Reliability and Validity of the Japanese Version of the Michigan Hand Outcomes Questionnaire: A Comparison with the DASH and SF-36 Questionnaires

Background: The Michigan Hand Outcomes Questionnaire (MHQ) has shown reliability, validity and responsiveness and has been used to assess surgical outcomes mainly in North America. We established a Japanese version of the MHQ and evaluated its reliability and validity compared with both the short-form 36 (SF-36) questionnaire and the Disabilities of the Arm, Shoulder and Hand (DASH) questionnaire in a Japanese-speaking population. Methods: The MHQ was cross-culturally adapted to a Japanese version according to guidelines. Sixty-eight patients with hand conditions were enrolled in this study and answered the MHQ, DASH questionnaire and SF-36 questionnaire. The MHQ was completed again with an interval of one or two weeks. Reproducibility and internal consistency were statistically assessed by the test-retest method and calculating Cronbach's alpha. Spearman's rank correlation was calculated to assess associations between the MHQ and the SF-36 questionnaire as well as the DASH questionnaire. Results: The intraclass correlation coefficients of MHQ subscales ranged from 0.68 to 0.93. Aesthetics subscale of the left hand showed the lowest intraclass correlation but still a good correlation. Cronbach's alpha values of the MHQ ranged from 0.81 to 0.96 for all subscales. The absolute values of Spearman's rank correlation coefficient between MHQ subscales and DASH function/symptoms scores ranged from 0.49 to 0.82. Spearman's correlation coefficients of the MHQ total score to subscales of the SF-36 questionnaire ranged from 0.42 to 0.68. The strongest correlations were found between work performance of the MHQ and role-physical of the SF-36 questionnaire. Conclusions: The Japanese version of the MHQ has adequate instrument properties for assessing hand outcomes compared with the SF-36 questionnaire as well as the DASH questionnaire.