Analysis of influencing factors and paths of social frailty in older adult patients with ischemic stroke: a cross-sectional study.

Objective: Analysis of influencing factors and action pathways of social frailty in older adult patients with ischemic stroke. Methods: A cross-sectional study was conducted among older adult inpatients with ischemic stroke in the Department of Neurology of two Grade A tertiary hospitals in Sichuan Province, China, from January to May 2025. Multiple stepwise linear regression analysis was used to analyze the influencing factors of social frailty in older adult patients with ischemic stroke, and structural equation modeling was employed to conduct path analysis on these influencing factors. Results: A total of 437 older adult patients with ischemic stroke completed the study. The prevalence of social frailty in older adult ischemic stroke patients was 38%. Regression analysis showed that activities of daily living, sleep quality, depression, social support, employee medical insurance, language function impairment and stroke recurrence are influencing factors of social frailty levels in older adult patients with ischemic stroke (p < 0.05). The structural equation model revealed that in the biological dimension, the direct effect of ADL on social frailty was significant (β = -0.154, p = 0.002), accounting for 16% of the total direct effects, while its indirect effect through social support was also significant (β = -0.061, p = 0.003, 95% CI: -0.114 to -0.019), representing 28% of the total effect in the ADL-social frailty pathway; the direct effect of sleep quality on social frailty (β = 0.057, p = 0.282) and the indirect effect via social support (β = 0.014, p = 0.546, 95% CI: -0.032 to 0.062) were not significant. In the psychological dimension, depression had a significant direct effect on social frailty (β = 0.390, p < 0.001), accounting for 40% of the total direct effects, along with a significant indirect effect via social support (β = 0.113, p < 0.001, 95% CI: 0.063-0.181), representing 22% of the total effect in the depression-social frailty pathway. In the social dimension, social support showed a significant direct effect on social frailty (β = -0.373, p < 0.001), representing 38% of the total direct effects. Conclusion: This study found that the prevalence of social frailty in older adult patients with ischemic stroke was 38%, and it is closely associated with health insurance type, stroke recurrence, language function, activities of daily living, depression, and social support. These findings provide a reference for clinical practitioners to design targeted interventions.