The landscape and outcome reporting in randomized controlled trials within women’s health field: a meta-epidemiological study. 4. evaluating the statistical significance and publication bias in obstetrics and gynaecology studies.

Randomized controlled trials (RCTs) aim to identify medical interventions that are safe and effective for patients. To achieve this goal, trialists should select, collect and report outcomes that are able to capture the harms and benefits of the intervention, should be relevant to patients with the affected disorder and should be consistent with the clinical practice. In reproductive medicine, live birth has been long considered the gold standard for measuring the effectiveness of infertility interventions. However, it takes longer to observe, is subject to random error and has a low statistical power to pick up an effect. Because of these limitations, clinical pregnancy or ongoing pregnancy was used more often. Despite that reporting live births in infertility trials has been a long-running recommendation, the extent to which this is adhered to is unclear. Outcomes in RCTs are not always consistently defined. For example, while one study considers 10 weeks of gestation as clinical pregnancy, another study might consider it as ongoing pregnancy. The absent or inconsistent outcome definitions in trials complicate the interpretation of the trial results and undermine the reliability of evidence synthesis. Our group recently published a meta-epidemiological study by investigating the abstract in reproductive medicine studies (Feng, Q., Mol, B. W., Ioannidis, J. P. A., & Li, W. 2023; Statistical significance and publication reporting bias in abstracts of reproductive medicine studies. Human Reproduction). We found that reporting of P-values alone prevailed and did not abate over time, while effect measures increased steadily, especially in RCTs and meta-analyses. Built on these findings, we will delve deeper in 1) mapping the landscape of RCTs in women’s health, including in reproductive medicine, gynaecology and obstetrics published in the last 10 years; 2) exploring the heterogeneity in outcome definitions and the quality of outcome reporting in the RCTs within women’s health field; 3) assessing the association in treatment effects between early outcomes such as clinical pregnancy and late outcome such as live birth; 4) evaluating the statistical significance and publication bias in obstetrics and gynaecology studies.