Article

Incidence and Progression of Chronic Kidney Disease in

Black and White Individuals with Type 2 Diabetes
Claire Gerber,1,2 Xuan Cai,1 Jungwha Lee,1 Timothy Craven,3 Julia Scialla ,4 Nao Souma,1 Anand Srivastava,1,2
Rupal Mehta,1,2 Amanda Paluch,5 Alexander Hodakowski,1 Rebecca Frazier,2 Mercedes R. Carnethon,5
Myles Selig Wolf,4 and Tamara Isakova1,2

Abstract 1
Center for
Background and objectives Type 2 diabetes and associated CKD disproportionately affect blacks. It is uncertain Translational
if racial disparities in type 2 diabetes-associated CKD are driven by biologic factors that influence propensity to Metabolism and
CKD or by differences in type 2 diabetes care. Health, Institute for
Public Health and
Design, setting, participants, & measurements We conducted a post hoc analysis of 1937 black and 6372 white Medicine, 2Division of
Nephrology and
participants of the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial to examine associations of Hypertension,
black race with change in eGFR and risks of developing microalbuminuria, macroalbuminuria, incident CKD Department of
(eGFR,60 ml/min per 1.73m2, $25% decrease from baseline eGFR, and eGFR slope ,21.6 ml/min per 1.73 m2 per Medicine, and
5
year), and kidney failure or serum creatinine .3.3 mg/dl. Department of
Preventive Medicine,
Feinberg School
Results During a median follow-up that ranged between 4.4 and 4.7 years, 278 black participants (58 per 1000 of Medicine,
person-years) and 981 white participants (55 per 1000 person-years) developed microalbuminuria, 122 black Northwestern
participants (16 per 1000 person-years) and 374 white participants (14 per 1000 person-years) developed University, Chicago,
macroalbuminuria, 111 black participants (21 per 1000 person-years) and 499 white participants (28 per 1000 Illinois; 3Department
of Biostatistical
person-years) developed incident CKD, and 59 black participants (seven per 1000 person-years) and 178 Sciences, Wake Forest
white participants (six per 1000 person-years) developed kidney failure or serum creatinine .3.3 mg/dl. School of Medicine,
Compared with white participants, black participants had lower risks of incident CKD (hazard ratio, 0.73; 95% Winston-Salem, North
confidence intervals, 0.57 to 0.92). There were no significant differences by race in eGFR decline or in risks of Carolina; and
4
Division of
microalbuminuria, macroalbuminuria, and kidney failure or of serum creatinine .3.3 mg/dl. Nephrology,
Department of
Conclusions Black participants enrolled in a randomized controlled trial had lower rates of incident CKD Medicine, Duke
compared with white participants. Rates of eGFR decline, microalbuminuria, macroalbuminuria, and kidney University Medical
failure did not vary by race. Center, Duke
University, Durham,
Clin J Am Soc Nephrol 13: 884–892, 2018. doi: https://doi.org/10.2215/CJN.11871017 North Carolina

Introduction
Type 2 diabetes mellitus is a major public health
problem and the leading cause of blindness, ampu-
tations and ESKD (1). Glycemic control, multifactorial
risk factor modification, and self-management are
the major corner stones of comprehensive treatment
approaches to reduce the risks of long-term compli-
cations in patients with type 2 diabetes (2). Despite
significant progress in type 2 diabetes care, retinop-
athy, neuropathy, nephropathy, and cardiovascular
disease continue to impose a significant burden on
affected individuals, their families, and the health
care system (3).
The prevalence of type 2 diabetes is almost twice as
high in non-Hispanic blacks as compared with non-
Hispanic whites (1), and blacks have a disproportion-
ate burden of diabetes-related complications (4). After
development of CKD, blacks with type 2 diabetes
have a risk of progression to ESKD that is approxi-
mately two to three-fold fold higher compared with
Correspondence:
Dr. Tamara Isakova,
whites (4–9). Accelerated deterioration of kidney Feinberg School of
function in black individuals with type 2 diabetes Medicine,
may be influenced by biologic factors that influence Northwestern
propensity to CKD and its severity or by differences University, 633 N. St.
Clair Street, Suite
in type 2 diabetes care (10,11). Previous studies
18-089, Chicago, IL
identified genetic predisposition, low birth weight, hy- 60611. Email: tamara.
pertension, obesity, low socioeconomic status, poor isakova@
access to high-quality health care, and high-risk northwestern.edu
health behaviors as contributors to CKD progres-
sion (4,12–15).
Randomized clinical trials provide protocol-driven
care that may eliminate or reduce differences in care
delivery and quality across groups, and may provide
an opportunity to study racial differences in CKD
development and progression. To our knowledge, an
evaluation of racial differences in kidney outcomes
in a type 2 diabetes population with low prevalence of
CKD at baseline and within the context of a random-
ized controlled trial has not been performed. We
conducted a post hoc analysis on a subset of the Action

