Chinese cross-cultural adaptation and validation of the Oxford shoulder score

BACKGROUND: The Oxford Shoulder Score (OSS) is a reliable and valid construct measuring non-specific shoulder pain, which are widely used to evaluate shoulder related quality of life. This study was to cross-culturally adapt and psychometrically validate a simplified Chinese version of the OSS (SC-OSS). METHODS: Cross-cultural adaptation was performed according to the international recognized guidelines. Consecutive patients with nonspecific shoulder pain were recruited to test the psychometric properties of SC-OSS. Item response trend and item-total correlation were evaluated to measure homogeneity. Principal component analysis (PCA) was used to evaluate the factorial structure. Cronbach's α and intra-class correlations were used to determine the reliability. Construct validity was analyzed by evaluating the correlations between SC-OSS and the Constant-Murley shoulder outcome score (CMSOS), the short form (36) health survey (SF-36) containing eight domains, and pain visual analogue scale (VAS). RESULTS: Overall, 121 patients were recruited. Each of the 12 items was properly responded and correlated with the total items. PCA extracted one factor for SC-OSS. SC-OSS had excellent reliability, with a Cronbach's α of 0.92 and intra-class correlation coefficient of 0.97 (95 % CI: 0.94-0.98). SC-OSS had a high correlation with CMSOS, physical functioning (PF) and bodily pain (BP) domains of SF-36 and VAS (r = -0.70, -0.65, -0.53, and -0.66, respectively). SC-OSS moderately correlated with role-physical (RP), social functioning (SF), general health perception (GH) and vitality (VT) (r = -0.45, -0.42, -0.39 and -0.36, respectively), but had a low correlation with role-emotional (RE) and mental health (MH) domains of SF-36 (r = -0.28 and -0.23, respectively). CONCLUSIONS: SC-OSS demonstrated excellent acceptability, internal consistency, reliability and construct validity, which can be recommended for application in mainland China.