Purpose
Retinal vein occlusion (RVO) is the second most common cause of vision loss from retinal vascular disease. Clinical trials often lack adequate representation of minority groups, limiting the external validity and equity of trial results. We aimed to characterize the evolution of ethnoracial demographic reporting, changes in the racial and ethnic composition of RVO trial participants, and correlations with trial sponsor and collaborator types.
Design
Retrospective trend study.
Subjects
Participants, sponsors, and collaborators in RVO clinical trials based in the United States (U.S.).
Methods
We assessed the ethnoracial distribution of patients and sponsors and collaborators in RVO clinical trials. Trial data was sourced from ClinicalTrials.gov and compared with 2022 U.S. Census Bureau data.
Main Outcome Measures
Change in ethnoracial reporting and representation, and comparison of representation within RVO clinical trials with 2022 U.S. Census demographics.
Results
Racial data was not documented for all 6 (12.2%) trials completed between 2008 and 2011. There was increased ethnoracial reporting between January 2012 and June 2017 ( n = 6/29, 20.7%) vs July 2017 to December 2022 ( n = 13/14, 92.9%) (OR 49.8; 95% CI: 5.39-460.47; P < .0001). The proportion of Asian participants increased from 4.2% to 15.9% (OR 4.37; 95% CI 2.49-8.37; P < .0001), and Black participants rose from 6.1% to 20.7% (OR 4.03; 95% CI 2.51-6.85; P < .0001). Conversely, White participants decreased from 73.8% to 56.0% (OR 0.45, 95% CI 0.34-0.59; P < .0001), and Hispanic/Latino participants decreased from 13.1% to 5.4% (OR 0.38; 95% CI 0.26-0.56; P < .0001). Compared to the 2022 U.S. Census data, the ethnoracial distribution of the total U.S. population is significantly different than the total enrollment in RVO clinical trials ( P < .0001). Government and academic involvement were correlated with less White and more Black representation respectively ( P ≤ .01).
Conclusions
From 2008 to 2022, there has been a notable increase in the documentation of racial and ethnic demographics in RVO clinical trials. Although the recruitment of Asian and Black patients has increased, the Hispanic population remains significantly underrepresented among RVO trial participants, highlighting the need for greater inclusivity in enrollment.
INTRODUCTION
Retinal vein occlusion (RVO) is the second most common cause of vision loss from retinal vascular disease after diabetic retinopathy (DR), affecting approximately 28 million people worldwide. RVO is an obstruction of the retinal venous system and is categorized into branch RVO, hemiretinal occlusion, and central RVO based on the occlusion site. Established risk factors include older age, hypertension, diabetes mellitus, and prior vascular diseases, with greater severity of comorbid disease correlated with heightened risk of developing RVO. , These disorders are well-documented to exhibit disparities between racial and ethnic groups. Diabetes disproportionately impacts Hispanic, Black, and Asian patients, and higher hypertension prevalence and lower blood pressure control rates have been documented among historically marginalized groups compared with non-Hispanic White populations. Similarly, the incidence of RVO is as least as prevalent, if not more so, in individuals from underrepresented populations. In a pooled analysis of 15 population-based studies, the estimated prevalence of any RVO was 3.7 per 1000 in Whites, 3.9 in Blacks, 5.7 in Asians, and 6.9 in Hispanics. Furthermore, RVO’s degree of visual impairment and resultant treatment burden have been observed to differ between racial and ethnic groups.
Clinical trials should strive to represent real-world populations to improve the generalizability of findings and support evidence-based care. Ensuring representative sampling of individuals facing the greatest disease burdens and severe manifestations is crucial, particularly when developing cutting-edge therapeutics that could significantly influence the trajectory of their illnesses. Recognizing this need, the 1993 NIH Revitalization Act was enacted to promote greater inclusion of racial minorities in NIH-funded studies. Subsequent legislation, such as Title VIII of the 2007 Food and Drug Administration (FDA) Amendments Act and the 2016 Final Rule, further strengthened requirements for reporting race and ethnicity as baseline characteristics in clinical trials. , Despite these efforts, individuals from racially and ethnically diverse backgrounds remain underrepresented in clinical research across many medical specialties. Addressing these disparities in ophthalmology requires greater transparency about the demographics of clinical trial participants.
Previous research has examined the distribution of race and ethnicity within RVO clinical trials, characterizing underrepresentation and overrepresentation within individual studies. To our knowledge, there is a lack of literature examining temporal trends in race and ethnicity documentation, the recruitment of historically underrepresented patient populations over time, and their correlation with clinical trial sponsors and collaborators. In this study, we aimed to characterize the trajectory of race reporting and ethnoracial composition of participants in RVO clinical trials while exploring associations with different categories of collaborators and sponsors. We compared our findings with similar studies that assess clinical trials for other ophthalmic diseases, including age-related macular degeneration (AMD) and diabetic macular edema.
METHODS
We conducted a retrospective cross-sectional study, modeled after Yu et al’s work, to assess the racial and ethnic representation in completed RVO clinical trials. Between December 2022 and January 2023, we reviewed the National Library of Medicine ClinicalTrials.gov database for trials completed before December 31, 2022. Trials that were ongoing, conducted outside of the United States (U.S.), or did not provide demographic information were excluded from analysis. Study completion year, trial location, number of enrolled subjects, and the presence or absence of ethnoracial data were recorded. When documented, the number of American Indian/Alaska Native, Asian, Native Hawaiian/Pacific Islander, Black/African American, White, Hispanic/Latino, and mixed-race participants were extracted. Sponsor and collaborator information were categorized into government, industry, private/group practice, academic, or nonprofit. Trials could be associated with multiple categories of sponsors or collaborators depending on the organizations or entities participating.
