EUVIMEDOpdag EUVIMED
Open access

Non-Physician Areas in Japan: Typological Classification Based on Regional Demographic Characteristics

Takashi Kuwayama, Kazuhiko Kotani

JMIR Formative Research · 2026-10-09 · 10 · e101570-e101570

DOI: 10.2196/101570 ↗

At a glance

Journal article in JMIR Formative Research, published online on 2026-10-09. Written by Takashi Kuwayama and 1 co-authors. Institution: Center for Community Medicine, Jichi Medical University, 3311-1 Yakushiji, Shimotsuke-City, Tochigi, 3290498, Japan, +81-285-58-7394. Major MEDLINE topics: Demography, Health Services Accessibility. 25 references. The full text is freely available in PubMed Central.

AbstractSubjectsBibliografiske oplysningerCiteRelated studies

Abstract (original language)

Background

Across Japan, there are “non-physician areas” where medical institutions are not present and health care access is difficult. Although these areas are defined as a single administrative category, they may vary according to regional demographic characteristics. Understanding this possible heterogeneity may be important for developing more context-sensitive strategies to improve health care accessibility.

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 identified from the national survey conducted by the Ministry of Health, Labour and Welfare of Japan (2019). Data on regional demographic characteristics were also obtained. After z score standardization, k-means clustering was used for classification. The optimal number of clusters was determined using silhouette scores. The data on non-physician areas were divided into training (354/590, 60.0% areas) and validation (236/590, 40.0% areas) datasets. The reproducibility of the cluster structure was evaluated using Jensen-Shannon distance and chi-square tests, and principal component analysis (PCA) was used to examine the overall structure of the clusters.

Results

The optimal number of clusters was 3 (silhouette score=0.266), indicating that non-physician areas could be classified into 3 types. The cluster distributions of the training and validation datasets were consistent (Jensen-Shannon distance=0.017; P =.76), supporting the reproducibility of the cluster structure. The PCA suggested that differences among non-physician areas could be interpreted along 2 axes, including population age structure and settlement size. The 3 clusters represented younger, intermediate, and older types, characterized by differences in demographic and settlement characteristics.

Conclusions

The present study demonstrated that administratively defined non-physician areas in Japan are not uniform and can be classified into 3 demographically distinct types. The typology may provide a framework for understanding differences not captured by the current administrative designation and for considering more context-sensitive health care strategies.

Subjects

MeSH terms (MEDLINE)

DemographyHealth Services AccessibilityJapanHumansClustering AlgorithmsCluster AnalysisSurveys and QuestionnairesPrincipal Component Analysis

Author keywords

clustermachine learningnon-physician areasregional characteristicsrural health care

Authors and institutions

  • Center for Community Medicine, Jichi Medical University, 3311-1 Yakushiji, Shimotsuke-City, Tochigi, 3290498, Japan, +81-285-58-7394.

Cite

APA

Kuwayama, T., & Kotani, K. (2026). Non-Physician Areas in Japan: Typological Classification Based on Regional Demographic Characteristics. JMIR Formative Research, 10, e101570-e101570. https://doi.org/10.2196/101570

Vancouver

Kuwayama T, Kotani K. Non-Physician Areas in Japan: Typological Classification Based on Regional Demographic Characteristics. JMIR Formative Research. 2026;10:e101570-e101570. doi:10.2196/101570

BibTeX
@article{Kuwayama2026,
  author = {Kuwayama, Takashi and Kotani, Kazuhiko},
  title = {Non-Physician Areas in Japan: Typological Classification Based on Regional Demographic Characteristics},
  journal = {JMIR Formative Research},
  year = {2026},
  volume = {10},
  pages = {e101570--e101570},
  doi = {10.2196/101570}
}
Start for free →