ABSTRACT
Dr. Chi-yuan Hsu describes controversies surrounding the race coefficient (“if African American”) in widely used kidney function estimating equations. He outlines recent research results on this topic by himself and others and comments on the relationship between activists and the academic medical community.
INTRODUCTION
I am very honored to be inducted into the American Clinical and Climatological Association (ACCA) and grateful to have the opportunity to present this paper.
I had initially debated what to title this piece to capture what I hope to describe. Potential alternatives included “Eliminating the Race Coefficient in Kidney Function Estimating Equations: What I Witnessed During the Storm” or “Eliminating the Race Coefficient in Kidney Function Estimating Equations: A Tale of Clashing Values.” I settled on calling it “Eliminating the Race Coefficient in Kidney Function Estimating Equations: The Center Did Hold” for reasons which I hope will become clear.
SERUM CREATININE AS A MEASURE OF KIDNEY FUNCTION
As almost everybody in this audience knows, serum creatinine (SCr) is one of the most ubiquitous blood tests in clinical medicine. By some estimates, it is ordered over 250 million times annually in the United States alone (1). For the last two decades, serum creatinine is typically used to estimate glomerular filtration rate (GFR) via equations such as the Modification of Diet in Renal Disease Study (MDRD) equation (2) or the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation (3). These equations take into account not only SCr but also age, sex, and race (as presumed proxies for non-GFR determinants of Cr) to derive an estimated GFR (eGFR). The race coefficients in these equations are in the form of a multiplier for “if African American” (Table 1 ) (2,3).
TABLE 1.
Glomerular Filtration Rate (GFR) Estimating Equations
| Sex | Serum Creatinine (mg/dL) | Serum Cystatin C (mg/L) | Equation | |
|---|---|---|---|---|
| MDRD equation | GFR = 175 × (Scr)-1.154 × Age -0.203 × 0.742 [if female] × 1.212 [if Black] | |||
| CKD-EPI creatinine equation | ||||
| Original | Female |
|
|
|
| Male |
|
|
||
| Refitted (without race) | Female |
|
|
|
| Male |
|
|
||
| CKD-EPI cystatin C equation 2012 | ||||
| Unchanged |
|
|
||
| CKD-EPI creatinine-cystatin C equation | ||||
| Original | Female |
|
|
|
| Male | ≤0.9 >0.9 |
|
|
|
| Refitted (without race) | Female |
|
|
|
| Male |
|
|
|
|
Abbreviations: MDRD, Modification in Diet in Renal Disease; CKD-EPI, Chronic Kidney Disease Epidemiology Collaboration.
Reproduced from reference (32).
The race coefficient was included in these equations because MDRD (2) and CKD-EPI (3) study investigators noted that at any given age, sex, and SCr, those identified as African Americans had higher measured GFR than those not identified as African American. (Most of the non-African American study participants in the MDRD and CKD-EPI databases were White.)
Their observations are consistent with earlier literature. In 1998, Dr. Camille Jones and colleagues analyzed nationally representative Third National Health and Nutrition Examination Survey data to show that among young and healthy individuals without diabetes or hypertension—who therefore were unlikely to have underlying kidney disease—Black Americans had higher mean SCr levels than White Americans (4). These authors wrote: “We have shown that there are ethnic differences in the population distribution of serum creatinine levels. If the ethnic differences in serum creatinine level are caused by differences in muscle mass or in renal tubular handling of creatinine rather than a true difference in renal glomerular function, then defining an abnormal cutoff value for creatinine level based on the total US population percentiles could cause … possibly iatrogenic morbidity because of overdiagnosis of renal disease in the non-Hispanic black population.”
QUESTIONING THE RACE COEFFICIENT
When the MDRD and CKD-EPI equations were first promulgated [in 1999 (2) and 2009 (3), respectively], there was not much concern raised in the mainstream nephrology or general medical community regarding the race coefficient. This may be because adjusting for age, sex, and race has been standard practice in epidemiology and clinical research.
