Prenatal stress, epigenetically-assessed glucocorticoid exposure at birth, and child psychiatric symptoms: A prospective, multi-cohort study

Abstract

Background

Recent work suggests that DNA methylation can be used as a proxy of fetal glucocorticoid exposure (MPS-GC), showing associations with maternal psychopathology during pregnancy. However, it is unknown whether the MPS-GC may act as a marker for broader prenatal stress and whether it partially mediates associations of prenatal stress with child internalizing and externalizing symptoms.

Methods

Using harmonized data from three prospective birth cohorts (N_pooled = 6086), we examined whether a cumulative measure of prenatal stress, and its individual stress domains, associate with the MPS-GC in cord blood at birth. Next, we examined (i) whether the MPS-GC at birth associates with child psychiatric symptoms, (ii) whether this association is moderated by postnatal stress, and (iii) whether the effect of prenatal stress on child psychiatric symptoms is partially mediated by the MPS-GC at birth.

Results

Our meta-analysis revealed no significant associations between the MPS-GC at birth and prenatal stress or the individual stress domains. Moreover, the MPS-GC did not significantly associate with later child internalizing or externalizing symptoms, and there were no moderating effects of postnatal stress. Additionally, while prenatal stress significantly associated with child psychiatric symptoms, we found no partial mediation via the MPS-GC at birth.

Conclusions

We did not find support that the MPS-GC in cord blood reliably proxies prenatal stress, associates with child psychiatric risk, or partially mediates the associations between prenatal stress and psychiatric risk.