The revised index for social engagement (RISE) in long-term care facilities: reliability and validity in Japan.

Background This study aimed to establish the validity and reliability of a revised index for social engagement (RISE) in the Japanese context. Methods We analysed the data of 1377 participants over 65 years of age who had been admitted to two types of long-term care facilities (LTCF) in Japan: four health facilities for older adults and eight nursing homes. Resident level data based on the Japanese version of the interRAI assessment instrument were collected from 623 residents in the former and 754 in the latter. From these data, we calculated RISE by adding six dichotomous items on social engagement in the assessment form. Factorial validity was evaluated by exploratory factorial analysis and confirmatory factor analysis, convergent validity by the correlation between average activity time and the RISE score, and discriminant validity by the correlation between cognitive levels and the RISE scores. Lastly, we assessed internal consistency using Cronbach's alpha. Results We identified a two-factor model in the exploratory factorial analysis with a factor loading >0.40, except for one RISE item. The confirmatory factor analysis confirmed that the two-factor model had appropriate model fits. The correlation between time involved in activities and the RISE score was r = 0.45, while the correlation between cognitive function and the RISE score was r = -0.32. The convergent and discriminant validities supported the use of Japanese LTCF. Cronbach's alpha ranged 0.70-0.72. Conclusions Although further revision may be needed to improve factorial validity, RISE is reliable and valid for assessing social engagement of older adults admitted to LTCF in Japan. By using the Japanese version of RISE, the positive aspects of social functioning can be appropriately assessed and provide more evidence for improving the quality of care in LTCF.