Keywords: randomized controlled trials, chronic kidney disease, end-stage kidney disease, kidney transplantation, acute renal failure, racial and ethnic disparities
Abstract
Significance statement
Racial and ethnic disparities in clinical trial enrollment are well described. However, whether these disparities are present in nephrology randomized clinical trials has not been previously reported. We performed a systematic review and meta-analysis of 380 randomized clinical trials involving different aspects of kidney disease published between 2000 and 2021. Our results indicate that worldwide reporting of race and ethnicity is poor and that White individuals account for most of the randomized participants with decreased enrollment of Black participants in more recent trials. However, trials conducted in the United States have representation of Black and Hispanic participants consistent with the population prevalence of disease and under-representation of Asian participants.
Background
Under-representation of racial and ethnic minorities in clinical trials could worsen disparities, but reporting and enrollment practices in nephrology randomized clinical trials have not been described.
Methods
PubMed was searched to capture randomized clinical trials for five kidney disease–related conditions published between 2000 and 2021 in ten high-impact journals. We excluded trials with <50 participants and pilot trials. Outcomes of interest were the proportion of trials reporting race and ethnicity and the proportions of enrolled participants in each race and ethnicity category.
Results
Among 380 trials worldwide, race was reported in just over half and ethnicity in 12%. Most enrolled participants were White, and Black individuals accounted for ≤10% of participants except in dialysis trials where they accounted for 26% of participants. However, Black participants were enrolled at high proportions relative to disease and population prevalence in US CKD, dialysis, and transplant trials representing 19% of participants in AKI, 26% in CKD, 44% in GN, 40% in dialysis, and 26% in transplant trials. Enrollment of Asian participants was low worldwide except in GN trials with marked under-representation in US CKD, dialysis, and transplant trials. Hispanic individuals represented only 13% of participants in US dialysis trials compared with 29% of US dialysis population.
Conclusion
More complete reporting of race and ethnicity in nephrology trials is needed. Black and Hispanic patients are well-represented in kidney disease trials in the United States. Asian patients are poorly represented in kidney trials both globally and in the United States.
Introduction
The global burden of CKD is estimated to be 700 million cases.1 In the United States, CKD is more common in Black and Hispanic patients than White patients, and adjusted risks of ESKD are approximately 25% to 50% higher in those with stage 4 or 5 CKD.2 Disparities in CKD outcomes are further evidenced by a higher rate of in-hospital mortality and progression to ESKD after AKI in Black individuals compared with White individuals and lower odds of receiving preemptive kidney transplants in Black and Hispanic patients.2,6 In addition, geographic variation in post-ESKD mortality is higher among Black, Hispanic, and Asian patients than White patients with ESKD.2–6 These numbers highlight pervasive socioeconomic disparities in the burden of kidney disease.7,8
The reasons for these disparities are complex. Differential exposure to risk factors and differential access to care have been recognized, but differential responses to therapies based on the differences in social determinants of health, genetic, and epigenetic factors across populations with different demographics may also be important.9–13 Representative enrollment of racial and ethnic minorities in nephrology trials is necessary to reduce disparities by ensuring that standard therapies are effective for all patients and that any differences in response are understood.
However, reporting of demographics and enrollment of racial and ethnic minorities in clinical trials has been low in other disciplines.14,15 Several guidance documents and policies regarding the reporting of trial outcomes for racial and ethnic minorities by National Institutes of Health (NIH), US Food and Drug Administration (FDA), and Health and Human services reflect emerging concern with this issue.16 Nevertheless, progress has been slow. In trials of new entities approved by the US Food and Drug Administration between 2015 and 2019, a majority of participants were White (69% to 79%).9 Although disparities in kidney disease care have recently garnered significant attention, we are unaware of any published reporting regarding trends in enrollment of minority populations in nephrology trials.
To better understand trends in trial reporting and enrollment, we performed a systematic review and meta-analysis of the reporting and enrollment of racial and ethnic minorities in nephrology randomized clinical trials published between 2000 and 2021.
Methods
Study Selection
This systematic review was reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement.17 The protocol was registered with the PROSPERO: the International Prospective Register of systematic reviews (CRD42022296363). Eligibility criteria were designed to facilitate the objective of estimating representation in trials likely to influence standards of care. We included randomized controlled trials (RCTs) enrolling participants aged 18 years or older in the following categories: (1) CKD, (2) AKI, (3) GN, (4) maintenance dialysis (hemodialysis [HD] and peritoneal dialysis [PD]), and (5) kidney transplantation. To avoid inclusion of pilot and feasibility trials which would be unlikely to directly influence treatment, we required that trials report clinically relevant outcome measures for inclusion. Outcome measures either primary or secondary could include mortality/survival, CKD progression, cardiovascular events, relapse and remission for GN trials, renal recovery, response to therapy, and rejection for transplant trials. For AKI trials, additional outcomes such as intensive care unit (ICU), ventilation or dialysis-free days, and ICU discharge were included. Studies with <50 participants were excluded a priori as likely to be pilot trials with limited direct effect on practice. Secondary/post hoc analysis of RCTs, protocols of RCTs, meta-analyses, trials that restricted enrollment to a specific racial or ethnic group by design, and those not meeting the adult age cutoff were also excluded.
Search Strategy
A medical librarian (J.N.) designed the literature search criterion in collaboration with the other authors. We searched PubMed/MEDLINE on January 27, 2022, with search terms designed to capture RCTs for the five kidney disease–related conditions described above. To focus on trials likely to influence standards of care, we limited our search to studies published in 10 high-impact journals with high impact factors: American Journal of Kidney Diseases, American Journal of Transplantation, The British Medical Journal, Clinical Journal of American Society of Nephrology, The Journal of American Medical Association, Journal of the American Society of Nephrology, Kidney International, The Lancet, Nephrology, Dialysis and Transplantation, and The New England Journal of Medicine, between January 1, 2000, and December 31, 202. The full search strategy and details of categories and outcomes are included in the Supplementary document.
