Gait velocity and walking distance to predict community walking after stroke.

Gait speed and walking distance were evaluated as predictors for levels of community walking after stroke. In this study, 103 stroke survivors were identified as limited (n = 67) or independent community walkers (n = 36). Ten meter and six min walk tests were used to measure gait speed and walking distance, respectively. The discriminative properties of gait speed and walking distance for community walking were investigated using receiver operating characteristic curves. Cut-off values of 0.87 m/s for community walking gait speed for walking distance had positive predictive values of 65% and 55%, respectively. The negative predictive value ranged from 89% for gait speed to 79% for walking distance. Gait speed and walking distance showed significant differences between limited and independent community walking. Gait speed was more significantly related to community walking than walking distance. The results of this study suggest that gait speed is a better predictor for community walking than walking distance in moderately affected post-stroke survivors.