OBJECTIVE Assessing the effectiveness of bracing treatment for tibiofemoral osteoarthritis (OA) and patellofemoral OA in patients with knee OA. METHODS This study was conducted within the Osteoarthritis Chronic Care Program (OACCP), a 52-week multidisciplinary non-operative program for OA patients. All participants had symptomatic, radiographic knee OA. Knee bracing with Ossur Unloader One and Tru-pull Lite was offered for participants with medial/lateral tibiofemoral and patellofemoral OA, respectively. Participants were assessed at weeks 0, 12, 26 and 52. The primary outcome was knee pain and function at week 52, as assessed by the Knee injury and Osteoarthritis Outcome Score (KOOS) pain and activities of daily living (ADL) scores. Linear regression models were used to compare effectiveness for pain and function between three groups (patellofemoral bracing, tibiofemoral bracing and no bracing). RESULTS There were 204 participants; 50 assigned patellofemoral bracing, 86 tibiofemoral bracing and 68 with no bracing. Mean baseline KOOS pain score was 52.9, 41.7 and 43.3 (0-100 scale where 100 represents normal) and mean baseline KOOS ADL score was 55.8, 43.7 and 43.1 for the three groups, respectively. Significant improvements were found in each group at week 52 for KOOS pain score and KOOS ADL. There was no significant difference in KOOS pain (P = 0.12) and ADL score (P = 0.13) at week 52 between the three brace types after adjusting for baseline variables. CONCLUSIONS A multidisciplinary non-operative program improved pain and function in persons with patellofemoral and tibiofemoral OA. However, wearing a patellofemoral or a tibiofemoral brace did not appear to provide additional benefits.