Screening and Data Collection
All study screening was performed by at least two authors independently using data collection and processing software (Covidence, Veritas Health Innovation, Melbourne, Australia). The initial stage of study screening was by title and abstract, screening out those that were clearly not relevant. The second stage of study screening was by full-text review, using the eligibility criteria developed a priori. Disagreements were resolved by consensus and finalized by the first and last authors. Data were abstracted for key variables by two authors (Q.H.S. and A.M.), including demographics age, sex, race, ethnicity, study design and population, sample size, study geographical region/location, number of sites, and outcomes of interest. Data were abstracted from original articles, supplementary material, protocol articles, and ClinicalTrials.gov website. As needed, authors were contacted for clarification regarding demographics, funding agency, and trial sites. In addition, lead authors of trials conducted in the United States were contacted to clarify enrollment demographics when publications reported racial categories incompletely.
Interventions were categorized as drug, device, dialysis prescription, procedure or surgery, care delivery, exercise/mindfulness, or other. Funding sources were categorized as NIH, non-NIH government (for studies outside of the United States and non-NIH government funding in the United States such as FDA, VA), industry, foundation, or other. Trial enrollment categories included United States only, International, Europe, Asia, and global (both the United States and International enrollment). Publication year was categorized as 2000–2006, 2007–2011, 2012–2016, or 2017–2021. Race and ethnicity categories were consistent with those used in the NIH Physician Health Services form for enrollment in clinical trials: White, Black, Asian, American Indians and Alaska Natives, Native Hawaiian or Other Pacific Islander, or Other. Ethnicity was classified as Hispanic and Not Hispanic or Latino.
Statistical Analysis
Given differences in patient populations, standard categorizations, and restrictions on collecting or reporting participant race or ethnicity in some countries, we analyzed results overall (worldwide) and for trials conducted solely in the United States. Trial characteristics were summarized according to reporting and enrollment of participants, enrollment sites, publication year, trial category, and intervention type. Outcomes of interest included the proportion of trials reporting race and ethnicity of enrolled participants and the proportion of enrolled participants according to categories of race and ethnicity. Additional analyses included proportions across trial categories, enrollment site, and publication date categories. Descriptive statistics are reported as n (%), mean±SD, or median (interquartile range).
Meta-analytic estimates of the proportion of participants were performed using the metaprop command in STATA.18,19 Pooled effect size estimates were calculated using random effects models with restricted maximum likelihood as the heterogeneity estimator. Confidence intervals for the binomial proportion were calculated using exact or Clopper–Pearson methods.18,19 Multivariable generalized linear regression models with logit link function with proportion of enrollment (White race or Hispanic ethnicity) as the dependent variable and publication year, trial category, intervention type, and enrollment region as independent variables were used to assess influence on diversity of enrollment. All trials reporting enrollment numbers for the relevant categories were included in the regression models.
We additionally performed a priori planned subgroup analyses according to trial category using the random effects model to further explore heterogeneity. Finally, for US trials, we qualitatively compared the percentage of enrollment by racial and ethnic categories with estimates of disease prevalence in the United States using data from the National Health and Nutrition Survey (NHANES) for CKD,20 the United States Renal Data System for maintenance dialysis,2 and the United Network for Organ Sharing for kidney transplant.21 We additionally compared enrollment with demographic data from the US census.22 Statistical analysis was performed using STATA version 17 (StataCorp, College Station, TX), and Microsoft excel. P<0.05 was considered significant.
Results
Trial Population, General Characteristics, and Trends in Reporting on Race and Ethnicity Worldwide and in the United States
As shown in Figure 1, the search strategy identified 4494 articles. Of these, 1200 were included in the full-text screening, and 380 randomized controlled trials worldwide met all inclusion criteria including 61 conducted solely in the United States. Total sample size was 350,519 overall and 43,715 for the trials in the United States. Worldwide slightly more than half (207 of 380; 54%) reported information on at least one racial category. Only 45 (12%) reported ethnicity. Categorization of race and ethnicity was self-reported in 42 trials, investigator reported in 11 trials, and not reported in 154. A total of 45 trials reported subgroup analyses according to race in the primary trial publication, whereas seven reported subgroup analyses according to ethnicity. Furthermore, nine trials stratified randomization by race and one by ethnicity.
Figure 1.
A total of 380 trials were included in the meta-analysis and 61 of these were in US alone. PRISMA figure. PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses. Figure 1 can be viewed in color online at www.jasn.org.
In the United States, 56 of 61 (92%) reported information on at least one racial category, and 25 of 61 (41%) reported ethnicity (Table 1, Supplemental Table 1). Race was reported in 100% of US CKD, GN, and dialysis trials, whereas ethnicity was reported in roughly half of trials with the lowest rate in transplant trials (Table 1). For US trials, 93% reported race and 21% ethnicity of participants in 2000–2006 compared with 100% and 65% of trials in 2017–2021. Worldwide, more recently published trials reported racial and ethnic characteristics more frequently with 44% of trials reporting race and 8% reporting ethnicity in 2000–2006 and 60% reporting race and 17% reporting ethnicity in 2017–2021.
Table 1.
