A transparent lens on estimated GFR

Dr. Inker, a member of the NKF-ASN task force, says her laboratory colleagues “agree with the importance of having a national conversation based on the totality of data, evidence, and considerations, and not having a unilateral move that’s quick.”

“The national conversation has widened, and it’s best that well-considered recommendations make the basis for what we do at Tufts, instead of quick changes made without careful consideration of all the issues.” She and her colleagues will be guided by the task force’s final report, she says.

“I think everybody understands that this is an important change to make,” she says. “But they want to make sure it’s patient-centric and doesn’t harm anybody.”

That doesn’t mean standing still. “This would be a great time to change to CKD-EPI, which is more accurate,” says Dr. Inker, who helped author the equation. She sees no reason to wait. It dismays her that some labs still use MDRD; any discussion about re-examining the race multiplier should be done in the context of finding the best way to estimate kidney function overall.

Likewise, she says, now would be a good time for institutions to boost use of cystatin C, and health care providers should increase their ordering. While assay variation was previously a concern, matters have greatly improved since 2018, Dr. Inker says, crediting the CAP Surveys.

As others have noted, estimated GFR is meant to be a preliminary assessment. It has, however, “taken on a status with respect to classifying patients and qualifying patients to be on a transplant list,” Dr. Miller says. He’s concerned that eGFR includes too much uncertainty to be used in such weighty decisions. It can offer insight into disease process and the impact of kidney disease on the population level, he says. “But when you start applying it at the individual patient level, it becomes difficult to justify some of the cutpoints that are being used.”

It bears repeating: It’s an estimate. “That sometimes gets forgotten” in clinical practice, Dr. Miller says. “If you calculate a confidence interval for the value, you’ll find it’s pretty large.” If the value doesn’t seem consistent with other indicators of kidney function, he suggests, it would be worth a follow-up test, like a cystatin C. It is, however, a low-volume test at his laboratory. “I can’t tell you why.”

The task force is viewing its work through a wide-angle lens and has “gone through a very methodical process of examining all the issues around race disparity in medicine, particularly in kidney disease,” Dr. Miller says.

As the interim report makes clear, this remains a work in process, a fact some find curious if not frustrating.

Says Dr. Hoenig: “I am disappointed that the interim report is just that—I would have hoped for more definitive response by now.”

Dr. Baird adds, “It doesn’t actually say anything. It is more of a plan to eventually say something.”

Using his own wide-angle lens, Lindo expresses concerns about relying on task forces and national organizations to create more equity in health care. “I think change comes from those closest to the population affected—the labs, the providers, the educators, and students.” He’s watched other laboratories drop the race multiplier and says it’s this type of groundswell that’s pressuring larger groups to act.

“I don’t know if I want groups like the NKF to be the ones to lead the work,” Lindo says. “I think we need a fresh critique with innovative ideas, and then it’s received and embraced by the much larger national organizations. I can imagine a task force being created, but if there isn’t, for example, a critical race scholar in these national organizations, are the conversations going to be the same as the ones that got us here?”

Dr. Miller says the group has heard from experts in areas ranging from race and ancestry and equation development to kidney disease and drug dosing, “to try to get a very thorough understanding of the complete picture of the impact of using estimated GFR.”

The members are evaluating more than 20 eGFR equations for potential use, including looking at performance characteristics, feasibility, the representation of racial groups used in developing the equations and validating data sets, and the impact on medical and drug dosing decisions as well as epidemiology.

Removing race doesn’t leave a problem-free equation in its wake. “I think this is one of those situations where there probably is not really a correct answer,” Dr. Miller says. “That’s why we’re looking at so many different aspects—to make sure the recommendation fits various uses of the equation and becomes the most practical solution to a difficult problem.”

Once the race multiplier is dropped, the question becomes, then what?

“This is a big question,” says Dr. Hoenig. “Are you going to have everybody be one number—the Black, or the non-Black, or the lower number? Or are you going to present the two numbers? Or average the two?”

Her institution opted to present both numbers as a range, accompanied by text that briefly explains kidney function. She concedes she’s in a fortunate spot—since they use a home-grown system, “we have a lot of real estate, if you will, in the lab report.”

Among those that use Epic, some simply drop the second number; some Epic reports provide a blood creatinine, an estimated GFR, and then add a line underneath telling them to “multiply by” if the patient is Black. “They don’t do the math for you. So a lot of physicians may just ignore that anyway,” Dr. Hoenig says.

At other institutions, however, Epic programs are linked to patients’ registration—patients may identify as Black, white, something else, or prefer not to answer. Epic may pull that information and present the eGFR based on self-reported race, unbeknownst to clinicians, Dr. Hoenig says. “Some think, We don’t even use race, but it turns out they do. They just didn’t know it was happening.”

Dr. Hoenig allows room for ambivalence. “We chose both numbers; I’m not sure that was the right decision. It just seemed like the better move.” Being able to provide a nuanced report helped. They included language around age-related decline in kidney function, since dropping the Black race coefficient without explanation runs the risk of suggesting kidney disease is present, she says. “You could suddenly label a lot of people with chronic kidney disease.”

