Reconsidering the Consequences of Using Race to Estimate Kidney Function
Reconsidering the Consequences of Using Race
to Estimate Kidney Function
Clinicians estimate kidney function to guide impor-
tant medical decisions across a wide range of settings,
including assessing the safety of radiology studies,
choosing chemotherapy, and reviewing the use of com-
mon nonprescription medications such as nonsteroidal
anti-inflammatory drugs. Because direct measure-
ment of kidney function is infeasible at the bedside, the
usual approach involves using estimating equations that
ely on serum creatinine. These equations assign a highe
estimated glomerular filtration rate (eGFR) to patients
who are identified as black. Yet in some medical and so-
cial science disciplines, a consensus has emerged that
ace is a social construct rather than a biological one.1 In
this Viewpoint, we argue that the use of kidney func-
tion estimating equations that include race as a vari-
able cause problems for transparency and unduly re-
strict access to care in some cases, yet offer only modest
enefits to precision.
Estimated GFR equations fulfill an important need
for clinicians to conveniently assess kidney function
and, secondarily, for public health authorities to assess
the prevalence of kidney disease. These equations,
such as the Chronic Kidney Disease Epidemiology Col-
laboration equation (CKD-EPI) and its predecessor, the
Modification of Diet in Renal Disease Study (MDRD)
equation, were generated in large cohorts of individu-
als who underwent gold-standard measurement of
“true” GFR by infusing iothalamate or another chemical
into the blood and quantifying its urine clearance.
Investigators found that black race was independently
associated with a slightly higher GFR at the same serum
creatinine level. This association has been justified by
the assertion that black individuals release more creati-
nine into the blood, perhaps because of more muscle
mass, although data remain inconclusive.2-4 The CKD-
EPI equation includes a race coefficient that increases
the eGFR in black patients by about 16%. Estimated
GFR equations also include age and sex because olde
individuals and women, on average, have less muscle
than younger individuals and men, respectively; these
generalizations have a stronger empirical basis than
that for race.
Classifying patients according to ancestry (rathe
than race or ethnicity) has legitimate purposes to iden-
tify individuals at risk of complications from rare gene
mutations like sickle cell trait or cystic fi
osis. How-
ever, eGFR equations are distinct because they instead
assert that existing organ function is different between
individuals who are otherwise identical except for race.
Population studies reveal only small differences in
gene distributions between racial groups while show-
ing greater variation between individuals of the same
ace. Meanwhile, the history of medicine offers abun-
dant evidence that racial categories were often gener-
ated a
itrarily and at times implemented to reinforce
social inequality.5
Racial categorization is often used in a nonstandard-
ized way. Consider a hypothetical 50-year-old woman
with a creatinine level of 2.0 mg/dL and no proteinuria.
Her father self-identified as black race and her mothe
self-identified as white race. If this patient is admitted
to the hospital, an administrator or clinician may assess
the patient’s skin tone or hair and label her as black in
the medical record. Alternatively, the patient may be
asked to identify her race. Yet she would have no way
to know that her answer would affect assessments of he
organ function or treatment. Furthermore, 3% of indi-
viduals in the United States identified as multiracial in the
2010 Census, whereas in Brazil and some other coun-
tries, the multiracial category exceeds
one-third of the population. Decision
support provides little guidance about
how to calculate the patient’s eGFR if she
is biracial, refuses to answer the ques-
tion about race, or self-identifies with a
ace that is different than that recorded
in the medical record.
Estimated GFR equations have major clinical
consequences. Many essential medications including
antibiotics are withheld from patients with a low eGFR
or are administered at reduced doses. The authorita-
tive Kidney Disease: Improving Global Outcomes
(KDIGO) guidelines recommend nephrology refe
al
if a patient’s eGFR is less than 30 mL/min/1.73 m2.
If the patient in the above example were considered
to be black, her eGFR would be 33 mL/min/1.73 m2,
ut if she were considered to be white, her eGFR
would be 28 mL/min/1.73 m2 with the CKD-EPI equa-
tion (ie, below the threshold for refe
al). In addition,
clinical trials commonly exclude patients with reduced
kidney function. If this patient were considered to
e black, she could enter some trials that would
exclude her if she were considered to be white.
Perhaps the most concerning implication of race
in eGFR is that it has the potential to reduce access
to kidney transplantation, for which racial disparities
are substantial. In the United States, being wait-listed
for a kidney transplant requires an eGFR of less than
Estimated GFR equations are distinct
ecause they assert that existing organ
function is different between individuals
who are identical except for race.
VIEWPOINT
Nwamaka Denise
Eneanya, MD, MPH
Renal-Electrolyte and
Hypertension Division,
Perelman School of
Medicine, University
of Pennsylvania,
Philadelphia; and
Palliative and
Advanced Illness
Research Center,
Perelman School of
Medicine, University of
Pennsylvania,
Philadelphia.
Wei Yang, PhD
Department of
Biostatistics,
Epidemiology, and
Informatics, Perelman
School of Medicine,
University of
Pennsylvania,
Philadelphia.
Peter Philip Reese,
MD, MSCE
Renal-Electrolyte and
Hypertension Division,
Perelman School of
Medicine, University
of Pennsylvania,
Philadelphia; and
Department of
Biostatistics,
Epidemiology, and
Informatics, Perelman
School of Medicine,
University of
Pennsylvania,
Philadelphia.
Co
esponding
Author: Peter Philip
Reese, MD, MSCE,
Center for Clinical
Epidemiology and
Biostatistics, University
of Pennsylvania,
423 Guardian Dr,
917 Blockley Hall,
Philadelphia, PA 19104
(peter.reese@uphs.
upenn.edu).
