Objectives: To characterize race and ethnicity reporting and representation in United States (US) gynecologic oncology (Gyn-Onc) clinical trials and compare reporting and representation across gynecology subspecialties. Methods: In this cross-sectional study, we investigated all US-based gynecology clinical trials that reported results on ClinicalTrials.gov from October 2007-March 2020, with a focus on Gyn-Onc. We evaluated two outcomes: (1) reporting of participant race/ethnicity and (2) representation (proportion of trial enrollees of a given race/eth- nicity). We analyzed associations of trial funding and subspecialty with each outcome using descriptive and multivariable logistic regression analyses. We modeled representation using logistic regression with a binary outcome (> or <20% Black, Indigenous, and People of Color [BIPOC] enrollment) controlling for a trial year, phase, and funder. Results: Of 8,515 registered gynecology trials, 3,214 were based in the US. Only 888 of 3,214 (27.6%) reported trial results, of which 505 (56.9%; 190,147 participants) included participant race/ethnicity data. There was a significant increase in race/ethnicity reporting over the study period (p<0.001). Race/ethnicity reporting varied by funder (government/academic 14.2%, industry 19.4%, p<0.001) and subspecialty (Gyn-Onc 13.9%, Benign Gynecology 16.6%, p=0.002), but these reporting differences were not persistent in multivariable analysis. Among trials that reported race/ethnicity, all BIPOC groups were underrepresented relative to their national population representation (Table). No significant difference in representation was seen by funding status. However, differences by subspecialty were noted; in aggregate, Benign Gynecology trials had more than twice as much diversity as Gyn-Onc trials (39.6% vs 17.7% average BIPOC representation per trial). Compared with Gyn-Onc clinical trials, REI (aOR: 2.78, 95% CI: 1.36-6.47), family planning (aOR: 3.68, 95% CI 1.87-8.16), and other benign gynecology trials (aOR: 1.79, 95% CI: 1.26-2.55) all had greater odds of enrolling at least 20% BIPOC participants. No difference was demonstrated between Gyn-Onc and Urogynecology trials. Conclusions: Reporting of race/ethnicity in US gynecology clinical trials is poor but improving. Among trials with race-ethnicity reporting, BIPOC participants were consistently underrepresented. Gyn-Onc trials demonstrated the least diversity of all gynecology subspecialties. This trend excludes BIPOC communities from health innovation conferred through trial participation, biases medical evidence, and worsens disparities in gynecology. Objectives: To characterize race and ethnicity reporting and representation in United States (US) gynecologic oncology (Gyn-Onc) clinical trials and compare reporting and representation across gynecology subspecialties. Methods: In this cross-sectional study, we investigated all US-based gynecology clinical trials that reported results on ClinicalTrials.gov from October 2007-March 2020, with a focus on Gyn-Onc. We evaluated two outcomes: (1) reporting of participant race/ethnicity and (2) representation (proportion of trial enrollees of a given race/eth- nicity). We analyzed associations of trial funding and subspecialty with each outcome using descriptive and multivariable logistic regression analyses. We modeled representation using logistic regression with a binary outcome (> or <20% Black, Indigenous, and People of Color [BIPOC] enrollment) controlling for a trial year, phase, and funder. Results: Of 8,515 registered gynecology trials, 3,214 were based in the US. Only 888 of 3,214 (27.6%) reported trial results, of which 505 (56.9%; 190,147 participants) included participant race/ethnicity data. There was a significant increase in race/ethnicity reporting over the study period (p<0.001). Race/ethnicity reporting varied by funder (government/academic 14.2%, industry 19.4%, p<0.001) and subspecialty (Gyn-Onc 13.9%, Benign Gynecology 16.6%, p=0.002), but these reporting differences were not persistent in multivariable analysis. Among trials that reported race/ethnicity, all BIPOC groups were underrepresented relative to their national population representation (Table). No significant difference in representation was seen by funding status. However, differences by subspecialty were noted; in aggregate, Benign Gynecology trials had more than twice as much diversity as Gyn-Onc trials (39.6% vs 17.7% average BIPOC representation per trial). Compared with Gyn-Onc clinical trials, REI (aOR: 2.78, 95% CI: 1.36-6.47), family planning (aOR: 3.68, 95% CI 1.87-8.16), and other benign gynecology trials (aOR: 1.79, 95% CI: 1.26-2.55) all had greater odds of enrolling at least 20% BIPOC participants. No difference was demonstrated between Gyn-Onc and Urogynecology trials. Conclusions: Reporting of race/ethnicity in US gynecology clinical trials is poor but improving. Among trials with race-ethnicity reporting, BIPOC participants were consistently underrepresented. Gyn-Onc trials demonstrated the least diversity of all gynecology subspecialties. This trend excludes BIPOC communities from health innovation conferred through trial participation, biases medical evidence, and worsens disparities in gynecology.
Read full abstract