Back

Advanced Causal Inference Methods in Obstetrics and Gynecology: A Simulation Study on Preeclampsia Prevention

Shi, W.

2025-10-09 obstetrics and gynecology
10.1101/2025.10.08.25337637 medRxiv
Show abstract

AO_SCPLOWBSTRACTC_SCPLOWPreeclampsia remains a leading cause of maternal and perinatal morbidity and mortality world-wide. Early preventive interventions, including low-dose aspirin therapy, reduce risk in high-risk pregnancies. Observational studies often face confounding, nonlinear covariate interactions, and heterogeneous treatment effects. We conducted a comprehensive simulation study comparing naive logistic regression (NL), propensity score weighting (PSW), regression adjustment (RA), doubly robust estimation (DR), targeted maximum likelihood estimation (TMLE), and causal forests (CF) in estimating aspirin effects on preeclampsia outcomes. Simulated cohorts reflected diverse confounding structures and treatment effect heterogeneity. Performance metrics included bias, standard deviation (SD), root mean squared error (RMSE), and coverage probability. TMLE and CF consistently yielded unbiased estimates and identified subgroup-specific effects, highlighting their potential in observational OB/GYN research.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.