Overall Characteristics of US trials, reporting of race and ethnicity
| Overall | Reporting Race, n (%) | Reporting Ethnicity, n (%) | ||
|---|---|---|---|---|
| No. of Trialsa (N=61), n (%) | Sample Size (N=43,715), n (%) | Total=56/61 (92%) | Total=25/61 (41%) | |
| Journal | ||||
| AJKD | 3 (5) | 811 (2) | 2 (67) | 1 (33) |
| AJT | 20 (33) | 4103 (9) | 19 (95) | 6 (30) |
| BMJ | 1 (2) | 6030 (14) | 1 (100) | 1 (100) |
| CJASN | 8 (13) | 3122 (7) | 7 (88) | 3 (38) |
| JAMA | 4 (7) | 4374 (10) | 4 (100) | 2 (50) |
| JASN | 8 (13) | 8046 (18) | 8 (100) | 3 (38) |
| KI | 4 (7) | 2495 (6) | 4 (100) | 1 (25) |
| Lancet | 2 (3) | 2789 (6) | 2 (100) | 2 (100) |
| NDT | 3 (5) | 4222 (10) | 2 (67) | 2 (67) |
| NEJM | 8 (13) | 7363 (17) | 7 (87) | 4 (50) |
| Publication date category | ||||
| 2000–2006 | 14 (23) | 5230 (12) | 13 (93) | 3 (21) |
| 2007–2011 | 8 (13) | 6503 (15) | 7 (88) | 2 (25) |
| 2012–2016 | 22 (36) | 1198 (3) | 19 (86) | 9 (41) |
| 2017–2021 | 17 (28) | 20,784 (48) | 17 (100) | 11 (65) |
| Trial category | ||||
| AKI | 9 (15) | 11,000 (25) | 7 (78) | 4 (44) |
| CKD | 16 (26) | 13,039 (30) | 16 (100) | 9 (56) |
| GN | 2 (3) | 213 (<1) | 2 (100) | 1 (50) |
| HD/PD | 8 (13) | 12,584 (29) | 8 (100) | 3 (38) |
| Transplant | 26 (43) | 6879 (16) | 24 (92) | 8 (31) |
| Funding source | ||||
| NIH | 20 (33) | 23,253 (53) | 20 (100) | 8 (40) |
| Non-NIH governmental | 6 (10) | 5535 (13) | 6 (100) | 4 (67) |
| Industry | 26 (43) | 15,766 (36) | 26 (100) | 10 (38) |
| Foundation | 3 (5) | 592 (1) | 3 (100) | 0 (0) |
| Otherb | 13 (21) | 5114 (12) | 9 (69) | 7 (54) |
| Intervention | ||||
| Care delivery | 11 (18) | 20,356 (47) | 10 (91) | 8 (73) |
| Dialysis | 5 (8) | 3274 (7) | 4 (80) | 1 (20) |
| Drug | 43 (70) | 19,747 (45) | 41 (95) | 15 (35) |
| Procedure/surgery | 2 (3) | 338 (1) | 1 (50) | 1 (50) |
HD, hemodialysis; PD, peritoneal dialysis; NIH, National Institutes of Health.
Percentages in the first two columns represent column percentages. Percentages in the last two columns represent row percentages. For other columns, row percentages are provided as n of trials meeting criterion/N of trials in the category.
Other=investigator/university funded. Some studies had more than one funding source.
Reporting practices of journals were consistent for US trials with five journals reporting race in 100% of trials and a low of 67% in two journals (Nephrology Dialysis and Transplantation and American Journal of Kidney Diseases). Worldwide reporting practices for race differed widely according to journal with 71% of articles published in the New England Journal of Medicine providing information compared with only 6% in Kidney International.
Reporting of Specific Race and Ethnicity Categories in US Trials and Worldwide
Among US trials, 77% reported information on the number of White and Black participants, but only 34% reported the number of Asian participants (Table 2). The proportion of White participants was reported in 85% of transplant trials, whereas the proportion of Black participants was reported in only 27% (Table 2, Figure 2). Completeness of reporting improved in US trials over time with 64% of trials reporting the number of Black participants between 2000 and 2006 and 100% between 2017 and 2021. The proportion of US trials reporting the number of Asian, American Indians and Alaska Natives, and Native Hawaiian or Other Pacific Islander participants was higher for each category in 2017–2021 than 2000–2006. However, only 11% of AKI trials reported the number of enrolled Asian participants. Similarly, no trials in which dialysis was the intervention and less than half of the drug trials reported the number of enrolled Asian participants.
Table 2.
Reporting of racial categories in US trials according to publication year, trial category, and enrollment site
| Category | No. trials,a (N=61), n (%) | White (n=47; 77%), n (%) | Black (n=47; 77%), n (%) | Asian (n=21; 34%), n (%) | AI/AN (n=7; 11%), n (%) | NH/PI (n=4; 7%), n (%) | Other race (n=34; 56%), n (%) |
|---|---|---|---|---|---|---|---|
| Publication category | |||||||
| 2000–2006 | 14 (23) | 12 (86) | 9 (64) | 5 (36) | 2 (14) | 0 (0) | 7 (50) |
| 2007–2011 | 8 (13) | 5 (63) | 6 (75) | 1 (13) | 0 (0) | 0 (0) | 4 (50) |
| 2012–2016 | 22 (36) | 15 (68) | 15 (68) | 6 (27) | 1 (5) | 2 (9) | 10 (45) |
| 2017–2021 | 17 (28) | 15 (88) | 17 (100) | 9 (53) | 4 (24) | 2 (12) | 13 (76) |
| Trial category | |||||||
| AKI | 9 (15) | 4 (44) | 5 (56) | 1 (11) | 0 (0) | 0 (0) | 1 (11) |
| CKD | 16 (26) | 13 (81) | 14 (88) | 7 (44) | 4 (25) | 2 (13) | 11 (69) |
| GN | 2 (3) | 2 (100) | 1 (50) | 1 (50) | 0 (0) | 0 (0) | 0 (0) |
| HD/PD | 8 (13) | 6 (75) | 5 (63) | 5 (63) | 0 (0) | 0 (0) | 6 (75) |
| Transplant | 26 (43) | 22 (85) | 7 (27) | 7 (27) | 2 (8) | 2 (8) | 16 (62) |
| Intervention | |||||||
| Care delivery | 11 (18) | 8 (73) | 9 (82) | 4 (36) | 0 (0) | 0 (0) | 6 (55) |
| Dialysis | 5 (8) | 3 (60) | 2 (40) | 0 (0) | 0 (0) | 0 (0) | 1 (20) |
| Drug | 43 (70) | 35 (81) | 35 (81) | 17 (40) | 7 (16) | 4 (9) | 26 (60) |
| Procedure/surgery | 2 (3) | 1 (50) | 1 (50) | 0 (0) | 0 (0) | 0 (0) | 1 (50) |
HD, hemodialysis; PD, peritoneal dialysis.