Though her institution moved early, she says it’s reasonable if labs want to wait for the final task force report, since it might bring more consistency to the whole process. On the other hand, “I know a lot of laboratorians who are saying, Why are we left holding the bag on race? So many laboratorians want the race thing gone now.”

In short, “Either of those stances would be fair,” Dr. Hoenig says.

Dr. Baird sounds like he’s done thinking about it. “Truth be told, since last March there’s been something else occupying the medical system,” he says with a laugh. “I’m not going to spend much time perseverating on when other guidelines come out, or when the rest of the systems in the country catch up. I know what we’re doing here. It took me a while, but I’m comfortable with it.”

Though the ultimate decision doesn’t rest with medical students, they’ve been a key part of the conversations. Dr. Miller has talked with members of VCU’s nephrology department, and they, along with the department chair, have met with medical students, residents, and physicians to address “the use of a race term in an estimating equation as potentially introducing disparity in medical treatments.”

The concern is legitimate, says Dr. Miller, whose own institution will await the final task force report before making a change. “I think it’s very encouraging that medical students are socially responsible in the way they want to practice medicine, and in the way they want to be taught how to practice medicine.” He calls their engagement refreshing and adds, “This is a societal issue. Kidney disease is not the only part of medicine where racial disparity has been known to occur.”

Certainly not. COVID-19 may be the biggest example, but it’s surrounded by others. Race-based benchmarks have even filtered into NFL concussion testing (Possin KL, et al. JAMA Neurol. 2021;78[4]:​377–378).

As the literature continues to swell with accounts of racial inequities in health care, what role might labs have in solving the problem?

Dr. Baird is unsure. “Many of our problems that have to do with racial and social justice in medicine are probably not related to mathematical biases in lab tests,” he says. Rather, he suggests, they’re access problems. “It’s not that A1c is intrinsically problematic. But if you do not have access to health care for a large swath of your population, that’s a bigger problem.”

Has the recent SARS-CoV-2 spotlight on labs given them a bullhorn to agitate for broader change? Dr. Baird responds with, “I’d say, ‘Yes—but.’”

“We could lead this,” he says, “but the question is, How? It turns out that every 102 years there’s probably an opportunity for laboratory testing to be front and center.”

It may take another actor—a relatively new presence on the medical stage—to help maintain momentum. Lindo’s position as assistant dean for social and health justice is one of only two such posts in the country, Lindo says. “You wouldn’t believe the number of emails I get from people seeking guidance, who say, I don’t know how to talk about this with my colleagues.” Last year he gave 163 talks by Zoom.

The pandemic hasn’t made his job any easier, he says, but it has highlighted the racial inequities in medicine. When he talks about it now, “I no longer sound like the person who’s presenting in the wrong room,” he says with a laugh. He no longer has to explain why he has a position in the Department of Family Medicine. “The pandemic means I no longer have to be shouting from the top of the mountain, saying, ‘We have issues.’ Now we’re talking about how to solve them.”

Still, challenges persist. He talks about encounters with those who agree that structural racism exists in medicine, but say that individual racism does not. “That’s fascinating,” he says. “So who’s doing the racism? Desks?”

Others suggest he’s looking for racism in places where it doesn’t exist. “They say, Why do you have to be such a nihilist? Do you just walk around and think the world is racist toward you? And I say, ‘I’m not thinking it—it shows me.’”

He uses the personal to pivot back to the larger problem. He and his wife (a surgeon who is Black; Lindo identifies as Latino) are looking to refinance their home. Aware that their racial identities put them at risk for receiving a lower appraisal (http://bit.ly/nyt-appraisal), they’ve begun contemplating how to avoid the risk, even considering having their white colleagues pose as the owner of the home during the appraisal.

When he shares this story with colleagues, some tell him he and his family should expose the obvious injustice. Lindo is skeptical. “I tell them, ‘Let’s do a test: How about the next time you refinance, you put pictures of a Black family in your home?’ It will prove the same point.”

“That’s the conversation in medicine,” Lindo continues. “Where is the burden being placed? On the backs of historically marginalized communities.” Even as the best-intentioned providers work to end racial injustice, “There’s still collateral damage along the way that we need to contemplate.”

The perpetually fresh voices of medical students could remain a force as well.

Dr. Baird recalls when he learned about the race multiplier in residency, “I just thought, Ah, that’s what we do, and moved on. I just absorbed it.” He says it took a wakeup call from students, as well as the events of last year—and perhaps 2021—to start asking his own questions. “I sort of glossed over that in the past,” he says, “and I think as a system we didn’t give it the due diligence, the questioning, that it could have used.”

Like Dr. Baird, Dr. Hoenig says it didn’t necessarily occur to her to question the race multiplier. “It was just sort of handed to me: This is the way we do it. I had never thought about this before.

“But once you see it, you can’t unsee it.”

Karen Titus is CAP TODAY contributing editor and co-managing editor.