884 Copyright © 2018 by the American Society of Nephrology www.cjasn.org Vol 13 June, 2018
Clin J Am Soc Nephrol 13: 884–892, June, 2018
to Control Cardiovascular Risk in Diabetes (ACCORD) trial
population that self-identified as non-Hispanic white or
non-Hispanic black. We examined the associations of black
race with longitudinal change in eGFR and with risks of
developing microalbuminuria, macroalbuminuria, incident
CKD, and kidney failure or serum creatinine .3.3 mg/dl.
We hypothesized that compared with white participants,
black participants with type 2 diabetes who received stan-
dardized multifactorial type 2 diabetes care within the context
of a randomized controlled trial would have faster eGFR
decline and be at greater risk of development and progression
of CKD during follow-up in the ACCORD trial.
Materials and Methods
Brief Description of the ACCORD Trial
The ACCORD trial was a randomized, multicenter,
double 232 factorial, parallel treatment trial that enrolled
10,251 high-risk participants with type 2 diabetes across
seven clinical center networks. The trial evaluated the
effects of intensive versus standard glycemic control,
fibrates versus placebo, and intensive versus standard BP
control on major cardiovascular disease events. The study
design, inclusion and exclusion criteria, predefined micro-
vascular outcomes and their frequency of assessment, and
results have been previously reported (16–18). Major
eligibility criteria were age between 40 and 79 years old,
known type 2 diabetes, defined according to the 1997
American Diabetes Association criteria (19), glycated he-
moglobin level of 7.5% or higher, type 2 diabetes duration of
.3 months, and presence of at least two cardiovascular
disease risk factors, high likelihood of cardiovascular
disease or history of cardiovascular disease. Individuals
with serum creatinine .1.5 mg/dl within 2 months before
the screening visit were excluded. Randomization occurred
from 2001 to 2005. Because of increased all-cause mortality,
the glycemic intervention was terminated in 2008 (17). The
lipid and BP interventions were completed in 2009. Written
informed consent was collected from all ACCORD trial
participants, and institutional review board approval was
obtained at all sites.
Study Population
We used ACCORD trial research materials obtained from
the National Heart, Lung, and Blood Institute to analyze data
of 8309 ACCORD trial participants who self-identified as
non-Hispanic blacks (n=1937) and non-Hispanic whites
(n=6372).
Exposure and Outcomes
We examined associations of black race with longitu-
dinal change in eGFR and with time-to-development of
microalbuminuria, macroalbuminuria, incident CKD, and
kidney failure or serum creatinine .3.3 mg/dl (18).
Among individuals with eGFR$60 ml/min per 1.73 m2
and no microalbuminuria (,30 mg/g creatinine) at the
baseline visit, we defined incident CKD as the new onset of
eGFR,60 ml/min per 1.73 m2 after the baseline visit,
$25% decrease from baseline, and an eGFR slope of
,21.6 ml/min per 1.73 m2 per year, which was the
median yearly change in eGFR calculated from baseline
and end of follow-up eGFR values.
CKD by Race in Diabetes, Gerber et al. 885
Measurements and Assessment of Baseline Covariates
Standardized questionnaires at the baseline visit evalu-
ated demographic characteristics, type 2 diabetes duration,
smoking status, and medical and medication history.
Height, weight, and BP were collected following a stan-
dardized protocol. eGFR was estimated from serum creat-
inine, measured by the Roche Creatinine Plus enzymatic
method (Roche Diagnostics, Basel, Switzerland), using the
CKD Epidemiology Collaboration equation (20). Urine
creatinine was measured enzymatically on a Roche Double
Modular P Analytics automated analyzer. Urinary albumin
was measured by immunonephelometry on a Siemens BN
II nephelometer. Hemoglobin A1c (HbA1c) was determined
by automated high-performance liquid chromatography.
Microalbuminuria and macroalbuminuria were defined
as urinary albumin-to-creatinine ratio (UACR) $30 and
$300 mg/g, respectively. To account for missing covariate
data (smoking, 13% and retinopathy, 12%), we added a
category of missing for these two covariates. The remain-
ing covariates had minimal missing data (,5%).
Statistical Analyses
We compared baseline characteristics of the study
population according to race. Statistical significance was
determined using t test for continuous variables with normal
distribution and chi-squared tests for categorical variables. To
ascertain racial differences in delivered care within the
context of the ACCORD trial, we compared achieved
HbA1c according to glycemia intervention arm, achieved
systolic BP according to BP intervention arm, and use of
angiotensin-converting enzyme inhibitors and angiotensin
receptor blockers. To determine whether there were racial
differences in outcome assessments either due to missed
visits or differences in survival, we compared the number
of serum creatinine measurements during follow-up and
the incidence rate of mortality by race.
Because examination of mean eGFR during follow-up
revealed that the eGFR slope varied with time (Figure 1),
we used linear mixed models (21) with separate slopes for
the period from baseline to 24 months and the period of 24
months to end of follow-up to examine differences between
black versus white participants in longitudinal change in
eGFR. To account for early hemodynamic changes in eGFR,
we derived adjusted mean annualized change in eGFR during
entire follow-up from a piecewise linear mixed model (21) with
two knots. The two knots were positioned at month 4 and
month 24, which allowed for different slopes before and after
the knots. We included race3time, race3first spline for time
and race3second spline for time interaction terms in the
piecewise linear mixed model. All models included a random
intercept for each participant and a random slope for time as a
continuous variable to account for within-participant correla-
tion. In model 1, we adjusted for trial interventions, including
randomized glycemia, BP, and lipid arms, and the seven
clinical center networks. In model 2, we adjusted for factors in
model 1 and for demographics, including age, sex, education,
and health insurance. In model 3, we adjusted for factors in
model 2 and for kidney-specific factors, including baseline
eGFR and presence of microalbuminuria and macroalbumi-
nuria at baseline. In model 4, we adjusted for factors in model 3
and for other baseline clinical factors, including systolic BP,
body mass index, HbA1c, smoking status, type 2 diabetes
886 Clinical Journal of the American Society of Nephrology

