Abstract
BACKGROUND: Balance assessment is crucial for predicting community ambulation outcomes in subacute stroke patients undergoing rehabilitation. This study aims to compare the accuracy of the Postural Assessment Scale for Stroke Patients (PASS) and the Berg Balance Scale (BBS) in predicting community walking ability at discharge from rehabilitation. METHODS: This prospective cohort study included 47 stroke patients admitted to a 4-week inpatient rehabilitation program. Patients were assessed with PASS and BBS at admission. Discharge assessments included the Functional Ambulation Categories and 6-Min Walk Distance tests. Statistical analysis involved calculating the area under the receiver operating characteristic curve (AUC), sensitivity, specificity, likelihood ratios, and cut-off scores. RESULTS: PASS and BBS demonstrated excellent predictive accuracy, with AUC values of 0.955 (95% CI [0.850-0.994]) for PASS and 0.991 (95% CI [0.906-1.000]) for BBS. Cut-off scores were >28 for PASS and >46 for BBS. Sensitivity was high for both (94.44%, 95% CI [72.7-99.9]), while BBS had superior specificity (96.43%, 95% CI [81.7-99.9]) compared to PASS (85.71%, 95% CI [67.3-96.0]). BBS also had a higher positive likelihood ratio (26.44 vs. 6.61). The difference in AUC values was non-significant (p = 0.093). CONCLUSIONS: PASS and BBS assessed at admission are highly accurate tools for predicting community ambulation at discharge in subacute stroke patients, with BBS demonstrating a slight advantage, particularly in its positive predictive value. These findings support the use of both scales to guide rehabilitative clinical decision-making.