Reliability and validity of the Japanese version of the INSPIRE measure of staff support for personal recovery in community mental health service users in Japan

Abstract

Background: Supporting personal recovery in people with mental health difficulties is central to mental health services. This study aimed to develop the Japanese version of INSPIRE and Brief INSPIRE measure of staff support for personal recovery and to evaluate its reliability and validity. Methods: A questionnaire survey was conducted from October to December 2015. The authors asked users to participate in the survey of 14 community mental health services in the Kanto region of Japan. The service users completed self-administered questionnaires that include the Japanese version of INSPIRE, the Recovery Assessment Scale, the Client Satisfaction Questionnaire, the patient version of the Scale to Assess Therapeutic Relationship in Community Mental Health care and the Short Form Health Survey. Internal consistency was assessed using Cronbach's alpha coefficient, and test-retest reliability was assessed using the intraclass correlation coefficient (ICC) and weighted kappa. Convergent validity was examined by assessing correlation with other scales. Factor validity was evaluated by exploratory factor analysis (EFA) with generalized least-squares mean and oblimin rotation. In addition, confirmatory factor analysis was used to check the fitness of the factor structure models derived from the EFA. Results: A total of 195 out of 212 users gave written informed consent and participated in the study. Data from 190 respondents were analyzed (response rate 89.6%). INSPIRE, Brief INSPIRE, and the subscales all showed Cronbach's alpha coefficient over 0.78. ICC and weighted kappa derived more than 0.92 for subscales and Brief INSPIRE. These numerical values indicated good reliability. The convergent validity of Brief INSPIRE and the subscales was significantly positively correlated with the other scales. Different from the previous study, the factor structure was extracted using EFA. Both factor structures were checked by CFA, but the degree of fitness index was not good in either. Therefore, the factor analysis did not show goodness of fit. Conclusions: This study found the Japanese version of INSPIRE and Brief INSPIRE to be reliable and valid for use among community mental health service users in Japan.