Postnatal maternal depressive symptoms and behavioural outcomes in term- and preterm-born toddlers

Abstract

AbstractObjectivesTo examine the association between maternal depressive symptoms in the immediate postnatal period and offspring’s mental health in a large cohort of term- and preterm-born toddlers.Design and ParticipantsData were drawn from the Developing Human Connectome Project. Maternal postnatal depressive symptoms were assessed at term, and children’s outcomes were evaluated at a median corrected age of 18.4 months (range 17.3 – 24.3).Exposure and outcomesPreterm birth was defined as <37 weeks completed gestation. Maternal depressive symptoms were assessed with the Edinburgh Postnatal Depression Scale (EPDS). Toddlers’ outcome measures were parent-rated Child Behaviour Checklist 11/2-5 Total (CBCL) and Quantitative Checklist for Autism in Toddlers (Q-CHAT) scores. Toddlers’ cognition was assessed with the Bayley Scales of Infant and Toddler Development – Third Edition (Bayley-III).ResultsHigher maternal EPDS scores were associated with toddlers’ higher CBCL (B=0.93, 95% CI 0.43-1.44, p<0.001, f2=0.05) and Q-CHAT scores (B=0.27, 95% CI 0.03-0.52, p=.031, f2=0.01). Higher maternal EPDS scores were not associated with toddlers’ cognitive outcomes. Maternal EPDS, toddlers’ CBCL and Q-CHAT scores did not differ between preterm (n=97; 19.1% of the total sample) and term participants.ConclusionsOur findings indicate that children whose mothers had increased depressive symptoms in the early postnatal period, including subclinical symptoms, exhibit more maternally-reported behavioural problems in toddlerhood. These associations were independent of gestational age. Further research is needed to confirm the clinical significance of these findings.Strengths and limitations of this studyProspective study with a large sample, using multiple imputation to reduce non-response bias.Maternal depressive symptoms assessed as a continuous variable, providing more nuanced information about the significance of subclinical symptoms.Maternal depressive symptoms assessed earlier than in previous studies, enabling recognition of early screening opportunities for families.Potential shared method variance bias through parent-completed child behavioural assessments.Unknown paternal and parental factors, such as comorbid psychiatric conditions, that may confound our findings.