Evaluation of intrarater and interrater reliability of the Wisconsin Gait Scale with using the video taped stroke patients in a Turkish sample.

Objective: To establish the intrarater and interrater reliability of Wisconsin Gait Scale (WGS) in hemiplegic patients.

Design: Repeated-measures reliability study using video data of stroke patients.

Setting: Rehabilitation department of the university hospital.

Participitants: Nineteen hemiplegic patients with 3-9 months stroke history and two physiatrists and two physical therapists.

Interventions: Video recordings were assessed twice, at an interval of 2 days, by the two physiatrists and two physical therapists.

Main Outcome Measure: Wisconsin Gait Scale.

Results: Internal consistency coefficients for the WGS were excellent; Cronbach scores were 0.91 and 0.94 for the first and third days. Coefficient of Repeatability (CR) for observers' WGS assessments were ranged between 4.23-5.76 and intraclass correlation coefficients for total WGS score were indicated very high interrater reliability at the begining and end, respectively 0.91 and 0.96. Intraclass correlation coefficients for fourteen items of WGS ranged from 0.81 to 1. "Hip hiking at mid-swing", "Circumduction at mid-swing" and "Hip extension of the affected leg" were the items with lowest correlation coefficients. Intrarater reliability for total WGS scores ranged from 0.75 to 0.90.

Conclusion: WGS was found excellent in reliability and may provide an objective means to document the findings from observational gait analysis, which is frequently used in clinical practice by rehabilitation teams.