Abstract
<title>Abstract</title> <p>Background Depression and physical inactivity are modifiable cognitive decline risks highlighted by the Lancet dementia commission. Existing research cannot clarify their relative within-person predictive impacts on cognitive trajectories among older adults, hindering targeted community dementia prevention strategies. Methods Data from 2,435 older adults (mean age 60.68 ± 7.27 years) in the China Health and Retirement Longitudinal Study were analyzed across three waves (2015 to 2020). Physical activity was measured using a modified International Physical Activity Questionnaire. Cognitive function was assessed through the Telephone Interview of Cognitive Status instrument. Depression was quantified using the 10-item Center for Epidemiologic Studies Depression Scale (range: 0–30 points). The Random-Intercept Cross-Lagged Panel Model was employed to examine between-person associations and within-person temporal dynamics. Results At the between-person level, cognitive function was negatively associated with depression (beta=-0.203, p < 0.01) and positively with physical activity (beta = 0.135, p < 0.01), while the physical activity-depression association was not significant. Within-person cross-lagged analyses revealed that depressive symptoms significantly predicted subsequent cognitive decline (beta=-0.046 to -0.043, p < 0.05), and cognitive function predicted subsequent physical activity engagement (beta = 0.059–0.073, p < 0.05). Notably, physical activity did not significantly predict subsequent cognitive function or depression at the within-person level. Conclusion When stable individual differences are accounted for, depressive symptoms emerge as a stronger within-person temporal predictor of cognitive decline than physical activity. These findings suggest that depression screening may warrant particular attention in dementia prevention initiatives, while highlighting the need for intervention studies to establish causal effects.</p>