ABSTRACT
The hypothalamic-pituitary-adrenal (HPA) axis in pregnancy has attracted considerable research attention, in part, because it may be a mechanism by which diverse prenatal exposures alter perinatal and child health outcomes. Symptoms of affective disturbance and stress are among the most-studied prenatal factors associated with HPA axis alterations, but there remains uncertainty about the nature of the association because of the limitations to, and variability in, data collection and analytic approaches. The current study capitalized on a prospective, longitudinal pregnancy cohort that examined salivary diurnal cortisol, collected at 5 time points across the day, at each trimester in a diverse sample of women. Detailed data on affective symptoms and major life events were collected at each trimester, as were data on health behaviors, medication, and socio-demographics. Results indicated modest stability of individual differences in diurnal cortisol across pregnancy, which was evident for diurnal slope (ICC = .20) and measures of total output (area under the curve, ICC = .25); substantial gestation-related increases in total cortisol output across pregnancy was also observed (p < .001). Adjusting for health behaviors, medication, and socio-demographic covariates, elevated levels of depressive symptoms and major life events were significantly (p < .05) associated with a higher morning awakening value and flatter diurnal slope, which was evident across all trimesters. In addition to the normative gestation-related changes in cortisol production, our results demonstrate selective but robust associations between psychological symptoms, stressors, and the HPA axis across gestation, and suggest both methodological and mechanistic strategies for future study.