Sedentary behavior, physical activity, psychological well-being, and gynecologic and obstetric disorders: A multivariable Mendelian randomization and mediation analysis.

The independent causal effects of interrelated lifestyle behaviors and psychological states on women's gynecologic and obstetric health remain incompletely understood. We conducted a comprehensive Mendelian randomization (MR) study to disentangle the effects of leisure screen time, moderate-to-vigorous physical activity (MVPA), experiencing mood swings, and the well-being spectrum on a wide range of these disorders. We performed 2-sample MR analyses using data from the UK Biobank and FinnGen cohorts for 20 outcomes, with causal estimates combined via meta-analysis. The analytical framework included univariable MR, multivariable Mendelian randomization (MVMR) to assess independent effects, and an exploratory mediation analysis. A comprehensive suite of sensitivity analyses was used to assess the robustness of the findings against pleiotropy and heterogeneity, and the results were corrected for multiple testing using the false discovery rate. Our analyses identified a core set of 6 disorders, including polycystic ovaries, endometriosis, and menstrual disorders, that were robustly associated with all 4 exposures. Overall, genetically predicted leisure screen time and experiencing mood swings were causally associated with an increased risk for 10 and 12 disorders. Conversely, MVPA and well-being spectrum were associated with a decreased risk for 9 and 12 disorders. These associations were largely independent in MVMR analyses. Exploratory mediation analysis highlighted that inflammation (via C-reactive protein) and hormonal pathways (via estradiol) mediated numerous associations, such as the C-reactive protein-mediated protective effect of MVPA on female infertility. This study provides genetically strengthened evidence for independent causal relationships between movement behaviors, psychological states, and a wide spectrum of gynecologic and obstetric disorders. Methodologically, our MVMR approach was instrumental in disentangling the effects of correlated exposures. Mechanistically, our findings point toward inflammation and hormonal pathways as potential mediators, which may inform future prevention and therapeutic strategies for women's health.