cesarean section, has begun to develop a new calculator that doesn’t include race.
“It’s important that the next steps involve policy change,” Vyas says. “I think that will — in many cases — require re-approaching the evidence and being open to creating new tools and amending tools that might have been in place for a long time.”