However, in retrospect, questions were being raised (5). For example, in 2015, Professor Dorothy Roberts, a sociology and law school professor now at the University of Pennsylvania, called out the race coefficient in eGFR equations as “race-based medicine” in her TED talk entitled “The Problem with Race-Based Medicine” (6).
The first paper to attract significant attention in mainstream medicine literature that questioned the use of race in estimating GFR was published by Dr. Nwamaka Eneanya and colleagues in the Journal of the American Medical Association in 2019 (7). They argued that “use of kidney function estimating equations that include race as a variable cause problems for transparency and unduly restrict access to care in some cases, yet offer only modest benefits to precision.” The issue of disadvantaging Black American patients from preemptive (pre-dialysis) transplant waiting was highlighted since the threshold to waitlisting depends on an absolute kidney function cutoff and is frequently operationalized as having an eGFR <20 ml/min/1.73m2.
More criticisms regarding the use of the race coefficient in eGFR equation medicine then appeared in the lay press (8,9). This trend accelerated with the murder of George Floyd in May 2020 and the racial reckoning that subsequently swept the country. The disproportionate impact of COVID-19 on communities of color added to the sense of urgency. High-profile pieces appeared in publications such as the New England Journal of Medicine (10,11), the New York Times (12), and Annals of Internal Medicine (13) questioning the consideration of race in medical decision making. The race coefficient in GFR estimating equations was highlighted as being emblematic of the problem. In September 2020, Congressman Richard E. Neal (D-MA), who was chair of the House Ways and Means Committee, pressed professional societies to seek alternatives to using race in clinical algorithms (14).
WHAT I WITNESSED DURING THE STORM
Abolishing the race coefficient in GFR estimation equations was a cause taken up by activists ranging from medical students (15-17) to postgraduate trainees to medical school (nephrology) faculty (18). Numerous petitions were sent to leaders in academic medicine including chancellors, chief medical officers, department chairs and nephrology division chiefs (myself included in the last group) (19).
Activists pushed for the race coefficient to be dropped from the MDRD (2) or CKD-EPI (3) equations and for the estimate for non-African Americans to be applied to African Americans. Some academic nephrologists advocated for this in somewhat inflammatory language such as: “The race multiplier for eGFR serves no purpose in medicine aside from falsely reassuring black patients that things are better than they are and delays transparent waitlisting. It should be dropped everywhere …” (20).
This step was ultimately taken at numerous medical centers around the country, at times without concurrence with nephrology division leadership. At the University of California San Francisco (UCSF) Health, we chose not to adopt this because we thought that purposely misclassifying estimated GFR only among African Americans was not an appropriate response. At UCSF Health, in October 2020, we did change the labelling in the electronic medical record for “eGFR if African American” to “eGFR – high estimate” and “eGFR if non-African American” to “eGFR – low estimate” while we awaited guidance from professional organizations. The two largest and most prominent nephrology professional societies in the United States—the National Kidney Foundation (NKF) and the American Society of Nephrology (ASN)—had established a joint task in July 2020 to reassess the inclusion of race in diagnosing kidney disease.
But waiting was not without controversy as some faculty members and trainees questioned why leadership appeared to be dawdling and not moving immediately to simply drop the race coefficient from existing equations and apply the same eGFR value to all.
A TALE OF CLASHING VALUES
In parallel, numerous research papers appeared in nephrology and general medicine journals probing different aspects of (removing) the race coefficient. As elaborated upon in a recent review article (21), one approach to organize this literature is through the lens of clashing values, which included: (i) racial justice, (ii) benefit of classifying more Black Americans as having chronic kidney disease (CKD) (or more severe CKD), (iii) accuracy compared with measured GFR, and (iv) financial cost.
(i) Racial justice: Critics have pointed out that the equations estimating GFR with race coefficients “reinforce the flawed idea that each racial and ethnic population is fundamentally homogeneous and that defined biological differences exist between races and ethnicities” (22). They see this practice as “rooted in a history of work from the fields of anthropometry and eugenics that set out to scientifically prove blacks are biologically distinct and separate from whites” (18).
