Childhood abuse predicts affective symptoms via HPA reactivity during mother-infant stress.

Despite extensive literature positing the hypothalamic-pituitary-adrenal (HPA) axis as a mechanism in the association between early childhood maltreatment and later adult psychopathology, empirical support for this full pathway is lacking. We tested indirect effects of childhood maltreatment on women's later affective symptomatology via HPA axis responding to a stressor involving their own infant. Women (n = 47) in a larger longitudinal study were assessed following the birth of their infant from 3 to 18 months postnatal. They reported childhood maltreatment history at 3 months and participated in a dyadic stress task with their infant at 12 months, at which time four salivary cortisol samples were collected to assess HPA response. Depression and anxiety symptoms at 18 months (controlling for symptom levels reported at 12 months) served as the primary outcome. Multilevel modeling was used to estimate both levels and dynamics of women's cortisol response trajectories. Tests of indirect effects revealed a significant effect of total Childhood Trauma Questionnaire (CTQ) scores on anxiety symptoms and a marginally significant effect on depression symptoms. Follow-up analyses with CTQ subscales revealed significant indirect effects of emotional and physical abuse on women's ongoing anxiety symptoms via more pronounced cortisol reactivity curves during the mother-infant stressor. We discuss methodological choices that may have allowed these effects to be detected in the present study and implications for stress-related risk and intervention.