Network Analysis of Depressive Symptoms Among Chinese Older Adults Across Different Intergenerational Support Patterns

Background Intergenerational support plays a crucial role in mental health in later life, yet both the heterogeneity of support patterns and the symptom-level interplay among depressive symptoms remain underexplored. Methods Using data from the China Health and Retirement Longitudinal Study (CHARLS 2020, N = 7,444), Latent Class Analysis (LCA) identified intergenerational support patterns. Network analysis and the Network Comparison Test (NCT) were utilized to estimate and compare depressive symptom networks (CESD-10) across subgroups. Results Three patterns emerged: Distant-Supportive (55.9%), Proximate-Interactive (24.2%), and Traditional Co-resident (19.9%). While “sadness” and “being bothered” were universal core symptoms, NCT revealed distinct structural disparities. The Proximate-Interactive group exhibited a uniquely strong “bothered–fearful” connection. The Distant-Supportive group showed pronounced “effort–cannot get going” pathways, whereas the Traditional Co-resident group displayed a stronger “bothered–trouble focusing” link. Conclusions Intergenerational support patterns are not only associated with overall depression severity but also with differences in the network topology of depressive symptoms. Differentiated interventions tailored to specific family structures are essential for geriatric mental health.