Examination of the relationship between participation in salons aimed at care prevention through the promotion of social participation and the subsequent cost of care: A 3-year prospective follow-up study in JAGES.

Many previous studies have found that social participation improves the health and functional maintenance of older people. However, to determine whether promoting social participation can prevent functional decline in the elderly, it is necessary not only to compare the prognosis of those who participate in social activities to those who do not but also to demonstrate that the intervention was effective in promoting social participation. Although the effect of social participation in preventing caregiving has been demonstrated, the key question is whether preventing functional decline through social participation can reduce care costs. This study aims to examine the relationship between participation in salons aimed at care prevention through the promotion of social participation and the subsequent cost of care. We use the Japan Gerontological Evaluation Study (JAGES) dataset to conduct longitudinal, individuallevel analysis. We focus on all residents 65 years and older who did not have a without disability in Taketoyo town and ran three regression analyses. First, a generalized linear model (GLM) with Tweedie distribution and log-link function, as well as robust estimation of variance components was used to estimate the dependent variables. Second, we used an inverse probability weighting (IPW) model to minimize selection bias. Finally, we performed the IV analysis. In this study, the GLM with IPW and IV models revealed link between salon participants and lower caregiving costs. The link between participation and caregiving costs was confirmed in a model with reduced selection bias, rather than in a simple GLM model.