Percentages in number of trials column represents column percentages. For other columns, row percentages are provided as n of trials meeting criterion/N of trials in the category.
Figure 2.
Majority of US trials reported demographic information on White and Black participants whereas only about one third reported on Asian participants. Reporting of racial demographic information in US trials according to trial category. Data are provided for reporting of White, Black, and Asian race. Figure 2 can be viewed in color online at www.jasn.org.
Looking at worldwide practices, reporting of race and ethnicity was highest in the global (US and International enrollment) trial category (Supplemental Table 1). Demographic information was often reported incompletely with detail selectively provided for the standard NIH categories (Supplemental Table 2). The proportion of White participants was reported in 195 of 380 (51%) of worldwide trials, whereas the proportion of Black participants was reported in 40%. By contrast, the proportion of participants who were Native Hawaiian and Pacific Islanders was reported in only 13 of 380 (3%) trials (Supplemental Figure 1).
Distribution of Race and Ethnicity among Enrolled Participants in US Trials and Worldwide
Among US trials, enrollment of Asian, American Indians and Alaska Natives, and Native Hawaiian or Other Pacific Islander participants was low throughout the analytic period. White individuals accounted for majority of participants in all trials (Table 3, Figure 3). Most of the participants in transplant trials were White (70%), whereas 26% participants were Black, and no Asian, American Indians and Alaska Natives, or Native Hawaiian participants or Other Pacific Islander participants were enrolled. Reporting of enrollment in AKI trials, in particular, was incomplete (e.g., reporting of the number of Black participants but not the number of White participants) such that the reported distribution in the published data were misleading (Supplemental Table 3).
Table 3.
Distribution of race and ethnicity in US trials
| Category | White N=25,415 Trials=51 |
Black N=11,113 Trials=49 |
Asian N=782 Trials=27 |
AI/AN N=64 Trials=12 |
NH/PI N=23 Trials=7 |
Other race N=1589 Trials=39 |
Hispanic N=4406 Trials=27 |
Non-Hispanic N=30,465 Trials=27 |
|---|---|---|---|---|---|---|---|---|
| Publication category | ||||||||
| 2000–2006 | 2905 (58) | 1792 (35) | 191 (4) | 18 (0) | 2 (0) | 132 (3) | 475 (12) | 3452 (88) |
| 2007–2011 | 2919 (51) | 2435 (43) | 17 (0) | 0 (0) | 0 (0) | 350 (6) | 328 (10) | 2852 (90) |
| 2012–2016 | 6306 (71) | 2097 (24) | 175 (2) | 4 (0) | 11 (0) | 266 (3) | 852 (10) | 7352 (90) |
| 2017–2021 | 13,285 (69) | 4789 (25) | 399 (2) | 42 (0) | 10 (0) | 841 (4) | 2751 (7) | 17,481 (93) |
| Trial category | ||||||||
| AKI | 6998 (73) | 1853 (19) | 177 (2) | 14 (0) | 7 (0) | 577 (6) | 809 (8) | 9134 (92) |
| CKD | 8050 (69) | 3049 (26) | 177 (1) | 24 (0) | 5 (0) | 382 (4) | 1898 (16) | 9542 (84) |
| GN | 91 (51) | 79 (44) | 8 (4) | 0 (0) | 0 (0) | 0 (0) | 28 (20) | 112 (80) |
| HD/PD | 6230 (53) | 4645 (40) | 383 (3) | 11 (0) | 0 (0) | 411 (4) | 1104 (12) | 8391 (88) |
| Transplant | 4046 (70) | 1487 (26) | 37 (0) | 15 (0) | 11 (0) | 219 (4) | 567 (12) | 3958 (88) |
| Intervention | ||||||||
| Care delivery | 12,585 (22) | 4548 (25) | 412 (2) | 0 (0) | 7 (0) | 753 (4) | 1976 (10) | 17,780 (90) |
| Dialysis | 1592 (53) | 1335 (44) | 54 (2) | 11 (0) | 0 (0) | 33 (1) | 185 (6) | 2784 (94) |
| Drug | 11,179 (64) | 5216 (30) | 316 (2) | 37 (0) | 16 (0) | 776 (4) | 2226 (17) | 10,492 (83) |
| Procedure/surgery | 59 (59) | 14 (14) | 0 (0) | 0 (0) | 0 (0) | 27 (27) | 19 (19) | 81 (81) |
Additional demographic data were obtained from authors when primary race categories were incompletely reported in the original publication. HD, hemodialysis; PD, peritoneal dialysis.
Figure 3.
White participants account for the majority of enrolled participants across all trial categories in the US. Participant demographics in US trials as reported in published articles (A) and demographics after obtaining additional data from authors (B). Figure 3 can be viewed in color online at www.jasn.org.
Among worldwide trials, White individuals accounted for most of the enrolled participants in all trial categories reporting race information (Supplemental Table 4): AKI 87%, CKD 74%, GN 42% HD/PD 63%, and transplant 81% (Supplemental Figure 2). Black individuals accounted for ≤10% of enrolled participants except in dialysis trials (26%). Enrollment of Asian individuals was low except in GN trials (39%). There were very few Native Hawaiian or Other Pacific Islander participants, while 4% to 9% of the trial participants were described as Other race—a designation often used to combine low-frequency categories including American Indians and Alaska Natives. Enrollment of Hispanic individuals was 13% overall, and it was low in CKD, AKI, and kidney transplant trials (2%, 6%, and 11% of participants, respectively).
In metaregressions, publication year, trial category, or intervention were not significantly associated with the proportion of White participants enrolled in US trials, whereas publication year (P<0.001) and trial category (P=0.01) were significantly associated with the proportion of enrolled participants who were White in worldwide trials (Supplemental Table 5 and 6) with White participants accounting for a higher percentage of enrolled participants and Black participants for a decreased proportion in more recent trials (Supplemental Figure 3). None of the tested characteristics were independently associated with enrollment of Hispanic participants (Supplemental Table 7 and 8).
