Analysis of the subject-object effects and chain mediation of psychosocial factors in couples during late pregnancy on depression: a cross-sectional study of pregnant women and their spouses

Abstract

Objective To examine, from a dual-perspective approach, the actor and partner associations of couples' perceptions of stress, rumination, self-efficacy, self-disclosure, and family intimacy with depressive symptoms during late pregnancy, and to test the chain mediation mechanism of “stress—family intimacy—rumination/self-disclosure—depression. Methods From July 2025 to May 2026, 1,331 pairs of pregnant women and their spouses attending prenatal checkups at the outpatient clinic of a Grade III Level A hospital in Nanjing were selected as study participants using convenience sampling. The Patient Health Questionnaire (PHQ-9), the Perceived Stress Scale (PSS), the Rumination Scale (RRS), the General Self-Efficacy Scale, the Distress Disclosure Scale, and the Family Affection and Cohesion Scale (FACES II) were used to collect data. Data analysis was performed using paired-sample t-tests, multiple linear regression, and the Agent-Patient Interdependence Model (APIM), with the Bootstrap method used to test for mediating effects. Results (1) Positive Detection Rate for Symptoms of Depression among women was 18.3%, and among their spouses, it was 16.7%. (2) Spouses' perceived stress and ruminative thinking were significantly higher than those of pregnant women (both P < 0.001), while pregnant women's self-disclosure, self-efficacy, family closeness, and adaptability were all significantly higher than those of their spouses (all P < 0.001). (3) Multiple linear regression analysis revealed that ruminative thinking (pregnant women: β = 0.337; spouses: β = 0.282) and perceived stress (pregnant women: β = 0.153; spouses: β = 0.240) were common core predictors of depression in both partners. (4) Univariate APIM analysis revealed that all subject effects were significant (P < 0.001), while only the spouses' rumination (β = 0.067, P = 0.003) and perceived stress (β = 0.052, P = 0.028) exhibited significant object effects. (5) The chained mediation model showed that the full pathway “stress → reduced family intimacy → increased rumination → worsened depression” exhibited significant indirect associations for both spouses, with the proportion of indirect effect higher in the pregnant women group (49.70%) than in the spouse group (35.66%); the pathway “stress → reduced family intimacy → decreased self-disclosure → increased depression” was also significant but showed a weaker effect. Cross-mediation analysis indicated that stress in one partner was indirectly associated with the other partner's depression levels, with this association occurring through the perceived family intimacy of the partner. Conclusion Depression among couples in late pregnancy is primarily driven by intra-individual associations, with rumination and perceived stress serving as core correlates; family intimacy may act as a key correlate linking stress to depression and cross-individual spillover patterns. Family intimacy was indirectly associated with depression levels in both partners, and these associations operated through rumination and self-disclosure levels. Clinical interventions should focus primarily on individual psychological adjustment to reduce stress and ruminative thinking in both the expectant mother and the father, while also emphasizing enhanced communication to improve marital intimacy and alleviate depressive symptoms in both partners.