Maternal and paternal depressive symptoms from pregnancy to late childhood in a UK-birth cohort: reciprocal and bidirectional effects

Background:
Evidence regarding the dynamics of the longitudinal reciprocal associations between maternal and paternal depression over time is lacking. The current study examines reciprocal longitudinal associations between maternal and paternal depressive symptoms (DS) from pregnancy (18 weeks-gestation) to 11 years 2 months (study child's age) to gain insights into the joint course of DS.

Methods:
The study sample comprised 6296 mothers and fathers from the UK-based birth cohort, the Avon Longitudinal Study of Parents and Children. Reciprocal longitudinal associations between maternal and paternal DS, assessed at nine time-points (Edinburgh Postnatal Depression Scale; continuous scores), were examined using random intercept cross-lagged panel models to differentiate between stable covariance (between-person differences in DS) and dynamic processes (within-person level). Analyses were adjusted for parental and socioeconomic confounders.

Results:
There were strong correlations between random intercepts of maternal and paternal DS (β = 0.305, SE = 0.015, p ≤ 0.001), suggesting positive associations between the overall mean levels of parental DS (i.e., a between-person association). There were strong stability effects for maternal and paternal DS across all nine measurement occasions (i.e., within-person associations), with larger effects evident for both maternal and paternal DS later in childhood (between ages 5 and 11 years). There was evidence for bidirectional longitudinal within-person effects between maternal and paternal DS at some but not other measurement occasions.

Conclusions:
These findings have important implications for screening, prevention and intervention programmes. Depression in one parent may require screening efforts directed at the other parent, with prevention and intervention programmes focused on the family, rather than one parent.