Demographic and Obstetric Factors Affecting Mental Health of Pregnant Women During COVID-19: EPDS Assessment Study

SRI JAHNAVI ADUSUMILLI, Saptarshi Purkayastha, Shruti Chaitanya, Nigama Pervala, Jamesetta Quiqui

Confirmation: I have read and agree with the IEEE BHI 2025 conference submission's policy on behalf of myself and my co-authors.

Keywords: Pregnancy, COVID-19, EPDS, machine learning, maternal mental health, XGBoost, Random Forest

Abstract: This study investigates the relationship between demographic and obstetric factors and maternal mental health during the COVID-19 pandemic. Using data from Canadian pregnant women (April 2020–April 2021) and advanced machine learning models, we found significant associations between sociodemographic characteristics and prenatal mental health, as measured by EPDS scores. Maternal education, household income, perceived threat levels, and anxiety scores emerged as key predictors of depression risk. XGBoost (89.46% accuracy) and Random Forest (AUC = 0.99 for severe depression) demonstrated strong predictive performance, offering promising tools for early intervention. However, the study’s cross-sectional design, geographically limited sample, and selected variables constrain generalizability and causal inference. Despite these limitations, the findings underscore the importance of integrating predictive models into prenatal care and highlight the urgent need for targeted mental health screenings during public health crises.

Track: 5. Public Health Informatics

Registration Id: D5NTT7XVTC2

Submission Number: 290