BACKGROUND: Patients with lower extremity osteoarthritis (OA) have increased all cause and disease specific mortality compared to the general population and epidemiological data suggests that OA management should include increased physical activity. AIM: To measure physical activity assessed as daily energy expenditure and accumulated steps day-1 in patients with severe OA of the hip or knee and compare with healthy controls. Furthermore, to investigate if gender, age, BMI and affected joint influence measures of physical activity. DESIGN: Cross sectional study. SETTING: Community. POPULATION: Fifty-one patients (49% women, age: 68±5 years, BMI: 28.4±4.6) with severe knee (N.=25) or hip (N.=26) OA scheduled for total joint replacement and 15 healthy population-based controls (53 % women, age: 68±5 years, BMI: 26.9±4.3). METHODS: Subjects wore an accelerometer (SenseWear™ Pro2 Armband) on 5 consecutive days. Time on body, total energy expenditure, average intensity (METS), time being sedentary, time above 3 METS and numbers of steps day(-1) were calculated. RESULTS: Patients had similar daily energy expenditure to controls (2632 vs. 2633 calories per day) but walked on average 29.3% fewer steps day(-1) (6632 vs. 8576 P=0.033). Gender and affected joint (hip or knee) did not influence the results. Age and BMI explained 15.4%-32.7% of the variation in total energy expenditure, average METS, time being sedentary, and steps day(-1). CONCLUSION: The present data indicate a need for interventions improving walking ability in patients with severe hip and knee OA. This need is independent regarding gender and affected joint but higher for older and heavier patients. Simultaneously assessed measures of physical activity gave different results, raising concern about validity of physical activity measures in patients with lower extremity OA and indicating caution when comparing results from studies applying different measures. CLINICAL REHABILITATION IMPACT: It is of potential interest for patients and practitioners that a relative high physical activity is possible for end-stage OA patients since this may reduce the increased all cause and disease specific mortality.