Aim: Building community social capital is important for socially inclusive societies, leading to dementia-friendly communities. However, quantitative evidence is currently lacking regarding the effectiveness of dementia-friendly communities, including the association with community social capital. The current study examined the ecological relationship between social capital and the proportion of people with cognitive decline.Methods: Data were obtained from population-based, cross-sectional surveys (mail-in questionnaire survey and home-interview survey) of community-dwelling older adults aged ≥65 years, living in 17 district areas in Machida City, Tokyo, Japan. Social capital included social networks (neighborly ties) and social support (emotional support and instrumental support). Cognitive function was assessed using the Mini-Mental State Examination, Japanese version (with a cut-off of 23/24). The proportions of people with high social capital and cognitive decline (Mini-Mental State Examination, Japanese version ≤23) were calculated for residential districts and used in the analysis.
Results:The district-level social capital indicators were positively and moderately correlated with the proportion of people with cognitive decline. After adjusting for the aging rate, educational level and population density of the district, an ordinal logistic regression analysis showed that higher proportions of people with strong neighborly ties and people with strong instrumental support were significantly associated with a higher proportion of people with cognitive decline. This trend was stronger among women than men.Conclusions: People can continue to live in communities with high social capital, even if they are experiencing cognitive decline. Although this study was preliminary, it provided empirical evidence for the benefits of promoting dementia-friendly communities. Geriatr Gerontol Int 2019; 19: 950-955.The odds ratios (OR) associated with an increase in a single unit of quartile (Q1 = 1; Q4 = 4) in each independent variable are shown. Each of the three social capital indicators were separately added into the models. Demographic factors were used in each model as covariates. CI, confidence interval; MMSE-J, Mini-Mental State Examination, Japanese version.