Am J Obstet Gynecol MFM. 2021 Jul 26. pii: S2589-9333(21)00145-2. [Epub ahead of print] 100450
BACKGROUND: Randomized controlled trials are considered the highest level of evidence but fewer than half are reproducible. Rigorous methodology improves trial quality but reproducibility may be limited by lack of transparency in reporting. The Consolidated Standard of Reporting Trials (CONSORT) guidelines define reporting standards, and pre-trial registration requires pre-defined methodology and outcomes.
OBJECTIVE: We evaluated obstetrics and gynecology trials in six journals for adherence to CONSORT guidelines. Secondarily, we evaluated pre-trial registration compliance and concordance between registry and publication. Further, we evaluated differences in trial characteristics of randomized controlled trials with highest versus lower compliance, and adherence to guidelines by journal type.
STUDY DESIGN: This was a cross-sectional study of obstetrics and gynecology trials published between 2017 and 2019 in six journals (AJOG, BJOG, Obstetrics and Gynecology, JAMA, The Lancet, NEJM). Randomized controlled trials were identified via PubMed and manual journal archive search. The primary outcome was adequate compliance with CONSORT guidelines defined as ≥ 80% of checklist items present. Secondary outcomes included completion of pre-trial registration, and concordance between pre-trial registration and publication for outcomes and sample size. We compared characteristics of trials with adequate versus inadequate compliance. Secondary analyses included comparisons of characteristics of trials in the top quartile of CONSORT compliance with those in lower quartiles, and compliance with guidelines in OBGyn versus non-OBGyn journals. In exploratory analysis, trends in CONSORT compliance across the study period were assessed. A post-hoc sensitivity analysis evaluated outcomes with exclusion of two retracted trials.
RESULTS: Of 170 trials included, 80% (95% CI 74-86%) were adequately compliant with CONSORT manuscript guidelines and 66% (95% CI 59-73%) with abstract guidelines. Nearly all (98%) trials reported pre-trial registration. Concordance between pre-trial registration and publication was identified for 77% of primary outcomes, 32% of secondary outcomes, and 60% of sample sizes. Trials with adequate compliance were more likely to be pre-registered, include an a priori power calculation, and use intent to treat analysis. Trials in the top quartile of CONSORT compliance were more likely to be multicenter, international, and government funded. More trials were in the top quartile of CONSORT compliance from non-OBGyn compared to OBGyn journals (64.9% vs 25.7%, p<0.001). No significant trends in adequate compliance were identified. Results did not differ significantly in sensitivity analysis.
CONCLUSIONS: Twenty percent of OBGyn trials, in 6 high-impact journals, were not compliant with CONSORT guidelines, and there were major discrepancies between pre-trial registration and publication. Transparency, reproducibility, and scientific rigor in obstetrics and gynecology trial reporting needs to be improved.
Keywords: accountability; methodology; obstetrics & gynecology trials; reporting guidelines; reproducibility crisis; research design