Abstract
BACKGROUND: Although previous studies have established a link between physical intimate partner violence (IPV) and adverse health outcomes for mothers and children, there is a lack of thorough comparative analysis in low-income and middle-income countries (LMICs) that examines how physical IPV experienced during pregnancy specifically differs from physical IPV at other times. This comparison is crucial to understanding the extensive impact of physical IPV during pregnancy on antenatal care, early initiation of breastfeeding (EIBF), exclusive breastfeeding for the first two days after birth (EBF2D), and place of delivery (POD). METHODS: This study conducted secondary analyses using cross-sectional data from the Demographic and Health Surveys (DHS) of 45 LMICs collected between 2012 and 2022 which utilized a two-stage stratified sampling method to include women who were interviewed for the domestic violence module and had a child 2 years old or younger. Multivariable log-binomial regression models were utilized to examine the associations between IPV, both during pregnancy and at other times, and the specified outcomes. FINDINGS: After applying survey weights, 100,199 women were included in the analyses. The prevalence of physical IPV during pregnancy was 6.07% (n = 6078). Adjusted for covariates, physical IPV during pregnancy was negatively associated with adequate antenatal care utilization (RR = 0.88; 95% CI = 0.84, 0.91), EIBF (RR = 0.92; 95% CI = 0.89, 0.96), and EBF2D (RR = 0.93; 95% CI = 0.90, 0.96). While physical IPV at other times also negatively impacted most outcomes, the effect was more pronounced when physical IPV occurred during pregnancy. INTERPRETATION: Physical IPV both during pregnancy and at other times pose significant barriers to maternal healthcare utilization and optimal breastfeeding practices in LMICs, with the impact of physical IPV during pregnancy being more severe. Targeted antenatal care interventions addressing physical IPV during pregnancy could improve health outcomes for both mothers and children. FUNDING: The manuscript was supported by the European Research Council Consolidator Grant IMPROVE_LIFE (Grant number 101124718).