Frailty index changes before and after stroke: Findings from four cohorts covering 18 countries.

BACKGROUND: Frailty is a dynamic predictor of adverse stroke outcomes, but its bidirectional relationship with stroke-how frailty progresses before and after stroke-remains underexplored.

AIMS: This study aims to examine longitudinal frailty index (FI) trajectories in individuals with and without stroke, and changes in frailty trajectories within individuals before and after a stroke event across four international longitudinal studies.

METHODS: This prospective cohort study analyzed data from four longitudinal cohorts: China Health and Retirement Longitudinal Study (CHARLS), English Longitudinal Study of Ageing (ELSA), Health and Retirement Study (HRS), and Survey of Health, Ageing and Retirement in Europe (SHARE). Frailty progression was assessed using validated FI scores. Incident strokes were identified through self-reported doctor diagnoses. Linear mixed models were used to evaluate changes in FI before and after stroke.

RESULTS: Among the 73,961 participants, 4374 (5.9%) incident stroke events were identified. Compared to stroke-free individuals, stroke survivors exhibited an observable increase in FI prior to the stroke event (e.g.

CHARLS: β = 0.016/year, 95% confidence interval (CI): 0.014-0.017). A sharp increase in FI occurred during the incident stroke event (HRS: β = 0.078, 95% CI: 0.074-0.083), followed by sustained post-stroke acceleration (ELSA: β = 0.019/year, 95% CI: 0.016-0.022). Sensitivity analyses confirmed robustness across cohorts.

CONCLUSION: Frailty accelerates significantly both before and after an incident stroke, suggesting a bidirectional relationship between stroke and frailty. Integrating frailty assessment into stroke risk stratification, rehabilitation, and secondary prevention to optimize patient outcomes, particularly in aging populations.Data access statement:The datasets generated and analyzed during the current study are available on the HRS website (https://hrs.isr.umich.edu/), CHARLS (https://charls.pku.edu.cn/en), SHARE (https://share-eric.eu/), and ELSA (https://www.elsa-project).