Descriptive statistics and odds ratios were used to identify temporal trends in reporting race and ethnicity information. To examine shifts in ethnoracial representation over time, we compared the proportion of participants in each racial and ethnic category between the first and second halves of the timeframe with available demographic data, with trials categorized based on year of completion. Chi-square tests were used to compare the demographics of participants in RVO clinical trials with publicly available data from the 2022 U.S. Census. Fischer exact tests were used to compare sponsor or collaborator categories between time periods, and t tests were used to evaluate differences in demographic proportions based on sponsor or collaborator types. Data analysis was performed through StataSE 17 (StataCorp). All statistical tests were 2-sided and used a significance threshold of P < .05. The study was exempt from institutional board review and adhered to the tenets of the Declaration of Helsinki.
RESULTS
The first RVO trial listed on ClinicalTrials.gov was completed in 2008; since then, a total of 49 trials have been conducted. Among these, 19 (38.8%) included at least one race as a baseline characteristic, whereas 30 (61.2%) did not report racial demographics ( Figure 1 ). Racial data was not documented for all 6 (12.2%) trials completed from 2008 to 2011. Between January 2012 and June 2017, 20.7% of trials (6/29) included race reporting; this significantly increased to 92.9% (13/14) between July 2017 and December 2022 (OR 49.8; 95% CI: 5.39-460.47; P < .0001).
A total of 3635 participants were enrolled in the 19 trials that reported racial demographics. White participants comprised the largest group ( n = 2092; 57.6%), followed by Black ( n = 706; 19.4%) and Asian ( n = 542; 14.9%) participants. Smaller proportions of participants identified as Hispanic/Latino ( n = 222; 6.1%), American Indian/Alaska Native ( n = 16; 0.4%), Native Hawaiian/Other Pacific Islander ( n = 9; 0.2%), and mixed-race ( n = 37; 1.0%) ( Figure 2 ). Over the years, the proportion of Asian patients increased from 4.2% to 15.9% (OR 4.37; 95% CI 2.49-8.37; P < .0001), and Black participants rose from 6.1% to 20.7% (OR 4.03; 95% CI 2.51-6.85; P < .0001) ( Table ; Figures 2 , 3 ). Conversely, White participants decreased from 73.8% to 56.0% (OR 0.45, 95% CI 0.34-0.59; P < .0001), and Hispanic/Latino patients decreased from 13.1% to 5.4% (OR 0.38; 95% CI 0.26-0.56; P < .0001). Mixed-race participants additionally decreased from 5.8% to 0.6% (OR 0.09; 95% CI 0.05-0.19; P < .0001). The distribution of American Indian/Alaska Native (OR 0.70; 95% CI 0.16-6.38; P = .65) and Native Hawaiian/Other Pacific Islander (OR 0.75; 95% CI 0.10-33.52; P = .56) trial participants remained relatively stable.
| Odds Ratio (95% CI) | P Value | |
|---|---|---|
| Hispanic/Latino | 0.38 (0.26-0.56) | <.0001 a |
| White | 0.45 (0.34-0.59) | <.0001 a |
| Black | 4.03 (2.51-6.85) | <.0001 a |
| Asian | 4.37 (2.49-8.37) | <.0001 a |
| American Indian/Alaska Native | 0.70 (0.16-6.38) | .65 |
| Native Hawaiian/Other Pacific Islander | 0.75 (0.10-33.52) | .56 |
| Mixed Race | 0.09 (0.05-0.19) | <.0001 a |
a Significant difference in proportion between January 1, 2012, to June 30, 2017, and July 1, 2017, to December 31, 2022 (all P < .0001).
According to the 2022 U.S. Census data, the U.S. population is 75.5% White, 19.1% Hispanic/Latino, 13.6% Black, 6.3% Asian, 1.3% American Indian/Alaska Native, 0.3% Native Hawaiian/Other Pacific Islander, and 3.0% mixed-race. The ethnoracial distribution of the total U.S. population is significantly different than the total patient enrollment in RVO clinical trials ( P < .0001), as well as compared with patient enrollment within the first ( P < .0001) and second ( P < .0001) halves of the evaluated periods.
Among the 49 clinical trials analyzed, industry sponsors and collaborators were the most prevalent, supporting 37 trials (75.5%). Academic institutions and private/group practices were each involved in 12/49 trials (24.5%), while government agencies (6/49, 12.2%) and nonprofit organizations (4/49, 8.2%) had the least involvement ( Figure 4 ). No statistically significant associations were found between sponsor or collaborator type and the presence or absence of demographic documentation (government: P = .16; industry: P = .43; private/group practice: P = .24; academic: P = .24; nonprofit: P = .17). Over time, private/group practice involvement decreased, starting from 2/6 trials (33.3%) between January 2008 and December 2011 and 10/29 trials (34.5%) from January 2012 to June 2017 and dropping to 0/14 trials after July 2017 ( P < .05). No significant trends were observed in government ( P = .35), industry ( P = .50), academic ( P = .70), or nonprofit ( P = .42) sponsorship or collaboration. However, industry sponsorship increased from 3/6 (50%) studies between January 2012 and June 2017 to 11/13 (84.6%) between July 2017 and December 2022.
