Early Discontinuation, Results Reporting, and Publication of Gynecology Clinical Trials From 2007 to 2020

The aim of this study was to characterize gynecology clinical trials over time, compare trials in various subspecialties, and analyze factors associated with early discontinuation, failure, or inadequacy to report results, as well as to publish and disseminate findings of gynecology clinical trials. The authors performed cross-sectional analysis of all gynecology trials registered on ClinicalTrials.gov between 2007 and 2020 and their resulting publications. Trials were analyzed using descriptive, Cox regression, and multivariable logistic analyses. The primary exposure variables were trial funding and gynecology subspecialty. The association of these variables with the 3 following outcomes was analyzed: early discontinuation, reporting results within 3 years of completion, and publication in a peer-reviewed journal indexed on PubMed. Only 3.7% of clinical trials (8174 of the 223,690) registered within the study time frame focused on gynecology, comprising benign gynecology (35.8%), gynecologic oncology (25.2%), reproductive endocrinology and infertility (17.5%), urogynecology (13.7), and family planning (7.9) trials. Fewer than half of completed trials (42.0%) disseminated results through reporting and publication of results. Of the 3 funding types (industry, government, and academic), industry-funded trials had the greatest likelihood of early discontinuation ( P < 0.001). The least likely studies to report results in ClinicalTrials. gov were academic-funded trials (adjusted odds ratio [aOR], 0.38; 95% confidence interval [CI], 0.30-0.50), but they were also the most likely to publish results (aOR, 1.62; 95% CI, 1.24-2.12). Between 2007 and 2020, the number of reproductive endocrinology and infertility trials increased the most of any subspecialty (6.4% compound annual growth rate). Compared with other subspecialties, reproductive endocrinology and infertility and family planning trials were themost likely to be to be discontinued early (respectively, adjusted hazards ratio, 2.08; 95% CI, 1.59-2.71 and adjusted hazards ratio, 1.55; 95% CI, 1.06-2.25). Completed reproductive endocrinology and infertility trials had the lowest likelihood of reporting results on ClinicalTrials.gov of any gynecologic subspecialty; the aOR was 0.58, with a 95% CI of 0.38 to 0.88. There were no significant differences in publication rates of different subspecialties. These findings raise concern of bias in the performance, reporting, publication, and dissemination of gynecology clinical trials. Only 3.7% of all clinical trials are gynecology trials. The paucity of gynecology clinical trials is consistent with decades of female underrepresentation in clinical research. The authors suggest that identifying and systematically addressing barriers to dissemination of gynecology clinical trials could decrease bias.