Researcher/local health professionals’ collaborations for community health promotion and mortality among older adults in Japan: JAGES community intervention study

Abstract

Introduction Health promoting interventions for long-term care prevention among older adults should target community social environment. However, in local government, many health sector staffs face challenges in essential activities including community diagnosis (risk and resource assessments) and intersectoral partnerships. Supporting municipality staffs for those actions, under the close researcher/municipality collaborations, may benefit them. However, the effectiveness of such support/collaborations has remained unclear. We evaluated that in terms of older residents’ community social participation and mortality risks. Methods In 2014, the half of the municipalities participating in the Japan Gerontological Evaluation Study (JAGES) received active supports from JAGES researchers for utilizing community diagnosis data and intersectoral collaborations with various organizations potentially contributing to health promotion for older adults. Participants were functionally independent community-dwelling older adults aged 65 years or more. First, we analyzed 107,292 older people in 25 municipalities participated in both 2013 and 2016 survey. Among them, 13 municipalities received active supports and community diagnosis data based on JAGES surveys (intervention) and remaining 12 municipalities obtained community diagnosis data only (control). We performed difference-in-difference (DID) analysis using multilevel Poisson regression to compare the changes in the prevalence of social participation among the older residents of the intervention and control group municipalities. Second, using the follow-up data (maximum 1267 days), we performed survival analyses applying Fine & Gray's proportional hazard model for assessing the relative risk for death, considering competing risks of moving out. In both analyses, to control selection bias we calculated propensity of receiving active supports from demographic data and performed inverse probability of treatment weighting (IPTW) analysis. We stratified all analyses by gender and adjusted for age, education, income, living alone or not, marital status, comorbidity, depressive symptoms, Instrumental Activities of Daily Livings (IADL), and considered the clusters of residential areas in 2013. Results In total, 4591 people have died for 235,534 person-year observation period. Among men, the estimated local activity participation was 46.6% (95% confidence interval [CI]: 45.5%, 47.7%) in 2013 and 57.1% (95% CI: 56.0%, 58.1%) in 2016, among residents in intervention group. In contrast, the participations were 46.4% (95% CI: 45.4%, 47.5%) in 2013 and 54.6% (95% CI: 53.0%, 55.3%) in 2016, among residents in the control municipalities (DID = 0.028, P = 0.006). Among women, there was no significant difference between the two groups (P = 0.131). The adjusted hazard ratio for death among those who lived in the supported municipality was 0.90 (95% CI: 0.84, 0.96) in men and 0.99 (95% CI: 0.90, 1.09) in women, compared to the control group. Conclusion Supporting health sector staffs in municipalities were associated with improved health risks among men. Better partnerships with various organizations, based on strategic risk and resource assessments, might lead to enrich community environment that provides more opportunities for social participation for older men. Building similar collaborating framework may be beneficial for advancing health promotion in local settings.