The dichotomous “if African American” classification was also criticized for ignoring racial admixture (18). Finally, there is concern that race may be assigned in many medical records or research databases by hospital staff or research teams in an arbitrary manner, thus leading to bias and misclassification (22).
(ii) Benefit of classifying more Black Americans as having chronic kidney disease (CKD) (or more severe CKD): Black Americans are at ∼three-fold higher risk of kidney failure than White Americans. Advocates for applying the “non-Black” estimate even to those who self-identify as Black stress the benefits associated with a systemic lowering of estimated GFR values among Black Americans, which would include facilitating access to subspecialty (e.g., nephrology) care and earlier preemptive transplant waitlisting (23).
Concerns about harms from systematically underestimating kidney function among Black Americans by dropping the race coefficient in the MDRD or CKD-EPI equations (such as underdosing of medications) (19) were acknowledged but felt to be worth the tradeoff (22,24).
(iii) Accuracy compared with measured GFR: Many who defended the status quo pointed out that the MDRD and CKD-EPI equations were created based on good-faith efforts to maximize accuracy in the study participants (25). Those who prized accuracy above all (26) felt that any “argument from social justice and history” was contrary to “a scientific approach” (27).
(iv) Financial cost: The same CKD-EPI research group that created the SCr-based CKD-EPI GFR estimating equation had previously reported that race is not needed when an alternative filtration marker, cystatin C, is used to estimate kidney function (28) (Table 1). While some have advocated for the wholesale switch over from SCr to cystatin C (18) as a way to eliminate race from estimating equations, cystatin C has not been more widely adopted, with a major barrier being higher cost (on the order of ∼5- to 10-fold more expensive per test).
THE CENTER DID HOLD
On September 23, 2021, the NKF-ASN Task Force led by Drs. Cynthia Delgado and Neil Powe issued its much-awaited final report (29,30). Appearing the same day in the New England Journal of Medicine were two original research publications. The first paper by my colleagues and me analyzed data from the multicenter Chronic Renal Insufficiency Cohort (CRIC) study to answers several questions of great interest to the field (31). The second paper by Dr. Lesley Inker and colleagues reported a new CKD-EPI equation that did not contain a race coefficient (32).
My colleagues and I showed that in CRIC, at the same measured GFR level (and age and sex), Black study participants had higher SCr than non-Black study participants (31). Although CRIC is not a nationally representative sample, this is consistent with nationally representative data alluded to above (4). Thus, having a race coefficient would maximize accuracy when using SCr as the filtration marker to estimate GFR. Over one-third of CRIC participants self-identified as being (non-Hispanic) Black. Since entry criteria into CRIC and SCr and GFR measurement methodologies did not differ by race, Black and non-Black CRIC enrollees are more similar than Black and non-Black study participants analyzed in prior databases, such as those used in the CKD-EPI meta-analysis (21, 22, 33).
Addressing the issue that a dichotomous race coefficient in GFR estimating equations ignored racial admixture and is subject to biased and erroneous assignment, we showed that replacing race with genetically derived ancestry (expressed as percentage of African ancestry) gave similar results. Specifically, in the full CRIC study sample, at the same SCr level (and age and sex), Black race was associated with a 12.8% higher measured GFR, and every 10% increase in percentage of African ancestry was associated with 1.6% higher measured GFR (31). [These results, however, should not be interpreted to mean that races are biologically distinct (21).]
Numerous authors have suggested the race coefficient could be replaced by better capturing any racial differences in non-GFR determinants of SCr such as height, weight (6), dietary factors, or creatinine metabolism (22). My colleagues and I also examined this and found that, in CRIC, the independent association between race and SCr was attenuated (from 12.8% to 8.7%) but not eliminated even after additionally considering factors such as dietary protein intake, 24-hour urine creatinine excretion, and measures of body composition not available in routine clinical care (such as bioelectrical impedance analysis phase angle) (31).
