Psychiatric standardized claim ratios by rurality and deprivation in Japan: A nationwide ecological study with contextual suicide mortality
Masahide Koda, Shuhei Nomura, Makoto Kaneko, Nahoko HaradaAbstract
Aim
We examined whether rurality was associated with psychiatric standardized claim ratios computed using medical‐institution location, and how these patterns related to area deprivation and residence‐based suicide mortality in Japan.
Methods
We conducted a nationwide ecological study of 335 secondary medical areas. Rurality, measured with the Rurality Index for Japan, and a census‐based Area Deprivation Index were population‐weighted from municipalities. Psychiatric specialty therapy claims (April 2023 to March 2024) from Japan's national health‐insurance claims open data were expressed as standardized claim ratios: observed claims billed by institutions in each area divided by national age–sex expected claims. Suicide deaths (April 2022 to March 2025) were expressed as age–sex standardized mortality ratios. Ecological rate‐ratio trends were estimated with quasi‐Poisson models.
Results
Outpatient claim ratios were lower in rural than urban areas but non‐monotonic (most urban, 108.4; fourth quintile, 76.9; and most rural, 81.3). Inpatient ratios peaked in the second quintile, and suicide mortality rose with rurality. Per 10‐point higher rurality score, outpatient volume was lower (rate ratio, 0.939; 95% confidence interval, 0.921–0.956) and remained lower after including deprivation (0.919; 0.896–0.942); the rurality–suicide association attenuated. Nearest‐three‐area pooling raised the most rural median outpatient ratio from 62.2 to 101.1.
Conclusion
More rural areas had lower outpatient psychiatric claim volume billed by local institutions relative to national age–sex benchmarks, but this contrast narrowed after nearest‐area pooling, consistent with cross‐boundary care. These institution–location ratios should prompt local review with residence‐based and capacity data, not be read as evidence of under‐provision.