Meta-analytic Estimates
All trials reporting enrollment information on race or ethnicity were included in a meta-analysis to better estimate proportional enrollment after accounting for the number of trials and trial size. The results of the weighted estimates were qualitatively similar to the crude estimates (Supplemental Tables 9 and 10). Heterogeneity quantified using I2 and Tau2 was high both for trials in the United States and overall with significant heterogeneity in trial demographics across trial type (data not shown).
Comparison of Enrollment with US Population Prevalence
White individuals accounted for 69% of participants enrolled in CKD trials from the United States compared with an estimated 67% of the CKD population (based on NHANES) and 59% of the overall US population (Figure 4A). By contrast, Black individuals were over-represented compared with estimates of population prevalence, and Asian individuals were under-represented. The proportion of Hispanic individuals in CKD trials was similar to their proportion of the US population but higher than their proportion in the US CKD population.20,22
Figure 4.
Enrollment of Asian participants was lower compared to disease prevalence or US population estimate. Comparison of enrollment by race or ethnicity in CKD, dialysis, and transplant trials and disease prevalence or demographics of overall US population from US census. (A) CKD trials; (B) dialysis trials; (C) transplant trials. Figure 4 can be viewed in color online at www.jasn.org.
By contrast, Black patients accounted for a disproportionate share of participants in US dialysis trials (42%) compared with the demographics of the US dialysis population between 1999 and 2019 (30% Black individuals). Enrollment of Asian and Hispanic participants was lower than their share of the US dialysis population (Figure 4B).2 Finally, White individuals accounted for a disproportionate share of participants in US kidney transplant trials whereas Asian participants lagged behind in comparison with their representation in the US transplant population.21 (Figure 4C).
Discussion
Under-representation of racial and ethnic minorities in clinical trials has been widely reported. Similar disparities have been presumed to be true for kidney disease trials and hypothesized to be a source of inequities in CKD outcomes. However, data to this effect have been lacking. To the best of our knowledge, this is the first published systemic analysis and description of the reporting of race and ethnic enrollment information for nephrology randomized controlled trials.
There were several important findings: Information on the race of enrolled participants is reported in most of the US trials. However, ethnicity reporting remains low and is provided in <50% of trials. Moreover, trials frequently fail to report enrollment information for the full spectrum of standard race and ethnicity categories, and among US trials, only 2 of 61 (3%) reported complete race information. By contrast, among trials worldwide, information on race is reported in only slightly more than half of kidney disease trials, primarily for White and Black participants, whereas data on ethnicity were reported in only 12%. Reporting has improved over time, but data on the number of participants who are Asian, native Hawaiian/Pacific Islanders, or American Indian/Alaska Natives remain infrequently reported. Contrary to our initial hypothesis, although White participants account for most of the trial participants both in the United States and worldwide, nephrology trials conducted in the United States seem to enroll relatively high proportions of Black and Hispanic individuals, but they markedly underenroll Asian patients. By contrast, the proportion of Black participants in nephrology trials worldwide seems to have decreased dramatically during the past two decades.
The selection of a study population in a clinical trial is critical because scientific rigor and clinical utility may be compromised if it is not representative of the disease being treated. Low reporting rates and unequal enrollment in clinical trials may thus directly affect ability to care for patients. This is particularly true for CKD because racial and ethnic minority populations in the United States are disproportionately affected compared with White populations, and there has been increasing attention to the negative effect of differences in social determinants of health on CKD outcomes. Our analysis highlights striking deficiencies in reporting of important demographic characteristics of participants in nephrology trials with incomplete reporting across categories for trials in the United States and low rates of reporting for trials conducted worldwide.
Incomplete reporting of race and ethnicity directly compromise the science diminishing its effect for the overall population nephrologists who are committed to serve and inhibits the ability of medical professionals to determine whether findings can be extrapolated to the populations they see. The underlying reasons for the low reporting of race and ethnicity in nephrology trials conducted outside of the United States and differential reporting across trial categories and publications are uncertain. However, greater awareness of the effect of racial and ethnic disparities on health outcomes in the United States, mandated collection of information on race and ethnicity by US funding and regulatory bodies, and prohibition of collection of race by some European agencies likely contribute. In addition, there may less of a perceived need to report this information for trials conducted in regions with more homogeneous populations than the United States. Conversely, we are unaware of any editorial policies that would explain variation in reporting. Requirements for universal reporting of standard categories at publication could readily address this issue and should be considered—at least for trials recruiting in the United States where populations are diverse and regulatory and funding agencies universally encourage such reporting.
Trends for enrollment of under-represented minorities seem to be more complex. In trials conducted in the United States, Black participants are enrolled at a higher proportion in CKD, dialysis, and transplant trials compared with the prevalence of disease—a relatively reassuring finding. Patients with Hispanic ethnicity were also enrolled in high proportions, except in US dialysis trials in which they are markedly underenrolled. Worldwide, participation of Black individuals has decreased over time—a finding potentially explained by a higher proportion of global trials (which may enroll in countries with few Black residents) conducted in recent years, although a similar trend was notable in the United States. Several socioeconomic factors have been proposed to explain the disparities in dialysis care among Hispanic or Latinx individuals including access to care and language barriers.23 Our findings suggest that these factors may have a similar effect on enrollment of Hispanic patients, at least in dialysis trials. Addressing language barriers might be particularly effective. For example, in one nonkidney disease study, a dedicated Spanish language line increased enrollment of Hispanic participants.24
In addition, we found strikingly low enrollment of Asian participants particularly in AKI, dialysis, and transplant trials compared with the disease prevalence for US trials. Low representation of Asian participants in US clinical trials has been reported previously.25 The factors underlying low enrollment of Asian participants are unclear because little research has been performed in this area in the United States. However, in one qualitative study, most of the participants reported that they knew nothing or only a little about clinical trials, and most did not report fluency in English language. Conversely, none of the participants reported religious or cultural influence on trial participation.26 Another study interviewed South Asian participants in the United Kingdom regarding clinical trials and reported that barriers seemed to include age, social class, and low fluency in English. The authors reported that South Asian participants were systematically excluded from trials because of the increased cost, time, and resources associated with their inclusion.27 Similarly, in a recent study, the prevalence and correlates of invitation to participate in clinical trials among a nationally representative sample of US adults showed Asian respondents had the lowest prevalence of invitation (2%)—a trend persistent from prior such analyses.28 Addressing language barriers to enrollment is likely to have the most immediate yield on enrollment of Asian participants in nephrology trials and might mirror the widespread availability of language services commonly used in hospital care. However, given the variety of languages and dialects spoken, additional, broader efforts are likely to be necessary to achieve full equity in enrollment.
