Abstract
OBJECTIVE: To identify factors associated with earlier independence in "real-life walking" during hospitalization in subacute stroke patients. DESIGN: Retrospective cohort study. SUBJECTS/PATIENTS: Two hundred and six hemiplegic patients. METHODS: Functional Independence Measure (FIM) walking items were measured biweekly from admission to discharge. Patients were grouped by achieving independent "real-life walking" (FIM-walking score ≥6). Time to independence, stratified by age, FIM motor score (FIM-M), FIM cognitive score (FIM-C), and Functional Ambulation Categories (FAC) scores were compared using Kaplan- Meier plots and log-rank tests. Hazard ratios were calculated via multivariable Cox proportional hazard models. RESULTS: The median time to independence was 4 weeks, with significant differences (p < 0.05) by age, FIM-M, FIM-C, and FAC stratification. Age ≤64 years (hazard ratio 1.92, 95% confidence interval 1.21-3.06), FIM-C ≥25 (hazard ratio 2.42, 95% confidence interval 1.52-3.86), and FAC ≥3 (hazard ratio 1.98, 95% confidence interval 1.22-3.21) significantly affected earlier walking independence (all p < 0.01). Impeding factors were FIM-M ≤38 (hazard ratio 0.23, 95% confidence interval 0.13-0.40; p < 0.01) and FAC = 0 (hazard ratio 0.184, 95% confidence interval 0.06-0.62; p < 0.01). CONCLUSION: Early improvement in "real-life walking" was associated with younger age, greater cognitive function, and greater "test-setting walking" ability on admission. Low activities of daily living independence and "test-setting walking" ability hindered early progress.