outcomes started at the baseline visit. We stratified the Cox

Figure 1. | The rate of change in eGFR over time was comparable in
black and white ACCORD trial participants. Mean absolute follow-
up values are shown for black (dashed line) and white (solid line)
participants. Error bars indicate SEM.
duration, history of heart failure, history of cardiovascular
disease (myocardial infarction, stroke, revascularization, or
angina), history of retinopathy, and use of medications, in-
cluding angiotensin-converting enzyme inhibitors, angioten-
sin receptor blockers, insulin, and thiazolidinediones.
We applied Cox proportional-hazards regression models
to analyze the associations of race with risks of microalbu-
minuria, macroalbuminuria, incident CKD, and kidney failure
or serum creatinine .3.3 mg/dl. Follow-up time for all
models by trial interventions, including randomized gly-
cemia, BP, and lipid arms, and the seven clinical center
networks. Then, we adjusted the Cox models for the same
covariates included in the linear mixed models. We report
hazard ratios (HRs) with 95% confidence intervals (95% CIs)
with white participants as the reference category. There was
no violation of the proportional hazards assumption, using
the Schoenfeld residual, for the effect of race.
All analyses were performed using SAS version 9.4 (SAS
Institute, Cary, NC). All statistical tests were two-sided,
and P values ,0.05 were considered statistically signifi-
cant. Because our objective in this post hoc analysis was to
obtain hypothesis-generating results, we did not adjust for
multiple testing. Reported P values are of nominal signif-
icance and serve as guides for possible associations.
Results
Baseline Characteristics of ACCORD Trial Participants
Consistent with ACCORD trial eligibility criteria (17), the
study population included middle-aged and older partic-
ipants with type 2 diabetes and associated complications
(Table 1). The mean age was 6367 years, the mean duration
of type 2 diabetes was 1168 years, and 35% of participants
had cardiovascular disease at baseline. Compared with
white participants, black participants had significantly
higher HbA1c levels (8.561.1 versus 8.260.9; P value
,0.001) and higher prevalence of retinopathy (16% versus

Table 1. Baseline characteristics of black and white ACCORD trial participants

Variable Black Participants n=1937 White Participants n=6372

Age, yr 6266 6367

Female, n (%) 967 (50) 2156 (34)
Education
Less than high school graduate, n (%) 401 (21) 631 (10)
High school graduate (or GED), n (%) 591 (31) 1635 (26)
Some college or technical school, n (%) 615 (32) 2286 (36)
College graduate or more, n (%) 326 (17) 1817 (29)
Insurance coverage, n (%) 1622 (84) 5746 (90)
Systolic BP, mm Hg 139617 135616
Body mass index, kg/m2 3265 3365
Hemoglobin A1c, % 8.561.1 8.260.9
Current smoking, n (%) 792 (48) 3168 (57)
Duration of diabetes, yr 1168 1167
Heart failure, n (%) 91 (5) 329 (5)
Cardiovascular disease, n (%) 580 (30) 2357 (37)
Baseline retinopathy, n (%) 260 (16) 588 (10)
Baseline eGFR, ml/min per 1.73 m2 87619 83617
Baseline serum creatinine, mg/dl 1.060.2 0.960.2
Microalbuminuria, n (%) 525 (28) 1541 (25)
Macroalbuminuria, n (%) 151 (8) 370 (6)
Urinary albumin-to-creatinine ratio, mg/g 15 (7–59) 14 (7–42)
Prevalent CKD, n (%) 756 (39) 2337 (37)
Angiotensin-converting enzyme inhibitor, angiotensin 1392 (72) 4424 (70)
receptor blocker use, n (%)
Insulin use, n (%) 390 (20) 1280 (20)
Thiazolidinedione use, n (%) 348 (18) 1511 (24)
Randomization to intensive glycemic control arm, n (%) 989 (51) 3184 (50)
Randomization to intensive BP arm, n (%) 546 (28) 1411 (22)
Randomization to fenofibrate arm, n (%) 387 (20) 1812 (28)

Value presented as number (percentage), mean6SD, or median (interquartile range). ACCORD, Action to Control Cardiovascular
Risk in Diabetes; GED, General Equivalency Diploma.
Clin J Am Soc Nephrol 13: 884–892, June, 2018 CKD by Race in Diabetes, Gerber et al. 887

Figure 2. | Intended reductions in mean HbA1C (A) and mean systolic BP (B) were attained in black and white ACCORD trial participants.
Mean absolute follow-up values for HbA1c (A) and systolic BP (B) are shown for black (dashed line) and white (solid line) participants. Error bars
indicate SEM.

10%; P value ,0.001). History of cardiovascular disease was
more common in white compared with black participants
(37% versus 30%; P value ,0.001), despite lower systolic BP
in white participants (135616 versus 139617; P value
,0.001). There were no significant differences by race in
allocation to intensive glycemic control arm. Imbalances in
other trial interventions were expected because of the 232
factorial design, which entailed that a nonrandom subset of
participants was included in the lipid trial and the remaining
participants were included in the BP trial.
In the entire study population, the average baseline
eGFR was 84617 ml/min per 1.73 m2. Compared with
white participants, black participants had higher baseline
eGFR (87619 ml/min per 1.73 m2 versus 83617 ml/min
per 1.73 m2; P value ,0.001). In contrast, microalbuminuria
and macroalbuminuria were more common in black
compared with white participants (28% versus 25% and
8% versus 6%; P value =0.02 and 0.002, respectively). The
prevalence of CKD, as defined by eGFR,60 ml/min per
1.73 m2 or UACR$30 mg/g at the baseline visit, was
888 Clinical Journal of the American Society of Nephrology

white participants (ten versus 11), and there was no

difference between black participants and white partici-
pants in the incidence rate of mortality during the study (14
events per 1000 person-years follow-up; 95% CI, 12 to 17
versus 16 events per 1000 person-years follow-up; 95% CI,
14 to 17; P value =0.35).

