Psychometric evaluation and adaptability of the simplified Chinese version of HOOS-12 in patients undergoing total hip arthroplasty
摘要
Considering the response burden of the full version of the HOOS scale, researchers developed a more concise HOOS-12 scale. Although the HOOS-12 has been verified to have good reliability, validity and responsiveness in most Western-country studies, it still needs further evaluation in other non-English populations. Therefore, our aim is to assess the content validity, internal consistency, structural validity, test-retest reliability, responsiveness, floor effect and ceiling effect of the Chinese version of HOOS-12 in the patient population whose native language is Chinese. The goal is to provide a high-quality evaluation tool for the global Chinese-using hip joint patient population, promote cross-cultural and cross-regional hip joint research, and ensure the comparability of the data.
MethodPatients with chronic hip diseases scheduled for total hip arthroplasty were included in the study. Content validity was evaluated using expert review to assess item relevance. CFA was employed to validate the construct validity of the CHOO-12. Convergent and discriminant validity of the CHOOS-12 were assessed using the SF-36, OHS, and WOMAC scales as reference instruments. The first HOOS-12 scores were used to evaluate content validity, internal consistency, and structural validity. The second set of scores was utilized to assess test-retest reliability and ceiling/floor effects. The third set of scores was applied to evaluate responsiveness.
ResultThe CHOOS-12 demonstrated good content validity. No ceiling or floor effects were observed for the CHOOS-12 total scale or its pain, function, and quality of life subscales. The internal consistency among items was high for both the overall CHOOS-12 and all its subscales. CFA indicated satisfactory performance for the pain dimension, though the unidimensionality of the function and quality of life dimensions was challenged. The CHOOS-12 scale exhibited good convergent and discriminant validity. With overall ES and SRM values of 2.08 and 2.42 respectively, the CHOOS-12 also demonstrated good responsiveness.
ConclusionCHOOS-12 demonstrated good reliability and validity in this study. In terms of construct validity, the pain subscale performed well, although the unidimensionality of the PF and QOL subscales was challenged, the core items still had measurement potential. Future research needs to conduct further validation in larger and more diverse samples.