Prenatal stress, intimate partner violence and maternal cortisol trajectories: insights from a prospective birth cohort from São Paulo, Brazil.

Prenatal stress has been implicated in alterations of maternal hypothalamic-pituitary-adrenal (HPA) axis functioning, yet evidence linking psychosocial stressors to long-term cortisol biology remains inconsistent. Hair cortisol concentration (HCC) provides an integrated measure of cumulative glucocorticoid exposure, allowing investigation of whether diverse prenatal stressors shape maternal cortisol trajectories during pregnancy and postpartum. We analyzed data from a prospective population-based birth cohort from São Paulo, Brazil. Pregnant women (n = 185) were assessed in the third trimester for exposure to intimate partner violence (IPV), emotional responses to pregnancy, and anxiety or depression using standardized instruments (WHO-VAW, MINI). Maternal HCC was quantified from 7 cm hair segments representing the last four months of pregnancy and the first two postpartum months. We compared cortisol levels across exposure groups using t-tests and Mann-Whitney tests, and estimated longitudinal associations with linear mixed-effects models. Cortisol levels declined significantly from pregnancy to postpartum (p < 0.001). Most prenatal stressors-including psychological IPV, anxiety, depression, and negative emotional responses to pregnancy-were not associated with HCC at individual time points nor in longitudinal models. Acute prenatal physical/sexual IPV was associated with lower HCC at specific prepartum time points and with lower mean HCC in longitudinal analyses. In this cohort, maternal cortisol biology showed limited sensitivity to most psychosocial stressors during pregnancy. Only acute physical or sexual IPV was associated with reduced cumulative cortisol levels. These findings highlight the specificity of severe interpersonal violence as a biological stressor and underscore the complexity of interpreting cortisol as a biomarker of prenatal stress.