Joint trajectories of frailty and depressive symptoms and risk of incident osteoarthritis in middle-aged and older adults.

BACKGROUND: Frailty frequently overlaps with depressive symptoms among middle-aged and older adults. However, it remains unclear how these two conditions develop concurrently over time and whether their combined longitudinal patterns are associated with the subsequent onset of osteoarthritis (OA). This study aimed to characterize the joint trajectories of frailty and depressive symptoms and to evaluate their prospective association with incident OA. METHODS: We analyzed 3,822 participants from the Health and Retirement Study (HRS) and 3,913 from the English Longitudinal Study of Ageing (ELSA). All participants were free of self-reported osteoarthritis (OA) at the analytical baseline. Frailty and depressive symptoms were assessed at four biennial waves using the frailty index (FI) and the 8-item Center for Epidemiologic Studies Depression Scale (CES-D-8), respectively. Group-based multi-trajectory modeling identified joint longitudinal patterns within each cohort. Cox proportional hazards models estimated hazard ratios (HRs) and 95% confidence intervals (CIs) for incident OA during a median follow-up of approximately 8 years. RESULTS: Three similar joint trajectory groups were identified in both cohorts: persistently low, moderate, and high combined frailty and depressive symptom burden. In fully adjusted HRS models, the HRs were 1.50 (95% CI, 1.28-1.76) for the moderate-burden group and 2.19 (95% CI, 1.68-2.86) for the high-burden group. The corresponding HRs in ELSA were 1.87 (95% CI, 1.60-2.17) and 3.40 (95% CI, 2.74-4.21). Sensitivity analyses using logistic regression, first-wave event exclusion, complete cases, and a four-group trajectory model produced broadly consistent associations. CONCLUSION: Persistently higher combined frailty and depressive symptom burden was prospectively associated with incident OA in two ageing cohorts. These findings describe an epidemiological association and do not establish causal, preventive, predictive, or treatment effects.