The purpose of this study was to examine the effect of the angle of a wedged insole on knee varus torque during walking. Fifteen healthy subjects were recruited. Knee varus torque was measured using three-dimensional motion analysis (Elite). Knee varus torque was normalized to gait cycle (0%: initial contact; 100%: ipsilateral initial contact) and stance phase (0%: initial contact; 100%: ipsilateral toe off). The average peaks of knee varus torque during the stance phase of the gait cycle according to the different insole angles (10 or 15 degrees) were compared using one-way ANOVA with repeated measures. The results showed that in the early stance phase, the average peak knee varus torque increased significantly for both the medial 10 and 15 degree wedged insole conditions and decreased significantly for both the lateral 10 and 15 degree wedged insole conditions as compared with no insole (p<.05). However, there were no significant differences between the 10 and 15 degree wedged insole conditions with either the medial or lateral wedged insole (p>.05). In the late stance phase, the average peak knee varus torque increased significantly for the medial 10 and 15 degree wedged insole conditions (p<.05), but not for the lateral 10 and 15 degree wedged insole conditions as compared with no insole (p>.05). We suggest that these results may be beneficial for manufacturing foot orthotic devices, such as wedged insoles, to control medial and lateral compartment forces in the knee varus-valgus deformity. Further studies of the effects of wedged insole angle on knee varus torque in patients with medial-lateral knee osteoarthritis are needed.