Finally, we showed that an ancestry term does not meaningfully improve the performance of a cystatin C-based equation (31). CRIC results confirmed and extended prior studies which showed that when cystatin C is used as the filtration marker, race is not required to produce GFR estimates of comparable accuracy to SCr-based equations.
To derive the new SCr (and SCr-cystatin C) CKD-EPI equations (Table 1), Inker et al used the same development datasets for their 2009 SCr-only CKD-EPI equation and their 2012 SCr-cystatin C combined equation, but they added more studies to the external validation dataset (32). The CKD-EPI investigators did not change their equation format, including where knots were placed (e.g., SCr of 0.7 mg/dl for women and 0.9 mg/dl for men) and used the same modeling approach as they had before. But the key analytic difference was not including race as an explanatory variable.
Furthermore, the decision was made to keep the proportion of Black participants in the development dataset higher than 13%, the actual U.S. population percentage of African Americans (32). This is important since this proportion (or a different weighing scheme) would affect how the systematic error is distributed between those who are African American and those who are not.
Figure 1 is reproduced from Inker et al to show the primary results (32). The original CKD-EPI SCr-based equation with age, sex, and race (ASR) was unbiased for both Black and non-Black participants (left most panel with both orange and green lines overlaying the line of identify). Dropping the race coefficient and applying the original CKD-EPI SCr-based equation “non-Black” estimate regardless of race means that bias was concentrated only among Black participants. This is illustrated in the middle panel (ASN-NB) in which the line representing Black participants (green line) is now off the line of identify but not the line representing non-Black participants (orange line). The right most panel shows the results from the refitted equation without race. Now there is measurement bias among both Black and non-Black participants, and both the green and orange lines do not overlay the line of identify. Inker et al reported that the new race-free SCr CKD-EPI underestimated measured GFR in Black participants by 3.6 ml/min/1.73 m2 and overestimated measured GFR in non-Black participants by 3.9 ml/min/1.73 m2 (32). The authors provided data showing how the magnitude of bias in each race group varied by the proportion of Black participants in the development data sets (the higher the proportion, the less bias there is among Black participants and the more bias there is among non-Black participants).
Fig. 1.
Comparing measured vs. estimated level by different estimating equations. Reproduced from reference (32). See text for details.
Inker and colleagues also quantified how much estimates of CKD prevalence among Black Americans would increase with the new SCr-based race-free CKD-EPI equation as compared with the existing CKD-EPI equation with race coefficient.
The two main recommendations of the NKF-ASN Task Force directly relevant to clinical care were to immediately adopt the race-free CKD-EPI SCr eGFR equation and to endorse more widespread use of cystatin C (29,30).
Because of the careful and thorough work of this task force, their recommendations were adopted immediately and without controversy. Most in the academic and clinical community viewed the task force as having an inclusive membership and were fair and rigorous in their approach. The controversy surrounding the race coefficient in kidney function estimation died down as both commercial laboratories [such as Quest and Labcorp (34)] and academic medical center laboratories quickly switched over to the race-free CKD-EPI equation (35).
Viewed through the lens of clashing values, the NKF-ASN Task Force recommendations were a successful compromise balancing racial justice (by eliminating the race coefficient and spreading the error among both Black and non-Black persons) and accuracy (it is “accurate enough”).
That the NKF-ASN Task Force moved in this direction appeared to have surprised some. For example, Dr. Vanessa Grubbs posted in her blog: “Given that the task force was led by some of the most outspoken critics of removing the race correction, I braced myself. … But, to my surprise, last week—after eight months of debates, petitions, news articles, medical journal publications, oral testimony, written testimony and tweets from thousands of medical students and physicians around the country—the task force released a statement of their decision: medical institutions should stop using GFR equations that called for a ‘race correction’” (36).
