Within-Assessment Variability of the GAITRite System in People With Stroke.

Abstract

The GAITRite is a validated tool for measuring gait. However, no information is available regarding the number of trials required. This study investigated the within-assessment variability of the GAITRite to establish the recommended number of trials. People with stroke undergoing inpatient rehabilitation were assessed on the GAITRite. Data of individuals admitted from January 2014 to January 2015 were extracted. Five trials were performed. A repeated-measures analysis of variance was used to investigate within-subject differences across the trials. Group means of each variable were plotted against trials. Significance was set at 0.05. Thirty-six people with stroke (mean [SD] age = 59.6 [13.3] yrs, 61.1% male) were assessed. Significant differences across trials were found for gait speed (P < 0.01), step length on the paretic (P < 0.01) and nonparetic sides (P < 0.01), stance phase time on the nonparetic side (P < 0.01), and single-limb support time of the paretic side (P = 0.02). The differences were found between the first two trials and the other trials. Depending on the gait variable of interest, the use of two warm-up trials and taking the average of two assessment trials could reduce the within-assessment variability of this instrument in people with stroke.