Fall history and risk of incident stroke in middle-aged and older adults with possible sarcopenia: A longitudinal cohort study based on CHARLS

Abstract

Background

Possible sarcopenia (PS) was defined per AWGS 2019 criteria as reduced muscle strength or physical performance without low muscle mass. PS and falls are prevalent in aging populations, yet their combined impact on long-term stroke risk remains underexplored, particularly in Asian cohorts. This study investigates the association between fall history and incident stroke among middle-aged and older Chinese adults with PS.

Methods

Using longitudinal data from the China Health and Retirement Longitudinal Study (CHARLS, 2015–2018), we analyzed 4605 participants with PS (mean age: 64.8 years). Fall history was assessed by self-reported response to “Have you fallen in the past two years?” Stroke was physician-diagnosed via self-report. Multivariable-adjusted logistic regression models evaluated stroke risk, adjusting for demographics, comorbidities (hypertension, diabetes), and biomarkers (low-density lipoprotein cholesterol [LDL-C], high-sensitivity C-reactive protein [hs-CRP]). Subgroup and sensitivity analyses assessed robustness.

Results

Stroke incidence was 9.3 % among fallers and 6.7 % among non-fallers, with a P-value of 0.036. Participants with fall history exhibited a 37 % higher risk of incident stroke compared to non-fallers (adjusted OR: 1.37, 95 % CI: 1.02–1.84). The association remained consistent across age, sex, BMI, and most comorbidity subgroups, except for dyslipidemia (interaction P = 0.022). Sensitivity analyses, including multiple imputation for missing data, exclusion of adults >80 years, and removal of hip fracture cases, consistently confirmed the robustness of the findings.

Conclusion

Fall history independently predicts stroke risk in individuals with PS, highlighting the potential of fall prevention to reduce stroke burden in this population. This study provides novel evidence of a prolonged stroke risk trajectory in Asian PS cohorts, complementing Western studies focused on short-term post-fall risks. The study has several limitations. Self-reported falls may introduce recall bias and lack mechanistic details. Unmeasured confounders such as diet and medication use constrain causal interpretation. Data constraints prevented differentiation between stroke subtypes.