CONCLUSION
In conclusion, my personal view is that the activists had a strong moral case for why the race coefficient should be eliminated from kidney function estimating equations. However, there was lack of acknowledgment for data showing that considering race does improve GFR estimation when SCr is used as the filtration maker (for reasons we still don't fully understand). And there was downplaying of harm resulting from simply dropping the race coefficient from existing equations and applying the “non-African American” value to all persons [which has been termed the “dominant race standard” (37)]. There were also incorrect assumptions made about the motivations of those who did not immediately adopt the same stance as advocated. Nevertheless, the activists’ core argument was ultimately persuasive and carried the day.
I believe that it is a laudable achievement that the NKF-ASN Task Force was able to find a solution that quelled the controversy. I also believe the events that unfolded over the last several years can be viewed at the end of the day as a success story for the medical establishment and an example of the ability of professional societies to lead and make positive changes. This may serve as a useful model as academic medicine grapples with other difficult societal and moral issues. After all, clinical medicine is not practiced in a vacuum but rather is—and has to be—influenced by forces that shape society at large.
I would like to end by thanking Drs. Anupam Agarwal, Neil R. Powe, and Talmadge E. King, Jr. for sponsoring my ACCA membership.
ACKNOWLEDGMENTS
The author would like to acknowledge with gratitude Drs. Peter Yang, Harold Feldman, and Alan Go (the other lead authors in the CRIC paper) and all the CRIC study participants, research staff, and investigators.
DISCUSSION
Rosenberg, Minneapolis: Thank you, Chi, you've told this story quite well. Nephrology is at the cutting edge of race-based medicine. It is amazing when you think about the transition that has occurred with elimination of the race coefficient for eGFR and the fact that people never really paid a lot of attention to it until recently. Nephrology has responded really well to this issue with elimination of the race coefficient being implemented at most U.S. hospitals. Race-based medicine equations still exist in other specialties such as pulmonary, obstetrics, and ophthalmology. What are some lessons from nephrology that can be applied to these other specialties?
Hsu, San Francisco: Thanks, Mark. I think each algorithm has to be looked at carefully and individually. As you know, race is a complicated and sensitive topic in the United States. Studies have shown that if you use these race-free GFR equations, you actually have less ability to detect racial disparities. There's a paper in JAMA (https://jamanetwork.com/journals/jama/fullarticle/2793290) by the CKD-PC group showing that, surprisingly, if the new race-free equation is used, there is no difference in risk of adverse outcomes among Black and White study participants at the same level of eGFR <60 ml/min/1.73m2. But you restore the expected racial difference (Black study participants had higher risks of kidney failure at any given eGFR level) if you use the old creatinine-based equation with race or the cystatin C equation. So, you might get misleading results using the new race-free equation because you are building in misclassification by systematically overestimating kidney function among those who are non-Black and systematically underestimating kidney function among those who are Black.
Again, it depends on the specific case and it might be difficult to generalize. There may be a price to pay for the cost of having race-free algorithms, and the cost-benefit ratio needs to be assessed on a case-by-case basis.
I absolutely agree with the point that one shouldn't jump from seeing an association with race to concluding that there must be a genetic explanation. Our results should not be interpreted to indicate this is “biologic” in some sense. There is a significant correlation between race and genetic ancestry, so it is not surprising that ancestry can replace race in eGFR equations, but we don't really understand why there are these kinds of difference when it comes to creatinine between those who self-identify as Black and those who don't. I think to this day, we still don't understand that.
I personally think more use of cystatin C is a good way to go. (I don't have any stocks in cystatin C measurement-making companies I should disclose!)
The way to deal with the many ways race is currently embedded in different medical algorithms is a difficult thing and should be addressed on a case-by-case basis.
