Validation of Chinese Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) in patients scheduled for total knee replacement

Objective Our aim was to cross-culturally validate Chinese Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) in patients with knee osteoarthritis (OA) scheduled for total knee replacement in Singapore. Methods Chinese WOMAC was translated from the original English version following standard guidelines. Patients were asked to complete a questionnaire containing the WOMAC (twice within 6 days), the Short Form 36 (SF-36), and the EuroQoL EQ-5D. Reliability was assessed using Cronbach's alpha and intraclass correlation coefficients (ICC), dimensionality using item-to-domain correlations and factor analyses. Convergent and discriminant construct validity was assessed using six each a priori hypotheses. Results Chinese WOMAC was well accepted by the patients in the pilot test. The Chinese and English versions were therefore administered to a consecutive sample of 131 Chinese- and 127 English-speaking subjects, respectively, with knee OA. Cronbach's alpha exceeded 0.7 for all domains except for Chinese pain domain, whereas the ICC exceeded 0.7 for all domains. Hypothesized item-to-domain correlations were observed for all items except for four items in Chinese physical function domain. The factor analyses yielded seven and five factors with eigenvalues of more than 1.0 in the Chinese and English versions, respectively. A total of 10/12 a priori construct validity hypotheses were satisfied for the Chinese version. Weak correlations between WOMAC pain and SF-36 bodily pain and moderate correlation between WOMAC pain and EQ-5D anxiety/depression were observed. Conclusions Chinese WOMAC was well accepted and demonstrated acceptable psychometric properties in Singaporean patients with severe knee OA.