Reliability, validity, and sensitivity of the Chinese Short-Form 36 Health Survey version 2 (SF-36v2) in women with breast cancer.

RATIONALE, AIMS, AND OBJECTIVES: The psychometrics of the Short-Form 36 Health Survey version 2 (SF-36v2) in female breast cancer patients remains unknown. This study aimed to test the reliability, validity, and sensitivity of the Chinese SF-36v2 in women with breast cancer.

METHODS: The sample included 326 eligible participants. The reliability and the item convergent and discriminant validity were estimated using Cronbach α (≥0.70) and the multi-trait multi-item matrix analysis, respectively. The structural validity was tested using confirmatory factor analysis. Sensitivity was determined via an analysis of variance and the relative efficiency for initially diagnosed cases (yes vs no) as well as the time since diagnosis (years) before and after stratifying by initially diagnosed cases.

RESULTS: The overall Cronbach α was 0.91 (eight scales range: 0.72-0.92). All hypothesized item-scale correlations were greater than the alternatives (r ≥ 0.40). With acceptable model fit indices, the physical functioning, role-physical, bodily pain, and general health subscales had stronger contributions to the physical component summary (goodness-of-fit index [GFI]: 0.92, parsimony goodness-of-fit index [PGFI]: 0.60, comparative fit index [CFI]: 0.91, Tucker-Lewis index [TLI]: 0.93, adjusted goodness-of-fit index [AGFI]: 0.91, root mean square error of approximation [RMSEA]: 0.06, normed chi-squared [NC]: 2.65), while the vitality, social functioning, role-emotional, and mental health subscales contributed more to the mental component summary (GFI: 0.91, PGFI: 0.62, CFI: 0.91, TLI: 0.92, AGFI: 0.91, RMSEA: 0.07, NC: 2.76). The relative efficiencies with significant F-statistics were found for mental health (relative efficiency: 34.28; initially diagnosed cases), physical functioning (12.88; time since diagnosis), and physical functioning (5.80), role-physical (5.15), bodily pain (7.70), social functioning (4.62), role-emotional (4.72), mental health (4.75), and physical component summary (6.96; initially diagnosed cases with time since diagnosis; P < 0.05 for all variables).

CONCLUSIONS: Chinese SF-36v2 has acceptable psychometric properties and is suitable for women with breast cancer.