Abstract / Summary
Social isolation and chronic multimorbidity severely affect older people’s health, while evidence on their longitudinal association and gender differences remains limited. Based on 2011–2020 CHARLS data, this study explored their dynamic relationship and gender heterogeneity. A total of 7125 older people aged 60 and above with at least three survey waves were enrolled. Random-intercept cross-lagged panel model (RI-CLPM), parallel latent growth model, and multi-group SEM were adopted. The RI-CLPM showed excellent fit (CFI = 0.992, TLI = 0.986, RMSEA = 0.034, SRMR = 0.020). Strong autoregressive effects were observed for both social isolation (β = 0.096–0.224) and multimorbidity (β = 0.619–1.035). Cross-lagged effects were largely non-significant, with the exception of a positive effect from multimorbidity to isolation in the final wave (β = 0.055, P = 0.002). Their developmental trajectories were synchronous (r = 0.102, P = 0.015). Gender-stratified analyses showed that women had higher initial levels and faster growth rates in both variables, with a significant gender moderating effect (Δχ²=68.34, P < 0.001). Social isolation and multimorbidity exhibit synchronous deterioration in older people, especially among females, though the within-person cross-lagged effects are limited. The co-occurrence appears to be primarily driven by stable between-person differences. Targeted interventions considering gender differences are needed to promote active aging.