Interrater and intrarater reliability and minimal detectable change of the Wisconsin Gait Scale when used to examine videotaped gait in individuals post-stroke

Abstract

Background: Often, interventions targeting the kinematic and temporal and spatial changes in gait commonly seen after a stroke are based on observations of walking. Having the capacity to objectively identify such changes and track improvements over time using reliable and valid measures is important. The Wisconsin Gait Scale (WGS), which is comprised of 14 items, was developed specifically to examine and document gait changes occurring after a stroke. The purpose of the study was to explore the interrater and intrarater reliability and minimal detectable change (MDC) of the WGS when used by physical therapists to examine gait in adults post-stroke. Methods: Fourteen physical therapists from 3 different acute inpatient rehabilitation centers rated videotapes of the gait of 6 adults post-stroke using the WGS. To minimize subject variability from fatigue, videotapes created by using 4 cameras provided right and left lateral, anterior, and posterior views of walking on a level surface. One complete ambulation trial from each subject post-stroke, which included 4 views of the same ambulation trial, was examined by the licensed physical therapists using the WGS. An opportunity was provided to review the tool and a practice trial was performed using an additional videotape not included in the analysis. Gait was examined on 2 different occasions separated by a period of approximately 21 days to minimize the effects of recall bias. Intraclass Correlation Coefficients (ICC) were used to examine the interrater and intrarater reliability of the WGS. Results: Interrater (ICC = 0.83) and intrarater (ICC = 0.91) reliability were both good. The standard error of the measurement (SEM) was 1.47 and the MDC95 was 4.24. There was no statistically significant difference between the scores on the WGS when comparing the 2 different sessions. Conclusions: The WGS shows promise as an instrument that can make observational gait analysis more reliable. High intrarater reliability and low SEM suggests that the WGS is stable when administered across multiple sessions by the same rater. The ICC for interrater reliability was also good, which suggests that multiple examiners can effectively use the instrument. With minimal training, the physical therapists in the study were able to produce highly reliable results using the WGS to objectively document gait dysfunction.