Abstract Background : Studies are now emphasizing on the involvement of risk factors like diabetes mellitus (DM) in knee osteoarthritis (OA). Patients with knee OA and those with DM have lower physical activity. However, whether patients with knee OA and DM have more decreased physical activity than do patients with only knee OA and whether DM is associated with physical activity in patients with moderate-to-severe knee OA who are scheduled for total knee arthroplasty (TKA) are unclear. The purpose of this study was to investigate differences in physical activity between patients with and without DM and to determine the association between physical activity and the presence of DM in patients with moderate-to-severe knee OA who are scheduled for TKA. Methods : We evaluated 183 patients with moderate-to-severe knee OA who were scheduled for TKA. We evaluated knee function (i.e., knee flexion/extension range-of-motion, knee-extension strength, knee pain) and performance-based physical function (e.g., Timed Up-and-Go [TUG] test), and objectively measured physical activity using an accelerometer. Physical activity by intensity was compared between knee OA patients with DM and those without DM. The association between physical activity and DM, including knee function and performance-based physical function, was evaluated. Results : The average daily step count was significantly fewer (p<0.001), and light-intensity physical activity (LPA) time (p<0.001) and moderate-to-vigorous-intensity physical activity (MVPA) time (p=0.006) were significantly shorter in knee OA patients with DM than in those without DM. After adjusting for age, sex, and body mass index, fewer average daily step count and shorter LPA time were significantly correlated with DM ( β = -0.200, p=0.006; β = -0.216, p=0.004) and a longer TUG test time ( β = -0.196, p=0.014; β = -0.208, p=0.011). Shorter MVPA time was significantly correlated with lower contralateral knee-extension strength ( β = 0.187, p=0.032). Conclusions : Knee OA patients with DM had lower physical activity than did knee OA patients without DM. The presence of DM was correlated with step count and LPA time in patients with moderate-to-severe knee OA scheduled for TKA. Our results suggested that the progression of decreased physical activity should be a concern for patients with both moderate-to-severe knee OA and DM.