Association between epidural labor analgesia and postpartum depressive symptoms at 6 weeks in primiparous women: a prospective preference-based cohort study

This study evaluated the association between epidural labor analgesia (ELA) and a positive Edinburgh Postnatal Depression Scale (EPDS) screen at 6 weeks postpartum among singleton primiparous women planning vaginal delivery and explored a parsimonious prediction model. This single-center prospective preference-based cohort enrolled 200 women with spontaneous onset of labor between July and December 2021. Participants chose whether to receive ELA. The primary outcome was a positive EPDS screen (score ≥ 9) at postpartum day 42. The primary logistic regression model included ELA, educational attainment, and antenatal fear of labor pain; post-exposure variables were excluded. Sensitivity analyses used inverse probability of treatment weighting (IPTW) based on baseline characteristics. The fixed three-variable model was also presented as an exploratory nomogram and internally validated using 1,000 bootstrap resamples. The final analysis included 184 women (ELA, n = 93; no-ELA, n = 91). Positive EPDS screens occurred in 13/93 women (14.0%) in the ELA group and 25/91 (27.5%) in the no-ELA group (Yates-corrected P = 0.038), with an absolute risk difference of − 13.49% points (95% CI − 24.92 to − 1.75). The adjusted odds ratio for ELA was 0.49 (95% CI 0.23–1.07; P = 0.075). IPTW estimates were directionally similar, but their confidence intervals included the null. The exploratory nomogram had an apparent area under the receiver operating characteristic curve of 0.720 and an optimism-corrected C-statistic of 0.708. ELA recipients had a lower unadjusted rate of positive EPDS screening, but adjusted and weighted confidence intervals included the null. These findings do not establish a causal protective effect and require cautious interpretation given the preference-based design, potential residual confounding, and limited sample size. The exploratory prediction model requires external validation before clinical use.