Our analysis of worldwide nephrology trials is consistent with the reporting and enrollment disparities identified in cardiology, oncology, and neurology trials.14,28,29 Consistently, systematic reviews have found that Black, Hispanic, and Asians patients are under-represented in trials compared with their disease burden. Lack of diversity in trial cohorts and enrolling homogenous populations without regard to disease prevalence may generate results not generalizable to the patients seen in routine practice and that undermine confidence of patients and providers in trial results. While our overall results are similar—with lower reporting and enrollment of the minority groups in international and global trials—they differ markedly with regard to the representative enrollment of Black individuals in US trials while showing persistent gaps in enrollment of Hispanic individuals in US dialysis trials and Asian individuals in multiple types of trials in the US. Quantifying these inequalities in the setting of nephrology trials and providing more granular understanding is a necessary first step toward change and improving outcomes in the future.
The underlying reasons for the high enrollment of Black individuals in US nephrology trials, particularly dialysis trials, are uncertain, but they may reflect US policies requiring collection and reporting of information on race as well as more uniform access to care and clinical trials and enhanced trust in providers and trial personnel in the setting of maintenance dialysis compared with other clinical settings: Specifically, we hypothesize that the regular contact of patients on dialysis with the usual recruiting sites (the dialysis unit) and associated personnel three times a week facilitates development of trust in the research and care teams. The lack of time for relationship building may underlie the somewhat lower enrollment of Black and Hispanic patients in ICU and AKI trials. Similarly, minority patients may have delayed diagnosis of kidney disease and lower access to a nephrologist earlier in the course of disease and thus lack access to nephrology care and trials or have lower levels of trust compared with some non-Hispanic White patients.
Careful consideration on enrollment strategies could mitigate barriers. Generically, approaches targeting enrollment might include making translation services readily available. Attention to hiring and training research investigators and coordinators from different backgrounds with the knowledge of the cultural norms and a sensitivity toward different belief systems could overcome prevalent stereotypes and biases and may be another key factor. Other approaches include designs that decrease entry barriers such as pragmatic trials and using local trusted community-based health advocates, using plain language and engaging study-related material with clear instructions on the consent process.30–33 Understanding regional enrollment and demographic patterns, better information on the demographics of different diseases (such as AKI), and incorporating dedicated enrollment metrics with active monitoring during recruitment may also be needed, particularly for global trials.
Several limitations of our analysis should be noted. We restricted our systematic review to 10 high-impact journals to focus on trials with the greatest likelihood of publishing practice changing results. Similarly, to eliminate pilot and feasibility trials, trial outcomes were required to meet rigid criteria that reflected clinically relevant or hard end points. We also required a sample size >50 patients, which might particularly have influenced the types of GN trials included, although only two GN trials were excluded based on this requirement at the full article review stage. Finally, we focused on trials for five specific conditions that are major foci of nephrology research but do not encompass the entire field. Our results may not be generalizable to trials not meeting these criteria, which collectively play an important role in advancing the field of nephrology. Comparisons with population prevalence should be interpreted cautiously but still provide important qualitative context on size of the gap between disease prevalence and trial enrollment. The low reporting of race and ethnicity in trials with enrollment outside of the United States might be reflective of local policies and rules about reporting of race and ethnicity over the study period.
In addition, most of the trials did not specify how race and ethnicity information were collected (e.g. self versus investigator reported). The binary categories of race and ethnicity used are based on US mandates and definitions. Although standard, they ignore more biologically relevant parameters and increasing recognition of race as a social construct. The potential for harm from using these constructs clinically should be recognized and results extrapolated cautiously—alternatives to race that more accurately capture the nuanced features underlying differences in disease progression and variation in outcomes are clearly needed. However, given the extensive literature regarding differential risks of kidney diseases in certain populations, better reporting of the traditional demographic categorizations is still needed to ensure that the evidence generated in nephrology trials is at least representative and can be applied appropriately while avoiding extrapolating to those for whom it might not be applicable.
Our analysis also has several strengths. It is the first study to collate and compute reporting on and enrollment of minorities in nephrology clinical trials. Trials were analyzed over a 20-year period which provides insights into temporal trends and the effects of historical legislation on reporting and enrollment. Finally, we evaluated information on several important trial categories including disease state and geographic location of enrollment, providing a comprehensive view of representation of minorities in nephrology research.
Trial enrollment of Black and Hispanic patients reflecting the prevalence of kidney disease conditions in the population should help address gaps in the care of populations with kidney disease and help us to better understand outcomes disparities across racial and ethnic classifications. It is critical that RCT eligibility criteria, enrollment processes, and randomization are transparent with standardized reporting and equitable enrollment that reflects disease prevalence. To overcome biases in trial conduct, systematic change is needed through a concerted effort from investigators, policy makers, and funding agencies.
Supplementary Material
Footnotes
See related editorial, “Mirror, Mirror on the Wall: Do Kidney RCT Populations Represent the People We Treat?,” on pages 1127–1128.
