Examination of Regional Differences in Performance of Health Systems by Prefecture Using Cluster Analysis

Abstract

OBJECTIVE: The Organisation for Economic Co-operation and Development (OECD) pointed out the deterioration of the performance of health systems (performance) in Japan. To correct this, we evaluated the performances in different prefectures. METHODS: By the OECD method, we set 27 indicators concerning "health status (HS)", " risk factors (RFs)", " access to care (AC)", " quality of care (QC)", and " health care resources (HRs)". Next, the relative value (RV, ±4 standard deviation) from the average value of each indicator in each prefecture was obtained. On the basis of this RV, the prefectures were divided into A, B, and C by cluster analysis. Then, the 27 indicators of the three groups were subjected to multiple comparison tests and the performances were evaluated. RESULTS: A included Hokkaido, Ishikawa, Kyoto, Shimane, Okayama, Hiroshima, Yamaguchi, Tokushima, Kagawa, Ehime, Kochi, Fukuoka, Saga, Nagasaki, Kumamoto, Oita, Miyazaki, Kagoshima, and Okinawa, B included Aomori, Iwate, Akita, Fukushima, Ibaraki, Tochigi, Saitama, Tokyo, Osaka, and Wakayama, C included Miyagi, Yamagata, Gunma, Chiba, Kanagawa, Niigata, Toyama, Fukui, Yamanashi, Nagano, Gifu, Shizuoka, Aichi, Mie, Shiga, Hyogo, Nara, and Tottori. The multiple comparison test results showed that HS and RFs were not significantly different between A and C. In A, AC and QC were poor, but HRs were excessive, and the local allocation tax was high. RFs, AC, QC, HRs, and the local allocation taxes were not significantly different between B and C, but HS was poor in B. CONCLUSIONS: The performance of health systems was deteriorating in the 19 prefectures included in A, and correction is necessary in these prefectures.