Propensity score analysis of psychological intimate partner violence and preterm birth

Abstract

Psychological intimate partner violence (IPV), a global public health problem, affects mothers during pregnancy. We evaluated its relationship with preterm birth. We established a cohort of 779 consecutive mothers receiving antenatal care and giving birth in 15 public hospitals in Spain. Trained midwives collected IPV data using the Index of Spouse Abuse validated in the Spanish language. Preterm was defined as birth before 37 completed weeks of gestation. Gestational age was estimated by early ultrasound. With multivariate logistic regression we estimated the relative association of IPV with preterm birth as adjusted odds ratios (AOR), with 95% confidence intervals (CI). In propensity score analysis, using weighting by inverse probability of exposure to IPV, the whole sample was used for estimating the absolute difference in probability of preterm amongst offspring born to mothers with and without IPV. Socio-demographic and other pregnancy characteristics served as covariates in both analyses. Preterm occurred in 57 (7.3%) pregnancies. Psychological IPV, experienced by 151 (21%) mothers, was associated with preterm birth (11.9% vs 6.5%; AOR = 2.4; 95% CI = 1.1–5.0; p = 0.01). The absolute preterm difference in psychological IPV compared to normal was 0.08 (95% CI = 0.01–0.16; p = 0.04). The probability of preterm birth was 8% higher on average in women with psychological IPV during pregnancy. As our analysis controlled for selection bias, our findings give credence to a causal inference. Screening and management for psychological IPV during pregnancy is an important step in antenatal care to prevent preterm birth.