Disclosures
D.M. Charytan reports the following—consultancy: Allena Pharmaceuticals (DSMB), Amgen, AstraZeneca, CSL Behring, Eli Lilly/Boehringer Ingelheim, Fresenius, Gilead, GSK, Janssen (steering committee), Medtronic, Merck, Nitto Biopharmaceuticals, Novo Nordisk, PLC medical (clinical events committee), Renalytix, and Zogenix; research funding: Amgen, Bioporto clinical trial support, Gilead, Medtronic clinical trial support, NovoNordisk; advisory or leadership role: CJASN and Expert Witness Fees Related to Proton Pump Inhibitors; and patents or royalties: royalties from UpToDate for authorship/edits on reviews. All remaining authors have nothing to disclose.
Funding
Q.H. Soomro is supported by ASN Ben J Lipps Fellowship grant and partly by UL1TR001445.
Author Contributions
Conceptualization: David M. Charytan, Qandeel H. Soomro.
Data curation: Amalya M. Charytan, David M. Charytan, Colin Keane, Angela McCarthy, Joey Nicholson, Giana Ramos, Qandeel H. Soomro, Dalila Varela, Javaughn Ways.
Formal analysis: Qandeel H. Soomro.
Methodology: David M. Charytan, Qandeel H. Soomro.
Software: Qandeel H. Soomro.
Supervision: David M. Charytan.
Visualization: Qandeel H. Soomro.
Writing – original draft: Qandeel H. Soomro.
Writing – review & editing: Amalya M. Charytan, David M. Charytan, Colin Keane, Angela McCarthy, Joey Nicholson, Giana Ramos, Qandeel H. Soomro, Dalila Varela, Javaughn Ways.
Data Sharing Information
The complete dataset used is available on request.
Supplemental Material
This article contains the following supplemental material online at http://links.lww.com/JSN/E417.
Supplemental Table 1. Overall Characteristics of worldwide studies, reporting of race and ethnicity.
Supplemental Table 2. Reporting of racial categories in worldwide trials according to publication year, trial category, and enrollment site.
Supplemental Table 3. Distribution of race and ethnicity in US trials. Data are presented according to as per published data for trials reporting information on race or ethnicity.
Supplemental Table 4. Distribution of race and ethnicity among enrolled participants in worldwide trials reporting information on race according to trial category.
Supplemental Table 5. Results of metaregression with the proportion of participants enrolled who were categorized as White in US trials as the dependent variable. N= 51 trials. Model AIC 1.14, BIC – 186.
Supplemental Table 6. Results of metaregression with the proportion of participants enrolled who were categorized as White in worldwide trials as the dependent variable. N= 194 trials (one trial reported that most of the participants were White individuals but did not provide quantitative data). Model AIC 0.92, BIC – 995.
Supplemental Table 7. Results of metaregression with the proportion of participants enrolled in US trials who were categorized as Hispanic as the dependent variable. N=27 trials. Model AIC 0.89, BIC -73.
Supplemental Table 8. Results of metaregression with the proportion of participants enrolled in worldwide trials who were categorized as Hispanic as the dependent variable. N=45 trials. Model AIC 0.80, BIC -148.
Supplemental Table 9. Pooled estimate from random effects meta-analysis. Effect sizes are based on proportions of each race and ethnicity category in US trials.
Supplemental Table 10. Pooled estimate from random effects meta-analysis. Effect sizes are based on proportions of each race and ethnicity category in worldwide trials.
Supplemental Table 11. Characteristics of all included trials (nr= not reported).
Supplemental Figure 1. Reporting of racial demographic information in worldwide trials according to trial category. Data are provided for reporting of White, Black, and Asian race.
Supplemental Figure 2. Participant demographics in worldwide according to trial category for trials reporting race of enrolled participants.
Supplemental Figure 3. Distribution of race and ethnicity by publication year in worldwide trials for trials reporting information on race or ethnicity.
References
- 1.GBD Chronic Kidney Disease Collaboration. Global, regional, and national burden of chronic kidney disease, 1990-2017: a systematic analysis for the Global Burden of Disease Study 2017. Lancet. 2020;395(10225):709-733. doi: 10.1016/S0140-6736(20)30045-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.United States Renal Data System. 2021 USRDS Annual Data Report: Epidemiology of Kidney Disease in the United States. National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases; 2021. [Google Scholar]
- 3. Centers for Disease Control and Prevention. Kidney disease mortality by state. Accessed August 10, 2022. https://www.cdc.gov/nchs/pressroom/sosmap/kidney_disease_mortality/kidney_disease.htm.
