Geographic variation and factors associated with home-based long-term care utilization across municipalities in Japan.

BACKGROUND

Japan's universal long-term care insurance (LTCI) system promotes aging in place through home-based services (HBS). However, geographic disparities in HBS utilization may persist despite universal coverage. This study examines the following research questions: (1) How does the utilization of HBS vary across Japan? (2) Do HBS utilization rates exhibit spatial clustering or random spatial patterns? (3) What factors are associated with the regional variation in HBS utilization?

METHODS

We conducted a cross-sectional study of 1,494 municipalities. Spatial autocorrelation analysis, including Global and Local Moran's I, was conducted to assess spatial clustering of HBS utilization. Non-spatial and spatial regression models were then estimated to examine regional factors associated with HBS utilization. Based on the Andersen-Newman Model of Health Services Utilization, these factors were categorized as need (older adults certified as having long-term care (LTC) needs), predisposing (sociodemographic characteristics), and enabling factors (municipality's system resources and LTC service supply). Stratified analyses were conducted to compare results between large urban, small urban, and rural municipalities.

RESULTS

HBS utilization varied widely across municipalities and showed significant spatial autocorrelation. Higher utilization clustered in western Japan, while lower utilization was observed in central and northern regions. The final spatial regression model explained 71.3% of the variance in HBS utilization. The need factor (certified as having LTC needs), predisposing factors (educational attainment), and enabling factors (financial strength of municipality, care manager availability, rurality, institutional care capacity, and non-institutional care provider density) were significantly associated with HBS utilization. Stratified analyses revealed that these associations varied by region type.

CONCLUSIONS

Despite universal LTCI coverage, substantial geographic heterogeneity persists in HBS utilization across Japan, associated with underlying municipal characteristics. This variation may pose a challenge to the goal of aging in place. Addressing it will require locally sensitive policy approaches that account for differences in care needs, socioeconomic context, and local LTC service availability across municipalities.