RESUMO
<b>Objective:</b> In Japan, the Integrated Community Care System aims to support residents to live as independently as possible at home. <i>Koreisya-Kyoshitsu</i> and <i>Fureaiikiiki</i> salons are two types of group activities for community-dwelling elderly. We investigated effective ways of conducting such activities.<br><b>Methods:</b> We analyzed 96 subjects from 8 salons and 354 subjects from 10 <i>Koreisya-Kyoshitsu</i>. Self-completed questionnaires included the following: attributes, the Motor Fitness Scale (MFS), revised Philadelphia Geriatric Center Morale Scale (PGCMS), Measurement of Psychological Independence (MPI), instrumental activities of daily living (IADL), and self-rated health status (SRH). Follow-up assessment was conducted 6 months later. Representatives from 8 salons and staff members from 10 <i>Koreisya-Kyoshitsu</i> answered an additional questionnaire on management.<br><b>Results:</b> In <i>Koreisya-Kyoshitsu</i>, physical performance (MFS) (<i>p</i> = 0.007) and subjective well-being (PGCMS) (<i>p</i> = 0.001) improved significantly, whereas psychological independence (MPI) deteriorated significantly (<i>p</i> = 0.015). The MFS scores significantly improved in the sub-group with a high number of sessions (7 or more) (<i>p</i> = 0.043), as well as in the non-volunteer sub-group (<i>p</i> = 0.004). The PGCMS scores significantly improved in the sub-group with a high number of sessions (<i>p</i> < 0.001). The MPI scores significantly deteriorated in the sub-group with a low frequency of sessions (6 or less) and in the non-volunteer sub-group (<i>p</i> = 0.013 and <i>p</i> = 0.010, respectively). In salons, the frequency of going out decreased significantly (<i>p</i> = 0.049). Functional status (IADL) significantly improved in the “twice or more a month” sub-group (<i>p</i> = 0.046), whereas it significantly deteriorated in the “once a month” sub-group (<i>p</i> = 0.004). The proportion of volunteers/organizers in <i>Koreisya-Kyoshitsu</i> (23.4%) was significantly lower than that in salons (39.6%).<br><b>Conclusion:</b> The frequency (number) of sessions, but not the volunteer/non-volunteer attribute, was a key factor in obtaining the health promotion effects of group activities in both <i>Koreisya-Kyoshitsu</i> and salons.