To evaluate psychometric properties of SF-12 as a generic health-related quality of life (HQL) measure in a patient population with osteoarthritis (OA) in clinical trials. METHODS: Data were aggregated from three clinical trials, evaluating efficacy of different NSAIDs in OA patients (N=651). Patient assessments were made using SF-36 and seven commonly used clinical measures in OA (patient and physician global assessments, pain intensity, time to walk 50 feet, knee pain on weight bearing and motion, functional capacity classification), at baseline, week 2, and week 6. The SF-12 items were extracted from the SF-36 items. For the SF-12, item missing rate, computability of component scores, factor structure, item convergent / discriminant validity, item-component correlations, and floor and ceiling effects were evaluated. Correlations of SF-12 physical (PCS12) and mental component summary scores (MCS12) with SF-36 component summary scores (PCS36 and MCS36), and clinical variables were used to establish construct and convergent validity of the SF-12 in OA patients. RESULTS: A low individual SF-12 item missing rate (0.46% to 2.3%) and a high percentage score computability (91%) were observed at baseline. No floor or ceiling effects at baseline, week 2, and week 6 were observed. The scree plot confirmed two factor structure of the SF-12 items. Items belonging to the physical component correlated more strongly with the PCS12 than the MCS12, and vice versa. The correlations between PCS12 and PCS36, and MCS12 and MCS36 ranged from 0.94-0.96 (p<0.0005), at baseline, week 2, and week 6. Significant correlations of -0.18 to -0.53 (p<0.05) between SF-12 component summary scores and clinical variables, at baseline, week 2, and week 6 were observed. A similar trend was observed between SF-12 and clinical variable change scores at week 2, and week 6. CONCLUSION: The SF-12 is a psychometrically sound tool for the assessment of HQL in osteoarthritis patients.