Background: Stress and anxiety in pregnancy and the postpartum period are associated with poor long-term maternal and child health outcomes. Social support has been shown to mitigate the effects of maternal stress and anxiety; however, the directionality and longitudinal associations among these variables are poorly understood. Using a novel multilevel modeling approach called dynamic structural equation modeling (DSEM), we examined within-person (state-level) autoregressive and cross-lagged associations among stress, anxiety, and social support in the perinatal period in order to elucidate directional associations over time.
Methods: Mothers from a longitudinal pregnancy cohort (N = 3,388) completed self-report measures of stress, anxiety, and social support across 4 time points from pregnancy to 12 months postpartum.
Results: Higher than average levels of stress and anxiety led to elevations in anxiety and stress and decreases in social support at subsequent time points. Importantly, earlier individual levels of partner and family support predicted subsequent decreases in stress and anxiety.
Limitations: Support was measured via maternal self-report thus extrapolations cannot be made to tangible or instrumental supports and lagged relationships represent average lags over time.
Conclusions: Using a novel statistical approach, these results suggest that increases in both partner and family support may be powerful protective factors for decreasing mental health difficulties in pregnancy and the postpartum, highlighting the importance of targeting and increasing this type of support from pregnancy to the postpartum period.
Keywords: Anxiety; Domestic partners; Postpartum period; Pregnancy; Social support; Stress.
Copyright © 2019. Published by Elsevier B.V.