Rathmell, Nashville: I really liked your talk, especially taking it into a scientific analysis. I was the interim chair for two weeks when I got that letter, like you got, but I actually found this particular issue to be somewhat easy to come down on because it felt like the issue (race, ancestry, and science) was really about math and what felt like lazy medicine, right? It seems that we were just populating something on a sheet for the doctors who didn't want to think about their patients and could just bucket them based on math that somebody else did for them. I wondered how much of this is really a different kind of problem—doctors who are really looking for more convenience rather than the opportunity to really examine their patients. The argument I made to our head of nephrology was if the nephrologists still wanted to use that calculation, then they could get a calculator out and do it; if they felt like that was better for their patient, then they could do that math, but it really bothered me very much that we were just saying the math problem was going to get in our way of understanding biology.
Hsu, San Francisco: Yes, thank you. In the UCSF chemistry lab, this was the only lab test that involved using race as a coefficient, which was a really distinct thing.
This idea about looking at each patient carefully is certainly valid. But the fact of the matter is, a doctor goes through so many clinical tests every day that it is very difficult to think through each one very deeply. eGFR is used by family medicine nurse practitioners, radiology technicians, etc. It's not realistic to expect many people to have to think about each Cr measurement carefully.
Now with kidney transplant, one actually can because there are fewer professionals involved. If you look at the rules, it doesn't say eGFR alone meets the criteria for listing. The rules are “below 20” but could be creatinine clearance, could be measured by 24-hour urine, or could be estimated by the Cockcroft-Gault equation. Since those dealing with transplant can be much more sophisticated than the average practitioner in knowing how creatinine relates to kidney function, another way to have dealt with any disadvantages Black patients have when it comes to waitlisting could have been focused on that particular step which has received so much attention, including in the lay press.
But I think your point is correct. Separate from the math, it's best that we get rid of race as much as we can in these sorts of algorithms. Fundamentally, I agree with the activists who brought our attention to this. As members of the establishment (which we all are here), our job is to convert advocacy into action in the way that makes the most scientific sense. The basic premise that we should try our best to get rid of race coefficients is right. It's quite remarkable that we didn't think consider this until society thought about racial reckoning more generally in recent years. We just accepted the equation with the race coefficient when it came out, back when I was a fellow and junior attending. To me, that is an important and interesting lesson and commentary.
Footnotes
Potential Conflicts of Interest: None Disclosed.
REFERENCES
- 1.Delgado C, Baweja M, Burrows NR, et al. Reassessing the inclusion of race in diagnosing kidney diseases: an interim report from the NKF-ASN Task Force. Am J Kidney Dis. 2021;78(1):103–15. doi: 10.1053/j.ajkd.2021.03.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Ann Intern Med. 1999;130(6):461–70. doi: 10.7326/0003-4819-130-6-199903160-00002. [DOI] [PubMed] [Google Scholar]
- 3.Levey AS, Stevens LA, Schmid CH, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med. 2009;150(9):604–12. doi: 10.7326/0003-4819-150-9-200905050-00006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Jones CA, McQuillan GM, Kusek JW, et al. Serum creatinine levels in the US population: Third National Health and Nutrition Examination Survey [erratum Am J Kidney Dis 2000;35:178] Am J Kidney Dis. 1998;32(6):992–9. doi: 10.1016/s0272-6386(98)70074-5. [DOI] [PubMed] [Google Scholar]
- 5.Martin T. The color of kidneys. Am J Kidney Dis. 2011;58(5):xxvii–xxviii. doi: 10.1053/j.ajkd.2011.08.018. [DOI] [PubMed] [Google Scholar]
- 6. Roberts D. The Problem with Race-Based Medicine. TEDMED 2015. Available at: https://www.ted.com/speakers/dorothy_roberts. Accessed November 27, 2022.
