Longitudinal effects of maternal depressive and anxious symptomatology on child hair cortisol and cortisone from pregnancy to 5-years: The EDEN mother-child cohort

Exposure to maternal depressive and anxious symptomatology in utero and after birth can affect child outcomes. One proposed mechanism is through changes in child stress hormone levels, however current studies present inconsistent findings, and further research is needed to better understand the impact of maternal mental health on child stress response. This study aims to add to the limited literature by analysing longitudinal data ranging from 24 weeks amenorrhea to 5 years postpartum among 281 mother-child pairs from the French EDEN motherchild birth cohort. Hair cortisol and cortisone data were collected from children at four time points: birth, 1, 3, and 5 years. Mothers reported depressive symptomatology via the Center for Epidemiologic Studies Depression Scale (CES-D) (at 24-weeks amenorrhea, 3-, and 5-year follow-up), and the Edinburgh Postnatal Depression Scale (EPDS) (at 4, 8 and 12 months postpartum). Prenatal anxiety symptomatology was measured via the State Anxiety Inventory (STAI) at 24 weeks amenorrhea. Group-based trajectory modelling indicated a 1-cluster classification of longitudinal child hair cortisol, cortisone and cortisol-to-cortisone ratio, as analyses did not reveal a classification by subgroups representing different child profiles. After inverse probability weighting, small effects showed prenatal depressive symptomatology was significantly associated to higher levels of child hair cortisone at one year. Prenatal anxiety symptomatology was significantly linked to higher levels of child cortisol measured at birth and cortisone at birth and at 1 year. Postpartum depressive symptomatology at 8 months was related to higher levels of cortisone among 3-year-olds. These effects were not moderated by child sex or maternal socio-economic status. Further research is needed to understand why there are associations at some time points and not others to determine any potential buffering factors.