Longitudinal relationships between maternal depressive symptoms, perceived social support, and sleep problems across mid-pregnancy to 6 weeks postpartum: A cross-lagged panel network analysis.

BACKGROUND

Maternal depressive symptoms, sleep problems, and social support are interrelated during the perinatal period, but their temporal and symptom-level interactions remain poorly understood. This study examined the longitudinal interrelationships among these three domains from pregnancy to postpartum.

METHODS

A prospective cohort of 3376 women was assessed at 24 gestational weeks (T1), 32 gestational weeks (T2), 1 week postpartum (T3), and 6 weeks postpartum (T4). Depressive symptoms were measured with the Edinburgh Postnatal Depression Scale (EPDS), sleep quality with the Pittsburgh Sleep Quality Index (PSQI), and social support with the Social Support Rating Scale (SSRS). Cross-lagged panel network (CLPN) models were estimated for three intervals (T1 → T2, T2 → T3, T3 → T4).

RESULTS

High depressive symptoms (EPDS ≥10) ranged from 21.6% to 31.8%, and sleep problems (PSQI ≥5) increased to 74.8% at T4. CLPN models revealed strong autoregressive pathways and dynamic within-community reinforcement among depressive and sleep symptoms. Support utilization increasingly predicted subsequent subjective support, strengthening from mid-pregnancy (β = 0.27) to early postpartum (β = 0.69). Cross-community effects varied by stage: difficulty coping and reduced enjoyment predicted lower subjective support during pregnancy, daytime dysfunction predicted anxious symptoms from late pregnancy to postpartum, and postpartum sleep disturbance was the strongest predictor of reduced subjective support (β = -0.36).

CONCLUSIONS

Depressive symptoms, sleep problems, and social support interact dynamically at the symptom level across the perinatal period. Stage-specific interventions targeting coping, sleep, and help-seeking behavior may improve maternal mental health outcomes.