- 7.Eneanya ND, Yang W, Reese PP. Reconsidering the consequences of using race to estimate kidney function. JAMA. 2019;322(2):113–4. doi: 10.1001/jama.2019.5774. [DOI] [PubMed] [Google Scholar]
- 8.Morris JE, Grubbs V, Hahn M, Richmond S. Abolish Race-Based Medicine in Kidney Disease and Beyond. San Franciso Examiner. November 27, 2019. Available at: https://www.sfexaminer.com/our_sections/forum/abolish-race-based-medicine-in-kidney-disease-and-beyond/article_9b9483a8-ff61-508b-8d5f-1a0f1709dd8e.html. Accessed November 27, 2022. [Google Scholar]
- 9.Richmond S. How Race-Based Medicine Continues to Harm Black Patients. Medium.com. November 17, 2019. Available at: https://medium.com/@sprichmond2/how-race-based-medicine-continues-to-harm-black-patients-af9f068fd10d. Accessed November 27, 2022. [Google Scholar]
- 10.Vyas DA, Eisenstein LG, Jones DS. Hidden in plain sight—reconsidering the use of race correction in clinical algorithms. N Engl J Med. 2020;383(9):874–82. doi: 10.1056/NEJMms2004740. [DOI] [PubMed] [Google Scholar]
- 11.Amutah C, Greenidge K, Mante A, et al. Misrepresenting race—the role of medical schools in propagating physician bias. N Engl J Med. 2021;384:872–8. doi: 10.1056/NEJMms2025768. doi: 10.1056/NEJMms2025768. [DOI] [PubMed] [Google Scholar]
- 12.Kolata G. Many Medical Decision Tools Disadvantage Black Patients. New York Times. July 17, 2020. Available at: https://www.nytimes.com/2020/06/17/health/many-medical-decision-tools-disadvantage-black-patients.html. Accessed November 27, 2022. [Google Scholar]
- 13.Sehgal AR. Race and the false precision of glomerular filtration rate estimates. Ann Intern Med. 2020;173(12):1008–9. doi: 10.7326/M20-4951. [DOI] [PubMed] [Google Scholar]
- 14.Neal RE, Morse M. Racial Health Inequities and Clinical Algorithms A Time for Action. CJASN. 2021;16(7):1120–1. doi: 10.2215/CJN.01780221. doi: 10.2215/CJN.01780221. Available at: https://journals.lww.com/cjasn/Fulltext/2021/07000/Racial_Health_Inequities_and_Clinical_Algorithms_.23.aspx. Accessed November 27, 2022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. New Generation of Doctor's Fight to Dismantle Vestiges of Racism in Medicine. NBC News. June 16, 2020. Available at: https://www.nbcnews.com/now/video/new-generation-of-doctor-s-fight-to-dismantle-vestiges-of-racism-in-medicine-85171781938 Accessed November 27, 2022.
- 16. Chadha N, Lim B, Kane M, Rowland B. Towards the Abolition of Biological Race in Medicine: Transforming Clinical Education, Research and Practice. Institute for Healing and Justice in Medicine. Availabe at: https://www.instituteforhealingandjustice.org/. Accessed November 27, 2022.
- 17.Kuehn BM. Medical students lead effort to remove race from kidney function estimates. KidneyNews. July. 2020;12(7):1–3. [Google Scholar]
- 18.Grubbs V. Precision in GFR reporting: let's stop playing the race card. CJASN. 2020;15(8):1201–2. doi: 10.2215/CJN.00690120. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Petition for Racial Justice in eGFR Reporting at UCSF and SFGH. Available at: https://docs.google.com/document/d/1KDelXTpZiz5UXaZhQ2V-1nqSrTagwXTz6Md18hVD3fI/edit. Accessed November 27, 2022.
- 20. Maripuri S. @saugarmaripuri. Twitter tweet. June 25, 2020. Available at: https://twitter.com/saugarmaripuri/status/1276340844841308160. Accessed November 27, 2022.
