Persistent prefectural ordering of billed urinary albumin quantification in Japan: a nationwide longitudinal ecological study

Nationwide Japanese claims studies have documented regional variation in quantitative urinary albumin or protein testing, but it is unclear whether geographic ordering persists over time or depends on the denominator used. We evaluated the persistence of prefectural ordering in billed urinary albumin quantification and examined its robustness using a population-based denominator.

This ecological provider-location study used National Database Open Data for 47 prefectures. FY2015-FY2022 was the primary historical interval; FY2023-FY2024 was a scope-aligned extension. The primary metric was urinary albumin quantification billing counts per 1,000 HbA1c billing counts. A sensitivity analysis used counts per 100,000 total population. Equal-weight dispersion, endpoint Spearman correlations, Kendall's W across annual ranks, and cross-metric rank correspondence were calculated.

During FY2015-FY2022, the national primary metric increased from 26.20 to 33.54 (+ 28.0%). Prefectural ordering was highly persistent (Spearman rho = 0.904; Kendall's W = 0.964). The population-referenced rate increased from 1,480.7 to 2,270.9 per 100,000 (+ 53.4%) and showed comparable persistence (rho = 0.920; W = 0.967). Annual rank correlations between the two metrics ranged from 0.893 to 0.915. FY2024 extension values were 36.19 per 1,000 HbA1c counts and 2,570.4 per 100,000 population, with persistence maintained.

Prefectural ordering of billed urinary albumin quantification persisted under both HbA1c-referenced and population-referenced definitions. These findings identify a stable geographic billing pattern but do not establish its demographic, clinical, or organizational determinants or estimate patient-level albuminuria-testing coverage.

The online version contains supplementary material available at https://doi.org/10.1007/s13340-026-00944-z.