Change in eGFR over Time

Figure 3. | Use of angiotensin-converting enzyme inhibitors and
angiotensin receptor blockers at each study visit did not differ by race
in ACCORD trial participants. Percentage of use of angiotensin-
converting enzyme inhibitors and angiotensin receptor blockers by
month in black (gray bars) and white (black bars) participants.
similar in black and white participants (39% versus 37%;
Table 1).
Achievement of HbA1c and Systolic BP Targets, Use of
Renoprotective Medications, Number of Follow-Up Serum
Creatinine Assessments, and Survival
Although mean values were modestly higher among
black participants compared with white participants, rapid
achievement of HbA1c and systolic BP targets was accom-
plished in black and white participants, and these targets
were maintained in both groups during follow-up (Figure
2). Throughout the duration of the trial, use of angiotensin-
converting enzyme inhibitors and angiotensin receptor
blockers was similar between white and black participants
(Figure 3). The average number of serum creatinine levels
obtained during follow-up was comparable in black and
Because the rate of change in eGFR was steeper between
baseline and month 24 compared with the rest of follow-up,
in addition to testing for differences in changes during the
entire follow-up, we examined slopes before and after 24
months (Figure 1). Unadjusted and adjusted mean annu-
alized changes in eGFR from baseline to month 24 and from
month 24 through to the end of follow-up are summarized
according to race in Table 2. During both time segments,
unadjusted and adjusted eGFR slopes did not differ sig-
nificantly between the groups. Similarly, adjusted mean
annualized change in eGFR during entire follow-up, de-
rived from a piecewise linear mixed model with two knots
at month 4 and month 24, was comparable in black and
white participants (21.45 and 21.79 ml/min per 1.73 m2
per year, respectively; P value =0.18).
Development and Progression of CKD
During a median follow-up that ranged between 4.4 and
4.7 years, 278 black participants (58 per 1000 person-years)
and 981 white participants (55 per 1000 person-years)
developed microalbuminuria, 122 black participants (16
per 1000 person-years) and 374 white participants (14 per
1000 person-years) developed macroalbuminuria, 111
black participants (21 per 1000 person-years) and 499
white participants (28 per 1000 person-years) developed
incident CKD, and 59 black participants (seven per 1000
person-years) and 178 white participants (six per 1000
person-years) developed kidney failure or serum creatinine

Table 2. Mean annualized change in eGFR by time periods in black and white ACCORD trial participants

Mean Annualized Change of eGFR, ml/min per 1.73 m2 per yr

P Value
(95% Confidence Interval)

Model Black, n=1937 White, n=6372

Month 0–Month 24
Unadjusted 23.59 (24.21 to 22.98) 23.94 (24.13 to 23.74) 0.11
Model 1 23.63 (24.24 to 23.02) 23.98 (24.17 to 23.78) 0.11
Model 2 23.65 (24.26 to 23.04) 23.98 (24.17 to 23.79) 0.12
Model 3 23.71 (24.36 to 23.07) 23.89 (24.09 to 23.68) 0.44
Model 4 23.91 (24.64 to 23.19) 24.10 (24.33 to 23.88) 0.46
Black, n=1862 White, n=6158
Month 24– End of Follow-Up
Unadjusted 20.44 (20.80 to 20.08) 20.38 (20.49 to 20.26) 0.63
Model 1 20.46 (20.81 to 20.10) 20.39 (20.51 to 20.28) 0.62
Model 2 20.45 (20.81 to 20.09) 20.39 (20.51 to 20.28) 0.67
Model 3 20.46 (20.82 to 20.09) 20.42 (20.54 to 20.31) 0.79
Model 4 20.38 (20.77 to 0.01) 20.43 (20.56 to 20.31) 0.69

Model 1: adjusts for glycemia trial, intensive BP trial, standard BP trial, lipid fenofibrate trial, network. Model 2: adjusts for factors in
model 1 and for demographics: age, female, education, and health insurance. Model 3: adjusts for factors in model 2 and for kidney-
specific factors: baseline eGFR, microalbuminuria, and macroalbuminuria. Model 4: adjusts for factors in model 3 and for other baseline
clinical factors: systolic BP, body mass index, hemoglobin A1c, smoking status, type 2 diabetes duration, history of heart failure, history of
cardiovascular disease (myocardial infarction, stroke, revascularization, or angina), history of retinopathy, and baseline use of med-
ications (angiotensin-converting enzymes, angiotensin receptor blockers, insulin, and thiazolidinediones). ACCORD, Action to Control
Cardiovascular Risk in Diabetes.
Clin J Am Soc Nephrol 13: 884–892, June, 2018 CKD by Race in Diabetes, Gerber et al. 889

Table 3. Risks of kidney outcomes in black versus white ACCORD participants

Development of Development of Kidney

Incident
Outcomes Microalbuminuria Macroalbuminuria Failure or
CKDa
(UAlb$30 mg/g)a (UAlb$300 mg/g)b SCr.3.3 mg/dl

N events/total N 1259/5276 496/7164 610/5215 237/8294

N events/total N, 278/1124 122/1590 111/1180 59/1937
black
N events/total N, 981/4152 374/5574 499/4035 178/6357
white

