The bidirectional dynamic relationship between depression symptoms and activities of daily living disability among the elderly: Evidence from three-wave longitudinal data

BACKGROUND: There remains a lack of multi-wave longitudinal studies capable of examining the dynamic bidirectional relationship between depression symptoms and ADL, and the understanding of how this association evolves over time remains incomplete. METHODS: Our study utilizes three-wave data from the China Health and Retirement Longitudinal Study (CHARLS) collected in 2015 (T0), 2018 (T1), and 2020 (T2). Correlation analyses were used to explore the association between depression symptoms and ADL disability. Cross-lagged model (CLPM) adjusted by inverse probability weighting (IPW) was employed to validate the dynamic relationship between the two over time. RESULTS: Changes in depression symptoms and ADL disability scores vary significantly over different time periods by sample characteristics. Correlation analyses revealed significant positive correlations between depression symptoms and ADL disability at all three waves. CLPM results revealed that depression symptoms and ADL disability mutually predict each other in the early phase (T0 Depression symptoms→T1 ADL disability: β = 0.048, P < 0.001; T0 ADL disability→T1 Depression symptoms: β = 0.043, P = 0.003) and negatively in the later phase (T1 Depression symptoms→T2 ADL disability: β = -0.057, P = 0.006; T1 ADL disability→T2 Depression symptoms: β = -0.069, P < 0.001). LIMITATIONS: Depression symptoms and ADL disability are assessed through self-reporting. CONCLUSION: Depression symptoms and ADL disability in elderly people exhibit a phased bidirectional predictive relationship over time, with depression symptoms having a stronger driving effect. A phased, integrated intervention strategy is required to break this cycle.