- 21.Hsu CY, Go AS. The race coefficient in glomerular filtration rate-estimating equations and its removal. Curr Opin Nephrol Hypertens. 2022;31(6):527–33. doi: 10.1097/MNH.0000000000000833. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Eneanya ND, Boulware LE, Tsai J, et al. Health inequities and the inappropriate use of race in nephrology. Nature Reviews Nephrology. 2022;18(2):84–94. doi: 10.1038/s41581-021-00501-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Diao JA, Wu GJ, Taylor HA, et al. Clinical implications of removing race from estimates of kidney function. JAMA. 2021;325(2):184–6. doi: 10.1001/jama.2020.22124. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Norris KC, Eneanya ND, Boulware LE. Removal of race from estimates of kidney function: first, do no harm. JAMA. 2021;325(2):135–7. doi: 10.1001/jama.2020.23373. [DOI] [PubMed] [Google Scholar]
- 25.Ojo A. Eliminating racial inequities in kidney health: much more than revising estimating equations. Ann Intern Med. 2022 doi: 10.7326/M21-4875. 10.7326/M7321-4875. [DOI] [PubMed] [Google Scholar]
- 26.Velasco JD, Snodgrass B. The use of race in eGFR: why racial justice requires accuracy. Am J Med. 2021;134(7):827–8. doi: 10.1016/j.amjmed.2021.02.013. [DOI] [PubMed] [Google Scholar]
- 27.Galishoff M. Look at the Evidence, Not the Politics. JAMA. September 2, 2020. Available at: https://jamanetwork.com/journals/jama/fullarticle/2769035. Accessed November 27, 2022. [Google Scholar]
- 28.Inker LA, Schmid CH, Tighiouart H, et al. Estimating glomerular filtration rate from serum creatinine and cystatin C. N Engl J Med. 2012;367(1):20–9. doi: 10.1056/NEJMoa1114248. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Delgado C, Baweja M, Crews DC, et al. A unifying approach for GFR estimation: recommendations of the NKF-ASN Task Force on reassessing the inclusion of race in diagnosing kidney disease. J Am Soc Nephrol. 2021;32(12):2994–3015. doi: 10.1681/ASN.2021070988. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Delgado C, Baweja M, Crews DC, et al. A unifying approach for GFR estimation: recommendations of the NKF-ASN Task Force on reassessing the inclusion of race in diagnosing kidney disease. Am J Kidney Dis. 2022;79(2):268–88. doi: 10.1053/j.ajkd.2021.08.003. [DOI] [PubMed] [Google Scholar]
- 31.Hsu CY, Yang W, Parikh RV, et al. Race, genetic ancestry, and estimating kidney function in CKD. N Engl J Med. 2021;385(19):1750–60. doi: 10.1056/NEJMoa2103753. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Inker LA, Eneanya ND, Coresh J, et al. New creatinine- and cystatin C-based equations to estimate GFR without race. N Engl J Med. 2021;385(19):1737–49. doi: 10.1056/NEJMoa2102953. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Delanaye P, Mariat C, Cavalier E, Glassock RJ, Gemenne F, Pottel H. The “race” correction in estimating glomerular filtration rate: an European point of view. Curr Opin Nephrol Hypertens. 2021;30(6):525–30. doi: 10.1097/MNH.0000000000000739. [DOI] [PubMed] [Google Scholar]
- 34. Labcorp Implements Race-Free Equation for Kidney Disease Test. Labcorp. March 2, 2022. Available at: https://www.labcorp.com/labcorp-implements-race-free-equation-kidney-disease-test. Accessed December 1, 2022.
- 35.Genzen JR, Souers RJ, Pearson LN, et al. Reported awareness and adoption of 2021 estimated glomerular filtration rate equations among US clinical laboratories, March 2022. JAMA. 2022;328(20):2060–2. doi: 10.1001/jama.2022.15404. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. The End of Black Kidney Function. Thenephrologist.com. March 9, 2021. Available at: https://www.thenephrologist.com/blog/2021/3/9/the-end-of-black-kidney-function. Accessed December 1, 2022.
- 37.Powe NR. Race and kidney function: the facts and fix amidst the fuss, fuzziness, and fiction. Med-Cambridge. 2022;3(2):93–7. doi: 10.1016/j.medj.2022.01.011. [DOI] [PubMed] [Google Scholar]