Given the increasing interest in treatment satisfaction research and the lack of a specific questionnaire in osteoarthritis (OA), we developed and explored the psychometric properties of the osteoARthritis Treatment Satisfaction (ARTS) questionnaire. The ARTS questionnaire which consists of 18 items was developed in French following the analysis of semi-structured interviews performed among 20 OA participants, five rheumatologists and five general practitioners. Psychometric properties were assessed in France on a cross-sectional sample of 797 OA participants and test-retest reliability was evaluated in an independent sample of 111 clinically stable OA participants who filled-in the questionnaire within a 7.7 (+/- 3.1) day interval. Using principal component analysis, four scales were identified: Treatment advantages (seven items), Treatment convenience (three items), Treatment confidence (two items) and Satisfaction with physician (six items). Item convergent and item discriminant validity were satisfactory. Internal consistency provided evidence of reliability and lack of redundancy (Cronbach's alphas ranged from 0.66 to 0.86). Test-retest reliability was acceptable for two out of four scales (intraclass correlations coefficients (ICC) ranged from 0.61 to 0.75). Significant between groups differences were found on the ARTS scales, demonstrating the known groups validity of the ARTS questionnaire. The responsiveness of the ARTS is still to be documented.