A Comprehensive Network Analysis of Biopsychosocial Factors Associated with Postpartum Depression

Background. The current study investigated relations between a broad set of postpartum depression (PPD) risk- and protective factors, their centrality and unique relations with PPD symptoms. Methods. Mixed graphical network models were estimated in cross-sectional data collected during Phase 7 (2012-2015) of the Pregnancy Risk Assessment Monitoring System. Half of the 57,518 women were included in exploratory model 1, the other half in model 2 for cross-validation. A broad selection of biopsychosocial factors were modelled, including sociodemographic variables, indices of maternal health (behavior), pregnancy course, support, infant variables, and stressors. Results. A densely connected network of risk- and protective factors was obtained. Pregnancy duration, infant intensive care unit placement, infertility treatment, birth weight, income, and childbirth classes were ranked amongst the most central variables in the model. Out of 35 biopsychosocial factors included in the model, 11 were robustly related with PPD across both samples. High income, pre-pregnancy physical activity, pregnancy intention, and prenatal care focusing on depression risk were related to lower depression severity. Several other variables, including prior history of depression, pre-pregnancy dieting, prenatal risk behavior, and personal stressors were uniquely related to increased depression severity. Depressive complaints reduced with increased time since delivery. Women experiencing depressive symptoms were more likely to rely on aid from health workers postpartum. Results from models 1 and 2 were largely identical. Conclusions. Although cross-sectional in nature, these findings shed light on the complex associations between key risk- and protective factors for PPD, with implications for early detection and prevention.