ABSTRACT
INTRODUCTION: Oral frailty leads to poor nutritional status, which, in turn, leads to frailty. This cross-sectional study aimed to determine regional differences in the prevalence of oral frailty and to identify factors associated with oral frailty using 3-level multilevel models. METHODS: This study comprised 165,164 participants aged ≥65 y without long-term care requirements in the Japan Gerontological Evaluation Study. The dependent variable was oral frailty, which was calculated based on age, number of teeth, difficulty in eating tough foods, and choking. The individual-level independent variables included sociodemographics, present illness, social participation, frequency of meeting friends, and social capital. The local district-level independent variable was social capital (n = 1,008) derived from exploratory factor analyses. The municipality-level independent variable was population density (n = 62). Three-level multilevel Poisson regression analysis was performed to calculate the prevalence ratios (PRs). RESULTS: The prevalence of oral frailty in municipalities ranged from 39.9% to 77.6%. Regarding district-level factors, higher civic participation was significantly associated with a lower probability of oral frailty. At the municipality level, the PR of the rural-agricultural area was 1.17 (95% confidence interval, 1.11-1.23) (reference: metropolitan). CONCLUSION: These results highlight the usefulness of oral frailty prevention measures in encouraging social participation in rural areas. KNOWLEDGE TRANSFER STATEMENT: The results of the present study showed regional differences in oral frailty. In particular, rural-agricultural areas show higher prevalence rates of oral frailty than those in metropolitan cities. Promoting measures of social participation among older adults may help prevent oral frailty in rural areas.
