In order to reduce pain caused by the affected hip joint, unilateral hip osteoarthritis patients (HOAP) adopt characteristic gait patterns. However, it is unknown if the knee and hip joint loading in the non-affected (limbnon-affected ) and the affected (limbaffected ) limb differ from healthy controls (HC) and which gait parameters correlate with potential abnormal joint loading. Instrumented 3D-gait analysis was performed on 18 HOAP and 18 sex, age, and height matched HC. The limbnon-affected showed greater first and second peak external hip adduction moments (first HAM: +15%, p = 0.014; second HAM: +15%, p = 0.021, respectively), than seen in HC. In contrast, the second peak external knee adduction moment (KAM) in the limbaffected is reduced by about 23% and 30% compared to the limbnon-affected and HC, respectively. Furthermore, our patients showed characteristic gait compensation strategies including reduced peak vertical forces (pvF), a greater foot progression angle (FPA), and reduced knee range of motion (ROM) in the limbaffected . The limbaffected was 5.6 ± 3.8 mm shorter than the limbnon-affected . Results of stepwise regression analyses showed that increased first pvF explain 16% of first HAM alterations, whereas knee ROM and FPA explain 39% of second KAM alterations. We therefore expect an increased rate of progression of OA in the hip joint of the limbnon-affected and suggest that the shift in the medial-to-lateral knee joint load distribution may impact the rate of progression of OA in the limbaffected . The level of evidence is III. © 2016 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 35:1764-1773, 2017.