Dynamic frailty and depressive symptoms in relation to incident stroke: findings from five harmonized longitudinal...

Frailty and depressive symptoms are common in later life and may be related to cerebrovascular risk. Evidence remains limited on whether frailty burden and frailty change are associated with incident stroke across diverse aging cohorts.

We analyzed harmonized longitudinal data from five population-based aging cohorts: the Health and Retirement Study (HRS), China Health and Retirement Longitudinal Study (CHARLS), Survey of Health, Ageing and Retirement in Europe (SHARE), English Longitudinal Study of Ageing (ELSA), and Mexican Health and Aging Study (MHAS). Frailty was measured using a harmonized 24-item deficit-accumulation frailty index (FI). The primary analysis used cohort-specific Cox proportional hazards models to estimate associations between baseline FI and first observed incident stroke during follow-up. Secondary exploratory analyses evaluated nonlinearity, FI change, competing mortality, depressive symptoms as a pathway marker, and two-wave cross-lagged associations.

The analytic sample included 81482 participants and 5,089 incident stroke events. In fully adjusted cohort-specific Cox models, each 0.1-unit increase in FI was associated with higher stroke risk in HRS, CHARLS, SHARE, and MHAS, but not in ELSA. Substantial between-cohort heterogeneity was observed; therefore, cohort-specific estimates were interpreted as the primary results and the random-effects pooled estimate was treated as descriptive. Fine-Gray sensitivity analyses treating death as a competing event supported positive frailty-stroke associations across all five cohorts. Restricted cubic spline (RCS) analyses suggested nonlinear associations for baseline FI and FI change. Exploratory pathway analyses indicated that depressive symptoms statistically accounted for part of selected frailty-stroke associations, although patterns varied by cohort and exposure definition. Two-wave cross-lagged panel models (CLPMs) suggested small, cohort-specific prospective associations between elevated frailty vulnerability and later depressive symptoms or stroke; these findings were interpreted as exploratory temporal associations rather than causal within-person effects.

Higher frailty burden was associated with incident stroke in most, but not all, harmonized aging cohorts, with substantial heterogeneity across populations. The findings support repeated frailty assessment and integrated mood evaluation in older adults while emphasizing the need for cohort-specific interpretation and confirmatory studies with adjudicated stroke outcomes.