BackgroundPerinatal mortality, encompassing late stillbirths and neonatal mortalities, is a key indicator of maternal and neonatal health. Despite advances in health care, there have been alarming increased in perinatal mortality rates in Mexico. This study investigated the influence of social determinants on perinatal death in a Latin American middle-income country, aiming to inform equitable and effective health policies.MethodsA prospective cohort study was conducted with pregnant women from Mexico City. Data on the following clinical and social factors were collected: pregestational body mass index (pBMI), lupus, antiphospholipid syndrome (APS), preeclampsia, foetal growth restriction (FGR), social vulnerability, poverty, household overcrowding, and gender-based violence. Nested logistic regression models were developed to identify significant predictors of perinatal mortality, with the results reported as odds ratios (ORs) and 95% confidence intervals (CIs).ResultsAmong the 3,890 participants, there were 76 cases of perinatal mortality. Significant clinical predictors of perinatal mortality included higher pBMI (OR = 1.088, 95% CI 1.026–1.153), APS (OR = 10.049, 95% CI 1.843–54.803), and FGR (OR = 2.929, 95% CI 1.399–6.135), whereas high social vulnerability (OR = 5.332, 95% CI 2.485–11.443) and medium social vulnerability (OR = 3.084, 95% CI 1.528–6.222) emerged as significant social predictors of perinatal mortality. A comprehensive model incorporating both clinical and social determinants achieved an AUC of 0.921, with a detection rate of 67.1% and a false-positive rate of 10%, this indicating a significant improvement in perinatal mortality prediction. The inclusion of social determinants progressively enhanced predictive performance, underscoring their critical role in risk assessment.ConclusionsClinical and social determinants significantly influence perinatal mortality. Addressing social inequalities and integrating social determinants into perinatal care could improve maternal and neonatal health outcomes. However, limitations such as reliance on self-reported data, ecological-level indicators for social vulnerability, and potential constraints on generalizability should be considered. These findings emphasize the need for targeted health policies to reduce social vulnerabilities and enhance health care access in middle-income countries.
Read full abstract