- 4.Yan G Shen JI Harford R, et al. Racial and ethnic variations in mortality rates for patients undergoing maintenance dialysis treated in US territories compared with the US 50 states. Clin J Am Soc Nephrol. 2020;15(1):101–108. doi: 10.2215/CJN.03920319 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Benjamins MR Lorenz P Saiyed NS, et al. Black-white inequities in kidney disease mortality across the 30 most populous US cities. J Gen Intern Med. 2022;37(6):1351–1358. doi: 10.1007/s11606-022-07444-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.King KL, Husain SA, Jin Z, Brennan C, Mohan S. Trends in disparities in preemptive kidney transplantation in the United States. Clin J Am Soc Nephrol. 2019;14(10):1500–1511. doi: 10.2215/CJN.03140319 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Powe NR. The pathogenesis of race and ethnic disparities: targets for achieving health equity. Clin J Am Soc Nephrol. 2021;16(5):806–808. doi: 10.2215/CJN.12640820 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Boulware LE, Mohottige D. The seen and the unseen: race and social inequities affecting kidney care. Clin J Am Soc Nephrol. 2021;16(5):815–817. doi: 10.2215/CJN.12630820 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Ramamoorthy A, Kim HH, Shah-Williams E, Zhang L. Racial and ethnic differences in drug disposition and response: review of new molecular entities approved between 2014 and 2019. J Clin Pharmacol. 2022;62(4):486–493. doi: 10.1002/jcph.1978 [DOI] [PubMed] [Google Scholar]
- 10.Norris KC, Agodoa LY. Unraveling the racial disparities associated with kidney disease. Kidney Int. 2005;68(3):914–924. doi: 10.1111/j.1523-1755.2005.00485.x [DOI] [PubMed] [Google Scholar]
- 11.Nicholas SB, Kalantar-Zadeh K, Norris KC. Racial disparities in kidney disease outcomes. Semin Nephrol. 2013;33(5):409–415. doi: 10.1016/j.semnephrol.2013.07.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Panico C, Thompson A. African Americans, kidney disease, and drug development: a regulatory perspective. Am J Kidney Dis. 2018;72(5):S33–S36. doi: 10.1053/j.ajkd.2018.06.023 [DOI] [PubMed] [Google Scholar]
- 13.Umeukeje EM, Young BA. Genetics and ESKD disparities in African Americans. Am J Kidney Dis. 2019;74(6):811–821. doi: 10.1053/j.ajkd.2019.06.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Zhang T, Tsang W, Wijeysundera HC, Ko DT. Reporting and representation of ethnic minorities in cardiovascular trials: a systematic review. Am Heart J. 2013;166(1):52–57. doi: 10.1016/j.ahj.2013.03.022 [DOI] [PubMed] [Google Scholar]
- 15.Turner BE, Steinberg JR, Weeks BT, Rodriguez F, Cullen MR. Race/ethnicity reporting and representation in US clinical trials: a cohort study. Lancet Reg Health Am. 2022;11:100252. doi: 10.1016/j.lana.2022.100252 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Food and Drug Administration. Collection of race and ethnicity data in clinical trials: Guidance for industry and food and drug administration staff, October 26, 2016. Accessed August 10, 2022. www.fda.gov/media/75453/download.
- 17.Page MJ McKenzie JE Bossuyt PM, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021;372:n71. doi: 10.1136/bmj.n71 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Nyaga VN, Arbyn M, Aerts M. Metaprop: a Stata command to perform meta-analysis of binomial data. Arch Public Health. 2014;72(1):39. doi: 10.1186/2049-3258-72-39 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Barker TH Migliavaca CB Stein C, et al. Conducting proportional meta-analysis in different types of systematic reviews: a guide for synthesisers of evidence. BMC Med Res Methodol. 2021;21(1):189. doi: 10.1186/s12874-021-01381-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Kibria GMA, Crispen R. Prevalence and trends of chronic kidney disease and its risk factors among US adults: an analysis of NHANES 2003-18. Prev Med Rep. 2020;20:101193. doi: 10.1016/j.pmedr.2020.101193 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Organ Procurement & Transplantation Network. National data. Accessed August 24, 2022. https://optn.transplant.hrsa.gov/data/view-data-reports/national-data/
- 22. United States Census Bureau. QuickFacts. Accessed August 30, 2022. https://www.census.gov/quickfacts/fact/table/US/PST045221.
- 23.Rizzolo K, Cervantes L, Shen JI. Racial and ethnic disparities in home dialysis use in the United States: barriers and solutions. J Am Soc Nephrol. 2022;33(7):1258–1261. doi: 10.1681/ASN.2022030288 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Sanossian N Rosenberg L Liebeskind DS, et al. A dedicated Spanish language line increases enrollment of Hispanics into prehospital clinical research. Stroke. 2017;48(5):1389–1391. doi: 10.1161/strokeaha.117.014745 [DOI] [PubMed] [Google Scholar]
- 25.Minocher Homji RS, Lakhoo S, Ray JG. Recruitment of immigrant and ethnic minorities in primary prevention trials of cardiovascular disease. QJM. 2011;104(6):469–476. doi: 10.1093/qjmed/hcr027 [DOI] [PubMed] [Google Scholar]
- 26.Ma GX, Seals B, Tan Y, Wang SY, Lee R, Fang CY. Increasing Asian American participation in clinical trials by addressing community concerns. Clin Trials. 2014;11(3):328–335. doi: 10.1177/1740774514522561 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Occa A, Leip A, Merritt AS, Stapleton JL. Prevalence and correlates of invitation to participate in clinical trials among US adults. Prev Med Rep. 2022;26:101742. doi: 10.1016/j.pmedr.2022.101742 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Loree JM Anand S Dasari A, et al. Disparity of race reporting and representation in clinical trials leading to cancer drug approvals from 2008 to 2018. JAMA Oncol. 2019;5(10):e191870. doi: 10.1001/jamaoncol.2019.1870 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Acton EK, Abbasi MH, Kasner SE. Evaluating age, sex, racial, and ethnic representation in acute ischemic stroke trials, 2010 to 2020: a systematic review and meta-analysis. J Am Heart Assoc. 2022;11(8):e024651. doi: 10.1161/jaha.121.024651 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Heller C Balls-Berry JE Nery JD, et al. Strategies addressing barriers to clinical trial enrollment of underrepresented populations: a systematic review. Contemp Clin Trials. 2014;39(2):169–182. doi: 10.1016/j.cct.2014.08.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Dember LM. The potential for pragmatic trials to reduce racial and ethnic disparities in kidney disease. J Am Soc Nephrol. 2022;33(9):1649–1651. doi: 10.1681/ASN.2022030301 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Thakur N Lovinsky-Desir S Appell D, et al. Enhancing recruitment and retention of minority populations for clinical research in pulmonary, critical care, and sleep medicine: an official American Thoracic Society research statement. Am J Respir Crit Care Med. 2021;204(3):e26–e50. doi: 10.1164/rccm.202105-1210st [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Butler III J, Fryer CS, Garza MA, Quinn SC, Thomas SB. Commentary: critical race theory training to eliminate racial and ethnic health disparities: the public health critical race praxis Institute. Ethn Dis. 2018;28(suppl 1):279–284. doi: 10.18865/ed.28.s1.279 [DOI] [PMC free article] [PubMed] [Google Scholar]