Median follow-up 4.4 4.7 4.7 4.7

time, yr
Incidence rate Incidence Rate P Value Incidence Rate P Value Incidence Rate P Value Incidence Rate P Value
events per 1000
person-years
follow-up
(95%
confidence
interval)
Black 58 (51 to 65) 0.48 16 (14 to 20) 0.18 21 (18 to 26) 0.008 7 (5 to 9) 0.48
White 55 (52 to 58) 14 (13 to 16) 28 (26 to 31) 6 (5 to 7)
Hazard Ratio P Value Hazard Ratio P Value Hazard Ratio P Value Hazard Ratio P Value
Unadjusted 1.04 (0.91 to 1.19) 0.53 1.15 (0.94 to 1.41) 0.17 0.76 (0.62 to 0.93) 0.009 1.11 (0.83 to 1.49) 0.50
Model 1 1.01 (0.87 to 1.17) 0.91 1.08 (0.87 to 1.35) 0.50 0.76 (0.60 to 0.95) 0.02 0.94 (0.69 to 1.29) 0.71
Model 2 1.01 (0.87 to 1.17) 0.92 1.13 (0.90 to 1.42) 0.28 0.70 (0.56 to 0.88) 0.002 0.94 (0.68 to 1.31) 0.73
Model 3 1.06 (0.91 to 1.23) 0.48 1.15 (0.92 to 1.45) 0.23 0.72 (0.57 to 0.90) 0.005 0.93 (0.67 to 1.31) 0.69
Model 4 1.05 (0.90 to 1.22) 0.57 1.15 (0.91 to 1.47) 0.25 0.73 (0.57 to 0.92) 0.009 0.90 (0.63 to 1.27) 0.53

Model 1: stratified by glycemia trial, intensive BP trial, standard BP trial, lipid fenofibrate trial, lipid placebo trial, and network. Model 2: stratified by glycemia trial,
intensive BP trial, standard BP trial, lipid fenofibrate trial, lipid placebo trial, and network, and adjusts for demographics: age, female, education, and health insurance.
Model 3: stratified by glycemia trial, intensive BP trial, standard BP trial, lipid fenofibrate trial, lipid placebo trial, and network, and adjusts for factors in model 2 and for
kidney-specific factors: baseline eGFR, microalbuminuria, and macroalbuminuria. Model 4: stratified by glycemia trial, intensive BP trial, standard BP trial, lipid
fenofibrate trial, lipid placebo trial, and network, and adjusts for factors in model 3 and for other baseline clinical factors: systolic BP, body mass index, hemoglobin A1c,
smoking status, type 2 diabetes duration, history of heart failure, history of cardiovascular disease (myocardial infarction, stroke, revascularization, or angina), history of
retinopathy, and baseline use of medications (angiotensin-converting enzymes, angiotensin receptor blockers, insulin, and thiazolidinedione). ACCORD, Action to
Control Cardiovascular Risk in Diabetes; UAlb, urinary albumin; SCr, serum creatinine.
a
No microalbuminuria, macroalbuminuria adjustment in model 3.
b
No macroalbuminuria adjustment in model 3.

.3.3 mg/dl (Table 3). In unadjusted analyses, compared
with white participants, black participants were not at
increased risk of microalbuminuria, macroalbuminuria,
and kidney failure or serum creatinine .3.3 mg/dl, and
their risk for incident CKD was reduced (Figure 4, Table 3).
After multivariable adjustment, risks of development of
microalbuminuria, macroalbuminuria, and kidney failure
or serum creatinine .3.3 mg/dl did not differ significantly
between the two groups (Table 3). Adjusted HRs for
incident CKD showed that black participants had lower
risks of these events compared with white participants
(Table 3).
Discussion
In this secondary analysis of the ACCORD trial, contrary
to our hypothesis, we found that during a median follow-
up period of 4–5 years, black race was not associated with
accelerated eGFR decline, and that, compared with white
participants, black participants had lower rates of incident
CKD. Despite higher prevalence of microalbuminuria and
macroalbuminuria in black compared with white partici-
pants at baseline, during follow-up there were no racial
differences in the development of albuminuria. Although
available follow-up time was limited, we did not identify
any significant differences between groups in risk of
progression of established CKD to kidney failure or serum
creatinine .3.3 mg/dl. Our results suggest that delivery of
standardized type 2 diabetes care before development of
CKD may lead to similar short-term kidney outcomes in
black and white individuals with type 2 diabetes. These
findings have implications for treatment strategies aimed at
prevention and management of CKD in type 2 diabetes.
The disproportionate burden of kidney disease in blacks
with type 2 diabetes is well described (4–15). Epidemiologic
studies have demonstrated that compared with whites,
blacks are more likely to have albuminuria, experience
early kidney function decline, and progress rapidly to
ESKD (4–12). According to US Renal Data System, a na-
tional data registry on ESKD population in the United
States, rates of ESKD due to diabetes are nearly three-fold
greater among blacks compared with whites (22). Although
many factors likely contribute to racial disparities in type 2
diabetes-associated CKD (4,12–15), recent discoveries of
the association of kidney risk variants in the gene encoding
apoL1 with accelerated progression of CKD in those with
and without diabetes (23) have refocused attention on the
effect of genetic background. Prior studies demonstrating
persistent racial disparities in settings of comparable health
care access (7,24–26) have also implicated the primacy of
biologic factors over factors related to delivery of type 2
diabetes care.
By demonstrating that black and white ACCORD trial
participants had comparable kidney outcomes, we now
provide evidence in support of the beneficial effects of
comprehensive type 2 diabetes care on eradicating racial
disparities in development and progression of type 2
diabetes-associated CKD. Our findings differ from earlies
890 Clinical Journal of the American Society of Nephrology

Figure 4. | Compared to white ACCORD trial participants, blacks were not at increased risk of microalbuminuria, macroalbuminuria, and
kidney failure, and their risk for incident CKD was reduced. Proportion of black (dashed line) and white (solid line) ACCORD trial participants
free from (A) microalbuminuria, (B) macroalbuminuria, (C) incident CKD, and (D) kidney failure or serum creatinine .3.3 mg/dl.

