Duke Activity Status Index cut-off scores for assessing functional capacity after stroke

Purpose: To determine the Duke Activity Status Index (DASI) cut-off scores and verify if they would be able to discriminate between chronic stroke individuals, who had poor, from those who had good functional capacity (FC). Materials and methods: Cross-sectional study, where 92 individuals with chronic stroke had their FC measured by DASI scores and were classified into community and non-community walkers. Both the receiver operating characteristic and the area under the curve were used to determine the best DASI cut-off values that could discriminate between individuals, who had poor, from those who had good FC. Binary logistic regression analysis is used to investigate the ability of DASI scores in predicting community walkers. Results: DASI cut-off value of 31.95 showed to be sensitive and specific to differentiate between individuals, who had poor from good FC. The regression analysis revealed that the chance of individuals, who had better FC (>= 31.95) be community walkers was 10.69 (95%CI:4-32.75) times higher than that of those who had poor FC (< 31.95). The model demonstrated good predictive ability. Conclusions: The DASI cut-off value of 31.95 showed to be sensitive and specific to distinguish between chronic stroke individuals, who had poor, from those who had good FC.