In a sample of 1100 families in England, the relationship between both mothers’ and fathers’ depressive symptoms (Edinburgh Postnatal Depression Scale, 3 months; General Health Questionnaire, 36 months) and children’s behavioural and emotional difficulties (Infant Characteristics Questionnaire, 3 months; Strengths and Difficulties Questionnaire, 51 months) was investigated. A series of bivariate and multivariate analyses found maternal postnatal depressive symptoms to be a significant predictor of children’s later internalising and externalising difficulties as reported by mothers, whilst controlling for the effects of marital difficulties and socioeconomic status. This association was attenuated when later maternal psychological symptoms were accounted for, with early maternal depressive symptoms remaining a significant predictor of total difficulties, externalising problems and peer problems in girls only. However, maternal postnatal depressive symptoms were not found to significantly predict father- or teacher-reported child difficulties. Paternal postnatal depressive symptoms did not predict any child difficulties when marital difficulties, socioeconomic status and later paternal
psychological symptoms were controlled. Marital difficulties (Dyadic Adjustment Scale, 3 months) was seen to be the strongest predictor of child difficulties across models. No child-to-parent effects were found, supporting the claim that the direction of effect was from parent-to-child. Although recent evidence has shown paternal depression to adversely affect offspring, the present findings suggest that interventions should primarily target depressed mothers of infant daughters. However, the multi-informant discrepancies seen here highlight the need to be cautious when interpreting results from a single informant.