studies (7,24–26), likely because we studied a population
with low prevalence of CKD and within the context of a
randomized clinical trial, which assured that existing
standard of care was delivered to all participants. We
found that, in spite of having more risk factors for adverse
CKD outcomes (higher HbA1 and BP, higher prevalence
of retinopathy and albuminuria), black participants had
lower rates of eGFR-driven end points, and there were no
racial differences in development of albuminuria-driven
end points and of kidney failure or serum creatinine .3.3
mg/dl. Although higher eGFR in black participants at
onset of follow-up may explain the reduced risks of eGFR-
driven end points that we observed (27), the totality of our
findings suggests that strategies that effectively deliver
standardized type 2 diabetes care may reduce racial
disparities in type 2 diabetes-associated CKD. This hy-
pothesis is supported by 54% reduction in rate of ESKD in
American Indians and Alaska Natives with type 2 diabetes
after implementation of the Indian Health Service first
Diabetes Standards of Care (28). By actively decreasing
BP and HbA1c, implementing routine reporting of eGFR
and yearly monitoring of UACR and prescription of
angiotensin-converting enzyme inhibitors and angioten-
sin receptor blockers (29), in addition to providing clinical
education programs and culturally relevant patient edu-
cation materials (30), the program eradicated long-standing
disparities in kidney outcomes in American Indians and
Alaska Natives with type 2 diabetes. Another setting where
equivalent access to health care was associated with ab-
sence of racial differences in the incidence of coronary heart
disease and ischemic stroke is the US Veterans Health Ad-
ministration (31). Reports of comparable compliance with
achievement of quality of care indicators for black and white
patients with stages 3 and 4 CKD receiving care in the
Department of Defense health system are also encouraging
(32), and suggest that delivery of standardized preventative
care is achievable.
Strengths of our analysis include evaluation of multiple
kidney outcomes, including change in eGFR, incident
albuminuria, strictly defined incident CKD, and kidney
failure, which we assessed during a median follow-up
time of 4–5 years. Additionally, we were able to adjust for
multiple covariates, including baseline retinopathy and
type 2 diabetes duration. We acknowledge some
Clin J Am Soc Nephrol 13: 884–892, June, 2018
limitations. Because we conducted a post hoc analysis, our
results are hypothesis generating. Our findings are further
limited by the low prevalence of CKD in the study
population and by the short duration of follow-up. The
ACCORD trial was terminated early because of the higher
mortality in the group receiving intensive glycemic ther-
apy, which resulted in limited total number of kidney
failure events (17). However, we were able to evaluate
other kidney outcomes. We were not able to assess
differences in access to and quality of care before trial
enrollment. Nevertheless, despite possible differences in
antecedent care, we did not find racial differences in
outcomes during the trial. To estimate kidney function, we
used the CKD Epidemiology Collaboration equation.
Although this method is likely imprecise for our study
population (33,34), we relied on change in eGFR for our
outcome definitions. Therefore, inherent error of cross-
sectional eGFR assessments was likely minimized. Finally,
this study is limited to middle-aged and older patients
with type 2 diabetes who are at high risk of cardiovascu-
lar disease and who enrolled in a clinical trial. Our results
may not be generalizable to a younger diabetic population,
patients with nondiabetic kidney disease and to diabetic
individuals in the community who are unlikely to volun-
teer to participate in a study.
We found that in a large racially diverse cohort of adults
with type 2 diabetes and low prevalence of CKD who
received protocol-driven type 2 diabetes care, black race was
not associated with accelerated development and progres-
sion of CKD. The results suggest that delivery of standard-
ized care for patients with type 2 diabetes may reduce racial
disparities in type 2 diabetes-associated CKD. Future studies
are needed to determine whether results similar to those
achieved in the American Indians and Alaska Natives with
type 2 diabetes could be achieved for the black race with
delivery of standardized type 2 diabetes care.
Acknowledgments
This work was supported by grants R01DK102438 (T.I.),
R01DK110087 (T.I.), R01DK081374 (M.S.W.), R01DK076116 (M.S.W.),
R01DK094796 (M.S.W.), K24DK093723 (M.S.W.), U01DK099930 (M.S.W.
and T.I.), and T32DK007169 (C.G.) from the National Institutes of
Health, a Strategically Focused Research Network Center Grant on
Health Disparities from the American Heart Association (M.R.C.,
M.S.W., and T.I.), and a Donald E. Wesson Research Fellowship from
the American Society of Nephrology (C.G.).
This manuscript was prepared using Action to Control Cardio-
