The aim of this study was to compare performance error and perceived difficulty during toe-out gait modification in individuals with knee osteoarthritis (OA) across three different types of visual feedback: mirror, raw video, and real-time biofeedback of toe-out angle. Twenty individuals with knee OA (11 female; age = 65.4 +/- 9.8 years) took part in this study at a university motion analysis laboratory. Seven participants had mild knee OA, nine had moderate knee OA, and four had severe knee OA. Participants were trained to walk on a treadmill while matching a target indicating a ten degree increase in stance phase toe-out compared to toe-out angle measured during self-selected walking. The target was provided visually by way of the three types of feedback listed above and were presented in a random order. Kinematic data were collected and used to calculate the difference between the target angle and actual performed angle for each condition (toe-out performance error). Difficulty was assessed using a numerical rating scale (0-10) provided verbally by participants. Toe-out performance error was significantly lower when using the real-time biofeedback compared to the other two methods (p = 0.025; mean difference vs. mirror = 2.05°; mean difference vs. raw video = 1.51°). Perceived difficulty was not statistically different between the groups (p = 0.51).
Though statistically significant, the 2 degree differences in toe-out gait performance error may not mean that the large economic and personnel costs of real-time biofeedback as a means to modify movement in clinical or research settings are necessary.