Abstract
BACKGROUND: Hand osteoarthritis (OA) has a wide spectrum of clinical presentations and physical function is one of the core domains where patients suffer. The Functional Index for Hand Osteoarthritis (FIHOA) is a leading assessment tool for hand OA-related functional impairment. Our objective was to make a Japanese version of FIHOA (J-FIHOA) and validate it among Japanese hand OA patients. METHODS: Forward and backward translation processes were completed to create a culturally adapted J-FIHOA. A prospective, observational multicenter study was undertaken for the validation process. Seventeen collaborating hospitals recruited Japanese hand OA patients who met the American College of Rheumatology criteria. A medical record review and responses to the following patient-rated questionnaires were collected: J-FIHOA, Hand20, Health Assessment Questionnaire (HAQ), numerical rating scale for pain (NRS pain) and Short Form 36 Health Survey (SF-36). We explored the structure of J-FIHOA using factor analysis. Cronbach's alpha coefficients and item-total correlations were calculated. Correlations between J-FIHOA and other questionnaires were evaluated for construct validity. Participants in clinically stable conditions repeated J-FIHOA at a one- to two-week interval to assess test-retest reliability. To evaluate responsiveness, symptomatic patients who started new pharmacological treatments had a 1-month follow-up visit and completed the questionnaires twice. Effect size (ES) and standardized response mean (SRM) were calculated with pre- and post-treatment data sets. We assessed responsiveness, comparing ES and SRM of J-FIHOA with other questionnaires (construct approach). RESULTS: A total of 210 patients participated. J-FIHOA had unidimensional structure. Cronbach's alphas (0.914 among females and 0.929 among males) and item-total correlations (range, 0.508 to 0.881) revealed high internal consistency. Hand20, which measures upper extremity disability, was strongly correlated with J-FIHOA (r = 0.82) while the mental and role-social components of SF-36 showed no correlations (r = - 0.24 and - 0.26, respectively). Intraclass correlation coefficient for test-retest reliability was 0.83 and satisfactory. J-FIHOA showed the highest ES and SRM (- 0.68 and - 0.62, respectively) among all questionnaires, except for NRS pain. CONCLUSIONS: Our results showed J-FIHOA had good measurement properties to assess physical function in Japanese hand OA patients both for ambulatory follow-up in clinical practice, and clinical research and therapeutic trials.