UNLABELLED: van Dijk GM, Veenhof C, Spreeuwenberg P, Coene N, Burger BJ, van Schaardenburg D, van den Ende CH, Lankhorst GJ, Dekker J, on behalf of the CARPA Study Group. Prognosis of limitations in activities in osteoarthritis of the hip or knee: a 3-year cohort study. OBJECTIVE: To describe the course of limitations in activities in elderly patients with osteoarthritis (OA) of the hip or knee over a follow-up period of 3 years, and to identify prognostic factors of the course of limitations in activities, focusing on body functions, comorbidity, and cognitive functioning. DESIGN: A longitudinal cohort study with 3 years of follow-up. Measurements were conducted annually. Statistical analyses included t tests, univariate regression analyses, and multivariate regression analyses. SETTING: Rehabilitation centers and hospitals (Departments of Orthopedics, Rheumatology, and Rehabilitation) in The Netherlands. PARTICIPANTS: Patients (N=237) with hip or knee OA. INTERVENTIONS: Not applicable. MAIN OUTCOME MEASURES: Patient-perceived change, self-reported limitations in activities measured by the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC), and observed limitations in activities (timed walking test). Prognostic factors: demographic data, clinical data, body function (pain, muscle strength, range of motion [ROM]), comorbidity, and cognitive functioning (cognitive decline, memory, attention). RESULTS: Self-reported limitations in activities measured by the WOMAC improved slightly after 3-year follow-up. In knee OA, reduced ROM at 1-year follow-up (beta=.120), increased pain at 1-year follow-up (beta=-.177), and higher morbidity count (beta=-.180) predicted worsening of self-reported limitations in activities. In hip OA, reduced ROM at 1-year follow-up (beta=.201 for hip external rotation and beta=.144 for knee extension), increased pain at 1-year follow-up (beta=-.134), higher morbidity count (beta=-.220), or the presence of moderate to severe cardiac disease (beta=-.214) and poorer cognitive functioning (beta=.181) predicted worsening of self-reported limitations in activities. Performance-based limitations in activities measured by the timed walking test did not change after 3 years of follow-up. In knee OA, decreased muscle strength at 1-year follow-up (beta=-.272) and higher morbidity count (beta=.199) predicted worsening of performance-based limitations in activities. In hip OA, better ROM (beta=.182), higher morbidity count (beta=.232), or the presence of moderate to severe cardiac and eye-ear-nose-throat disease (beta=.210 and beta=.188, respectively) and older age (beta=.355) predicted worsening of performance-based limitations in activities. CONCLUSIONS: Overall, at the group level, limitations in activities of patients with OA of the hip or knee recruited from hospitals and rehabilitation centers seem fairly stable during the first 3 years of follow-up. However, at the level of individual patients, considerable variation occurs. Prognostic factors for worsening of limitations in activities include increased pain, reduced ROM, and decreased muscle strength at 1-year follow-up; higher morbidity count; and to a lesser extent poor cognitive functioning.