vascular Risk in Diabetes (ACCORD) trial research materials ob-
tained from the National Heart, Lung, and Blood Institute (NHLBI)
Biologic Specimen and Data Repository Information Coordinating
Center and does not necessarily reflect the opinions or views of the
ACCORD trial or the NHLBI. The funders had no role in the design,
analysis, or interpretation of data, preparation and review of the
manuscript or in the decision to submit the manuscript for publi-
cation.
Disclosures
T.I. received grant support from Shire. M.S.W. has served as a
consultant or received honoraria from Akebia, Amag, Amgen, Arde-
lyx, Diasorin, Incyte, Keryx, Luitpold, Pfizer, Sanofi, and Ultragenyx
and received grant support from Shire. R.M. has interest in Abbot
Laboratories, AbbVie, Inc. and Teva Pharmaceuticals Industries Ltd.
CKD by Race in Diabetes, Gerber et al. 891
References
1. US Department of Health and Human Services Centers for
Disease Control and Prevention: National diabetes fact sheet:
National estimates and general information on diabetes and
prediabetes in the United States, 2014. Available at: https://www.
cdc.gov/diabetes/pubs/statsreport14/national-diabetes-report-
web.pdf. Accessed July 19, 2017
2. Inzucchi SE, Bergenstal RM, Buse JB, Diamant M, Ferrannini E,
Nauck M, Peters AL, Tsapas A, Wender R, Matthews DR; Amer-
ican Diabetes Association (ADA); European Association for the
Study of Diabetes (EASD): Management of hyperglycaemia in type
2 diabetes: A patient-centered approach. Position statement of
the American Diabetes Association (ADA) and the European
Association for the Study of Diabetes (EASD). Diabetologia 55:
1577–1596, 2012
3. Centers for Disease Control and Prevention: Chronic disease
prevention and health promotion, 2016. Available at: https://
www.cdc.gov/chronicdisease/resources/publications/aag/
diabetes.htm. Accessed July 19, 2017
4. Krop JS, Coresh J, Chambless LE, Shahar E, Watson RL, Szklo M,
Brancati FL: A community-based study of explanatory factors for
the excess risk for early renal function decline in blacks vs whites
with diabetes: The Atherosclerosis Risk in Communities study.
Arch Intern Med 159: 1777–1783, 1999
5. Brancati FL, Whittle JC, Whelton PK, Seidler AJ, Klag MJ: The
excess incidence of diabetic end-stage renal disease among
blacks. A population-based study of potential explanatory factors.
JAMA 268: 3079–3084, 1992
6. Collins AJ, Foley RN, Herzog C, Chavers B, Gilbertson D, Ishani A,
Johansen K, Kasiske B, Kutner N, Liu J, St Peter W, Ding S, Guo H,
Kats A, Lamb K, Li S, Li S, Roberts T, Skeans M, Snyder J, Solid C,
Thompson B, Weinhandl E, Xiong H, Yusuf A, Zaun D, Arko C,
Chen SC, Daniels F, Ebben J, Frazier E, Hanzlik C, Johnson R,
Sheets D, Wang X, Forrest B, Constantini E, Everson S, Eggers P,
Agodoa L: US renal data system 2012 annual data report. Am J
Kidney Dis 61(1 Suppl 1): A7, e1–e476, 2013
7. Young BA, Katon WJ, Von Korff M, Simon GE, Lin EH,
Ciechanowski PS, Bush T, Oliver M, Ludman EJ, Boyko EJ: Racial
and ethnic differences in microalbuminuria prevalence in a
diabetes population: The pathways study. J Am Soc Nephrol 16:
219–228, 2005
8. Nicholas SB, Kalantar-Zadeh K, Norris KC: Socioeconomic
disparities in chronic kidney disease. Adv Chronic Kidney Dis 22:
6–15, 2015
9. United States Renal Data System: USRDS annual data report:
Epidemiology of kidney disease in the United States, 2016.
Available at https://www.usrds.org/2016/view/Default.aspx.
Accessed July 19, 2017
10. Patzer RE, McClellan WM: Influence of race, ethnicity and
socioeconomic status on kidney disease. Nat Rev Nephrol 8:
533–541, 2012
11. Udler MS, Nadkarni GN, Belbin G, Lotay V, Wyatt C, Gottesman O,
Bottinger EP, Kenny EE, Peter I: Effect of genetic African ancestry on
eGFR and kidney disease. J Am Soc Nephrol 26: 1682–1692, 2015
12. Fedewa SA, McClellan WM, Judd S, Gutiérrez OM, Crews DC:
The association between race and income on risk of mortality in
patients with moderate chronic kidney disease. BMC Nephrol 15:
136, 2014
13. Evans K, Coresh J, Bash LD, Gary-Webb T, Köttgen A, Carson K,
Boulware LE: Race differences in access to health care and dis-
parities in incident chronic kidney disease in the US. Nephrol Dial
Transplant 26: 899–908, 2011
14. Iyengar SK, Sedor JR, Freedman BI, Kao WH, Kretzler M, Keller BJ,
Abboud HE, Adler SG, Best LG, Bowden DW, Burlock A, Chen
YD, Cole SA, Comeau ME, Curtis JM, Divers J, Drechsler C,
Duggirala R, Elston R, Guo X, Huang H, Hoffmann MM, Howard
BV, Ipp E, Kimmel PL, Klag MJ, Knowler WC, Kohn OF, Leak TS,
Leehey DJ, Li M, Malhotra A, März W, Nair V, Nelson RG,
Nicholas SB, O’Brien SJ, Pahl MV, Parekh RS, Pezzolesi MG,
Rasooly RS, Rotimi CN, Rotter JI, Schelling JR, Seldin MF, Shah
VO, Smiles AM, Smith MW, Taylor KD, Thameem F3, Thornley-
Brown DP, Truitt BJ, Wanner C, Weil EJ, Winkler CA, Zager PG, Igo
RP Jr ., Hanson RL, Langefeld CD; Family Investigation of Ne-
phropathy and Diabetes (FIND): Genome-wide association and
trans-ethnic meta-snalysis for advanced diabetic kidney disease:
892 Clinical Journal of the American Society of Nephrology
