OBJECTIVE: To estimate minimal clinically important differences (MCID) of effects measured by the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) in patients with osteoarthritis (OA) of the lower extremities undergoing a comprehensive inpatient rehabilitation intervention. METHODS: A prospective cohort study assessed patients’ health by the WOMAC at baseline (entry into the clinic) and at the 3 month followup, and by a transition questionnaire asking about the change of “health in general related to the OA joint” during that time period. The WOMAC section score differences between the “equal” group and the “slightly better” and “slightly worse” groups resulted in the MCID for improvement and for worsening. RESULTS: In total 192 patients were followed up. The MCID for improvement ranged from 0.80 to 1.01 points on the continuous WOMAC numerical rating scale from 0 to 10, reflecting changes of 17 to 22% of baseline scores. The MCID for worsening conditions ranged from 0.29 (6%) to 1.03 points (22%). In the transition reply subjectively unchanged patients reported a “pessimistic bias” of 0.35 to 0.51 points, except for the stiffness section. Both MCID and pessimistic bias showed regression to the mean and baseline dependency. CONCLUSION: The assessment of MCID using the transition method is a heuristic and valid strategy to detect particular rehabilitation effects in patients with OA of the lower extremities with the use of the WOMAC, and it is worth implementing. The size of the MCID and of the systematic bias is comparable to that assessed by other methods and in other therapeutic settings.