Abstract
Background: Depressive symptoms, pain, and frailty interactions in middle-aged and older adults do have longitudinal research support, yet the currently available evidence remains insufficient for a comprehensive understanding. This study aimed to examine their interrelationships and underlying mechanisms. Methods: This study utilized data from the China Health and Retirement Longitudinal Study (CHARLS), which includes four assessments for depressive symptoms, pain, and frailty over 7 years. We included 4961 participants aged 45 years and older in our analysis. We employed cross-lagged panel models (CLPMs) and random intercept CLPMs (RI-CLPMs) to analyze the bidirectional temporal relationships at the between-person and within-person levels. Results: Cross-lagged panel analysis revealed bidirectional associations between depressive symptoms and pain. A reciprocal predictive relationship was also identified between frailty and pain. After controlling individual differences with the RI-CLPM, depressive symptoms continued to positively predict pain during subsequent periods. However, the predictive effect of pain on subsequent depressive symptoms turned nonsignificant. Although a cross-lagged relationship remained between pain and frailty, it showed a declining trend. Limitations: While engagement in depressive symptoms, pain, and frailty was assessed via questionnaire, long measurement intervals may not capture short-term fluctuations in the state of each variable. Conclusions: This study differentiated within- and between-individual effects, uncovering distinct lagged effects of pain, depression, and frailty across levels. It underscored the importance of jointly assessing these conditions and integrating within- and between-individual differences to formulate and execute targeted interventions.