Family Investigation of Nephropathy and Diabetes (FIND). PLoS
Genet 11: e1005352, 2015
15. Crook ED, Patel SR: Diabetic nephropathy in African-American
patients. Curr Diab Rep 4: 455–461, 2004
16. Buse JB, Bigger JT, Byington RP, Cooper LS, Cushman WC,
Friedewald WT, Genuth S, Gerstein HC, Ginsberg HN, Goff
DC Jr ., Grimm RH Jr ., Margolis KL, Probstfield JL, Simons-Morton
DG, Sullivan MD; ACCORD Study Group: Action to Control
Cardiovascular Risk in Diabetes (ACCORD) trial: Design and
methods. Am J Cardiol 99[12A]: 21i–33i, 2007
17. Gerstein HC, Miller ME, Byington RP, Goff DC Jr ., Bigger JT, Buse
JB, Cushman WC, Genuth S, Ismail-Beigi F, Grimm RH Jr .,
Probstfield JL, Simons-Morton DG, Friedewald WT; Action to
Control Cardiovascular Risk in Diabetes Study Group: Effects of
intensive glucose lowering in type 2 diabetes. N Engl J Med 358:
2545–2559, 2008
18. Ismail-Beigi F, Craven T, Banerji MA, Basile J, Calles J, Cohen RM,
Cuddihy R, Cushman WC, Genuth S, Grimm RH Jr., Hamilton BP,
Hoogwerf B, Karl D, Katz L, Krikorian A, O’Connor P, Pop-Busui R,
Schubart U, Simmons D, Taylor H, Thomas A, Weiss D, Hramiak I;
ACCORD trial group: Effect of intensive treatment of hyperglycaemia
on microvascular outcomes in type 2 diabetes: An analysis of the
ACCORD randomised trial. Lancet 376: 419–430, 2010
19. American Diabetes Association: Clinical practice recommen-
dations 1997. Diabetes Care 20[Suppl 1]: S1–S70, 1997
20. Levey AS, Stevens LA, Schmid CH, Zhang YL, Castro AF 3rd,
Feldman HI, Kusek JW, Eggers P, Van Lente F, Greene T, Coresh J;
CKD-EPI (Chronic Kidney Disease Epidemiology Collaboration):
A new equation to estimate glomerular filtration rate. Ann Intern
Med 150: 604–612, 2009
21. Fitzmaurice G, Laird NM, Ware JH: Applied Longitudinal Anal-
ysis, New York, Wiley, 2004
22. Ozieh MN, Dismuke CE, Lynch CP, Egede LE: Medical care expen-
ditures associated with chronic kidney disease in adults with diabetes:
United States 2011. Diabetes Res Clin Pract 109: 185–190, 2015
23. Parsa A, Kao WH, Xie D, Astor BC, Li M, Hsu CY, Feldman HI,
Parekh RS, Kusek JW, Greene TH, Fink JC, Anderson AH, Choi MJ,
Wright JT Jr ., Lash JP, Freedman BI, Ojo A, Winkler CA, Raj DS,
Kopp JB, He J, Jensvold NG, Tao K, Lipkowitz MS, Appel LJ; AASK
Study Investigators; CRIC Study Investigators: APOL1 risk variants,
race, and progression of chronic kidney disease. N Engl J Med 369:
2183–2196, 2013
24. Young BA, Maynard C, Boyko EJ: Racial differences in diabetic
nephropathy, cardiovascular disease, and mortality in a national
population of veterans. Diabetes Care 26: 2392–2399, 2003
25. Choi AI, Karter AJ, Liu JY, Young BA, Go AS, Schillinger D: Ethnic
differences in the development of albuminuria: The DISTANCE
study. Am J Manag Care 17: 737–745, 2011
26. Lewis EF, Claggett B, Parfrey PS, Burdmann EA, McMurray JJ,
Solomon SD, Levey AS, Ivanovich P, Eckardt KU, Kewalramani R,
Toto R, Pfeffer MA: Race and ethnicity influences on cardiovas-
cular and renal events in patients with diabetes mellitus. Am Heart
J 170: 322–329, 2015
27. Isakova T, Craven TE, Lee J, Scialla JJ, Xie H, Wahl P, Marcovina
SM, Byington RP, Wolf M: Fibroblast growth factor 23 and
incident CKD in type 2 diabetes. Clin J Am Soc Nephrol 10: 29–38,
2015
28. Bullock A, Burrows NR, Narva AS, Sheff K, Hora I, Lekiachvili A,
Cain H, Espey D: Vital signs: Decrease in incidence of diabetes-
related end-stage renal disease among American Indians/Alaska
natives - United States, 1996-2013. MMWR Morb Mortal Wkly
Rep 66: 26–32, 2017
29. Narva AS: Reducing the burden of chronic kidney disease among
American Indians. Adv Chronic Kidney Dis 15: 168–173, 2008
30. Sequist TD, Cullen T, Acton KJ: Indian health service innovations
have helped reduce health disparities affecting american Indian
and alaska native people. Health Aff (Millwood) 30: 1965–1973,
2011
31. Kovesdy CP, Norris KC, Boulware LE, Lu JL, Ma JZ, Streja E, Molnar
MZ, Kalantar-Zadeh K: Association of race with mortality and
cardiovascular events in a large cohort of US veterans. Circulation
132: 1538–1548, 2015
32. Gao SW, Oliver DK, Das N, Hurst FP, Lentine KL, Agodoa LY,
Sawyers ES, Abbott KC: Assessment of racial disparities in
chronic kidney disease stage 3 and 4 care in the department of
defense health system. Clin J Am Soc Nephrol 3: 442–449,
2008
33. Delanaye P, Mariat C, Maillard N, Krzesinski JM, Cavalier E: Are
the creatinine-based equations accurate to estimate glomerular
filtration rate in African American populations? Clin J Am Soc
Nephrol 6: 906–912, 2011
34. Stevens LA, Claybon MA, Schmid CH, Chen J, Horio M, Imai E,
Nelson RG, Van Deventer M, Wang HY, Zuo L, Zhang YL,
Levey AS: Evaluation of the Chronic Kidney Disease Epide-
miology Collaboration equation for estimating the glomerular
filtration rate in multiple ethnicities. Kidney Int 79: 555–562,
2011
Received: October 20, 2017 Accepted: March 1, 2018
Published online ahead of print. Publication date available at www.
cjasn.org.
See related editorial, “Race in America: What Does It Mean for
Diabetes and CKD?,” on pages 829–830.