The Dynamic Interplay Among Cognitive Function, Depressive Symptoms, and Self-Rated Health in Middle-Aged and Older Chinese Adults: A Cross-Lagged Panel Study

Background The frequent co-occurrence of cognitive decline and depressive symptoms in aging suggests a complex, potentially bidirectional relationship. Self-rated health (SRH) is holistically linked to both. However, the longitudinal dynamics among these three constructs, particularly distinguishing within-person processes from stable between-person differences and the role of gender, remain insufficiently understood. Methods Using two-wave longitudinal data (2018 and 2020) from the China Health and Retirement Longitudinal Study (CHARLS), this study analyzed 4,161 adults aged 45+. We employed a dual-model strategy: a traditional Cross-Lagged Panel Model (CLPM) and a Random-Intercept Cross-Lagged Panel Model (RI-CLPM) to isolate within-person dynamics. Multi-group analysis tested for gender invariance. Covariates included age, education, and functional limitations. Results Both models revealed a significant reciprocal negative relationship between cognitive function and depressive symptoms over two years. Crucially, the RI-CLPM showed this negative cycle was substantially stronger at the within-person level than CLPM estimates. SRH consistently predicted subsequent depressive symptoms but showed no direct effect on future cognitive function. The cognitive autoregressive path reversed from positive in the CLPM to negative in the RI-CLPM, indicating within-person "regression to the mean." Core dynamic pathways were invariant across genders. However, the association between education and the stable trait component of cognitive function was significantly stronger for female. Conclusion The interplay between cognition and depression in later life is primarily a within-person, mutually reinforcing cycle. SRH influences depressive symptoms but not directly cognitive trajectories. While the core dynamics are similar across genders, the foundation of cognitive reserve shaped by education shows gendered patterning. Findings underscore the value of within-person methodologies and highlight the need for integrated interventions addressing mood and cognition, considering gender-specific life-course factors like education.