The Longitudinal Relationships Between Cognitive Impairment and Depression in Older Adults: A Cross-Lagged Panel Network Analysis

Objectives To investigate the prospective associations and network characteristics between cognitive impairment and depression among individuals aged 60 years or older. Design A longitudinal study employing a cross-lagged panel network (CLPN) analysis. Setting Community-based setting in China. Participants A total of 6,297 older adults participated, recruited via cluster sampling. Measurements Cognitive function was assessed using the Ascertain Dementia 8-item questionnaire (AD-8). Depressive symptoms were measured using the Patient Health Questionnaire-9 (PHQ-9). A cross-lagged panel network analysis was employed to establish the symptom network spanning time-point T1 to T2. Centrality metrics—including out-strength, in-strength, betweenness, and closeness—were quantified. Results The results show that cognitive impairment (particularly AD-8 item 5: time disorientation and item 1: impaired judgment) was identified as the strongest prospective predictor of depressive symptoms, whereas the reverse influence of depressive symptoms on cognition was comparatively weak. Impaired judgment showed high out-predictability, suggesting its potential as a candidate target for early intervention; conversely, the high in-predictability of fatigue indicates it may be a sensitive marker of upstream network changes. Conclusions This study provides evidence that cognitive impairment prospectively predicts depression in an asymmetric manner and offers a conceptual cascade model. These findings support further research into a public health shift toward staged interventions, from early screening to targeted strategies, to alleviate the population burden of this comorbidity.