Depopulation masks persistent geographical inequality in physician supply: an exact decomposition of Japan’s physician maldistribution index between its 2024 and 2026 publication rounds

Abstract Background Geographical inequality in physician supply is monitored almost everywhere with density indicators, and these indicators determine which areas are designated underserved. In depopulating societies their numerator and denominator move in opposite directions, so an indicator can improve without any change in deployment. We quantified this for Japan’s national physician maldistribution index. Methods The index equals the standardised physician count divided by the product of population and a standardised utilisation ratio, which permits an exact logarithmic decomposition into supply, population-size and demand-structure effects. Using the official component data for 47 prefectures and 330 secondary medical areas at two publication rounds (January 2024 and April 2026), we decomposed observed change, constructed counterfactual indices holding population constant, and compared dispersion and β-convergence. Sensitivity analyses included a headcount–hours split of the supply effect, an exact decomposition of change in the utilisation ratio, population-weighted dispersion and a prefecture-clustered bootstrap. Results The identity reproduced all 377 published values exactly. Nationally, 43.0% of the rise from 255.6 to 266.8 was attributable to population decline; across medical areas the share was 67.3%. Among the 16 highest-index prefectures the mean supply effect was zero, reflecting headcount rather than the hours adjustment. Observed convergence among medical areas (β = −0.085, t = − 5.74) was largely absent in the counterfactual (β = −0.025, t = − 1.72; cluster-bootstrap 95% interval − 0.065 to + 0.008; difference from the observed coefficient − 0.060, 95% interval − 0.068 to − 0.048). In the smallest population quintile the index rose by 10% although the standardised physician count fell, and population decline accounted for 83% of the rise. Conclusions Under depopulation, density indicators can register improving equity even when the dispersion of physician supply, net of population change, is essentially unchanged; the discrepancy is greatest at the spatial scale at which scarcity is officially designated. Equity monitoring should report dispersion alongside levels, publish decomposed indicator change, and anchor essential-service indicators on capacity rather than on shrinking utilisation denominators.

Authors

Institutions

Publication Details

Journal
Research Square
Published
2026-11-13
DOI
https://doi.org/10.21203/rs.3.rs-11481394/v1
Primary Topic
Global Health Workforce Issues
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Depopulation masks persistent geographical inequality in physician supply: an exact decomposition of Japan’s physician maldistribution index between its 2024 and 2026 publication rounds

Taeuk Kang, Sookja Choi
Research Square
Global Health Workforce Issues
preprint

Depopulation masks persistent geographical inequality in physician supply: an exact decomposition of Japan’s physician maldistribution index between its 2024 and 2026 publication rounds

Taeuk Kang, Sookja Choi
preprint en

Abstract

Abstract Background Geographical inequality in physician supply is monitored almost everywhere with density indicators, and these indicators determine which areas are designated underserved. In depopulating societies their numerator and denominator move in opposite directions, so an indicator can improve without any change in deployment. We quantified this for Japan’s national physician maldistribution index. Methods The index equals the standardised physician count divided by the product of population and a standardised utilisation ratio, which permits an exact logarithmic decomposition into supply, population-size and demand-structure effects. Using the official component data for 47 prefectures and 330 secondary medical areas at two publication rounds (January 2024 and April 2026), we decomposed observed change, constructed counterfactual indices holding population constant, and compared dispersion and β-convergence. Sensitivity analyses included a headcount–hours split of the supply effect, an exact decomposition of change in the utilisation ratio, population-weighted dispersion and a prefecture-clustered bootstrap. Results The identity reproduced all 377 published values exactly. Nationally, 43.0% of the rise from 255.6 to 266.8 was attributable to population decline; across medical areas the share was 67.3%. Among the 16 highest-index prefectures the mean supply effect was zero, reflecting headcount rather than the hours adjustment. Observed convergence among medical areas (β = −0.085, t = − 5.74) was largely absent in the counterfactual (β = −0.025, t = − 1.72; cluster-bootstrap 95% interval − 0.065 to + 0.008; difference from the observed coefficient − 0.060, 95% interval − 0.068 to − 0.048). In the smallest population quintile the index rose by 10% although the standardised physician count fell, and population decline accounted for 83% of the rise. Conclusions Under depopulation, density indicators can register improving equity even when the dispersion of physician supply, net of population change, is essentially unchanged; the discrepancy is greatest at the spatial scale at which scarcity is officially designated. Equity monitoring should report dispersion alongside levels, publish decomposed indicator change, and anchor essential-service indicators on capacity rather than on shrinking utilisation denominators.

Research Square
Sungshin Women's University (KR), Chung-Ang University (KR)
Reduced inequalities
Global Health Workforce Issues
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon