Accepted for/Published in: JMIR Formative Research
Date Submitted: May 16, 2026
Date Accepted: Aug 27, 2026
Typological classification of non-physician areas in Japan
ABSTRACT
Background:
Across Japan, there are “non-physician areas” where medical institutions are not present, but medical care should be provided. These areas are considered to vary depending on the regional demographic characteristics.
Objective:
This study examined whether non-physician areas can be classified based on regional demographic characteristics using an unsupervised machine learning model.
Methods:
A total of 590 non-physician areas were seen from the Ministry of Health, Labour, and Welfare Survey of Non-physician Areas in Japan (2019). Data on regional demographic characteristics, including population composition and automobile ownership, were also obtained. After z-score standardization, K-means clustering was used for the classification. The optimal number of clusters was determined using the silhouette values. The data on non-physician areas were divided into training (354 areas) and validation (236 areas) data. The reproducibility of the cluster structure was evaluated using Jensen–Shannon distance and chi-square tests, and a principal component analysis (PCA) was used for confirmation of the results.
Results:
The optimal number of clusters was three (silhouette=0.240), indicating that non-physician areas could be classified into three types. The cluster distributions of the training and validation data were consistent (JS distance=0.025, p=0.563), confirming the reproducibility of the cluster structure. In the PCA, non-physician areas can be interpreted along two axes: population age structure and settlement size and living infrastructure. The three clusters represented population types characterized by a relatively younger age structure, an advanced aging structure, and an intermediate structure.
Conclusions:
Three types of non-physician areas were identified using the unsupervised machine learning model in the study. These findings suggest that non-physician areas were not uniform and that the design of medical care systems might need to be adopted to regional demographic characteristics, particularly in rural areas. Clinical Trial: -
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.