The Bidirectional Relationship Between Subjective Well-Being and Depression: A Cross-Sectional and Cross-Lagged Network Analysis

Network modeling has been suggested as an effective approach to uncover intricate relationships among emotional states and their underlying symptoms. This study aimed to explore the dynamic interactions between subjective well-being (SWB) and depressive symptoms over time, using cross-sectional and cross-lagged network analysis. Data were drawn from three waves (2016, 2018, and 2020) of the China Family Panel Studies (CFPS), including 13,409 participants aged 16 and above. SWB was measured through indicators like life satisfaction and future confidence, while depressive symptoms were assessed using the CES-D8 scale. Symptom-level interactions were analyzed via cross-sectional network analysis at each wave, and cross-lagged panel network analysis was employed to examine the temporal dynamics and bidirectional relationships between SWB and depressive symptoms. The cross-sectional symptom network analysis showed that the number of non-zero edges at T1, T2, and T3 were 50, 44, and 49, respectively, with network densities of 0.90, 0.80, and 0.89. The core symptom "feeling sad" (D7) consistently had a significantly higher strength than other symptoms. The negative correlation between "life satisfaction" (Z2) and depressive symptoms was particularly evident at T3. The cross-lagged symptom network analysis revealed the key roles of "feeling lonely" (D5) and "feeling sad" (D7), as well as "feeling unhappy" (D4) and "not enjoying life" (D6) across different time periods, which may form a negative feedback loop. "Life satisfaction" (Z2) and "confidence in the future" (Z3) exhibited significant protective effects, forming a positive feedback loop that suppresses negative emotions through mutual reinforcement. Stability analysis showed that the network structure was stable, with a centrality stability coefficient of 0.75. The study reveals a dynamic, bidirectional relationship between SWB and depressive symptoms. These results offer valuable insights for targeted interventions and public health initiatives aimed at improving mental well-being.