Opinion
jama.com (Reprinted) JAMA July 9, 2019 Volume 322, Number 2 113
© 2019 American Medical Association. All rights reserved.© 2019 American Medical Association. All rights reserved.
Downloaded From: https:
jamanetwork.com/ by a Virginia Commonwealth University User on 10/13/2021
mailto: XXXXXXXXXX
mailto: XXXXXXXXXX
http:
www.jama.com/?utm_campaign=articlePDF%26utm_medium=articlePDFlink%26utm_source=articlePDF%26utm_content=jama XXXXXXXXXX
20 mL/min/1.73 m2. If this hypothetical patient were encountered
5 years later and she had a creatinine level of 2.8 mg/dL, her eGFR
would be 18 mL/min/1.73 m2 if she were white and would be
21 mL/min/1.73 m2 if she were black, and she would qualify fo
wait-listing if considered white but would be ineligible for wait-
listing at that time if considered black.
Supporters of the status quo may affirm that taking account of
ace enables more precise estimation of kidney function and is
thereby worthwhile, but historical mistreatment of racial minority
groups suggests that race-based treatment prescriptions need very
strong justification.6 Using race to guide clinical care is justified only
if (1) the use confers substantial benefit; (2) the benefit cannot be
achieved through other feasible approaches; (3) patients who re-
ject race categorization are accommodated fairly; and (4) the use
of race is transparent.
Kidney function equations fail this test. As a result, investiga-
tors should develop new eGFR equations that substitute objective
data such as height and weight for race. The Figure illustrates how
much higher eGFR would be if patients were assigned black race
and emphasizes thresholds at which key clinical decisions are
made. There may be reason for optimism in improving these equa-
tions. One study showed that the race coefficient was reduced
from 20% to 3.3% when body composition variables were added
to the eGFR equation.7
Research is also needed to quantify the benefits and harms of
abandoning race in GFR estimation. If race is excluded, black pa-
tients might have enhanced access to transplantation but also might
eceive inappropriately low antibiotic dosing. If race is excluded, more
lack patients could also be falsely labeled as having kidney dis-
ease or having a more advanced stage of disease, potentially lead-
ing to anxiety or unnecessary treatment. In addition, clinicians should
take opportunities to discuss how race is used with their patients
to more effectively engage in shared decision-making.
For nearly 20 years, eGFR equations have helped clinicians
screen for kidney disease and care for patients. The problems of ra-
cial classification related to eGFR have not been closely examined.
The value of racial labels should be measured and alternatives to
using this variable should be carefully considered before commit-
ting to the same algorithms in the future.
ARTICLE INFORMATION
Published Online: June 6, 2019.
doi:10.1001/jama XXXXXXXXXX
Conflict of Interest Disclosures: Dr Eneanya
eported receipt of lectureship fees from Fresenius
Medical Care North America. Dr Yang reported
eing a statistical editor for the American Journal of
Kidney Diseases. Dr Reese reported receipt of
investigator-initiated grants from Merck, A
Vie,
and CVS Caremark to the University of
Pennsylvania; consulting for Collaborative
Healthcare Research & Data Analytics
(COHRDATA); and being an associate editor for the
American Journal of Kidney Diseases.
Funding/Support: Dr Reese receives support from
the Greenwall Foundation to support bioethics
esearch.
Role of the Funde
Sponsor: The Greenwall
Foundation had no role in the collection,
management, analysis, and interpretation of the
data; preparation, review, or approval of the
manuscript; or decision to submit the manuscript
for publication.
REFERENCES
1. Fontanarosa PB, Bauchner H. Race, ancestry, and
medical research. JAMA. 2018;320(15): XXXXXXXXXX.
doi:10.1001/jama XXXXXXXXXX
2. Gallagher D, Visser M, De Meersman RE, et al.
Appendicular skeletal muscle mass: effects of age,
gender, and ethnicity. J Appl Physiol XXXXXXXXXX;83
(1): XXXXXXXXXXdoi:10.1152/jappl XXXXXXXXXX
3. Barondess DA, Nelson DA, Schlaen SE. Whole
ody bone, fat, and lean mass in black and white
men. J Bone Miner Res. 1997;12(6): XXXXXXXXXXdoi:10.
1359/jbmr XXXXXXXXXX
4. Hsu J, Johansen KL, Hsu CY, Kaysen GA,
Chertow GM. Higher serum creatinine
concentrations in black patients with chronic
kidney disease: beyond nutritional status and body
composition. Clin J Am Soc Nephrol. 2008;3(4):
XXXXXXXXXXdoi:10.2215/CJN XXXXXXXXXX
5. Braun L, Fausto-Sterling A, Fullwiley D, et al.
Racial categories in medical practice: how useful are
they? PLoS Med. 2007;4(9):e271. doi:10.1371
journal.pmed XXXXXXXXXX
6. Institute of Medicine. Unequal Treatment:
Confronting Racial and Ethnic Disparities in Health
Care. Washington, DC: National Academies Press;
2003. doi: XXXXXXXXXX/10260.
7. Anderson AH, Yang W, Hsu CY, et al; CRIC Study
Investigators. Estimating GFR among participants in
the Chronic Renal Insufficiency Cohort (CRIC)
Study. Am J Kidney Dis. 2012;60(2): XXXXXXXXXXdoi:10.
1053/j.ajkd XXXXXXXXXX
Figure. Relationship Between Racial Categories and Estimation
of Kidney Function Across the Spectrum of Chronic Kidney Disease
20
15
10
5
0
Di
ff
e
en
ce
B
et
w
ee
n
2
eG
FR
E
st
im
at
es
,
m
L
m
in
1
.7
3
m
2
Measured GFR, mL/min/1.73 m2
XXXXXXXXXX120
Circles indicate how much higher estimated glomerular filtration rate (eGFR) is
for patients when assigned black race instead of nonblack