情報量クラスター分析による医療圏域の階層構造の把握

概要

論文の詳細を見る
Areal subdivision of a region is essential for the rational allocation of health resources. The objective of this paper is to develop a quantitative method for identifying the hierarchical structure of hospital catchment areas by the cluster analysis using the Kullback's minimum discriminant information as a criterion of proximity. As a data set for application, a patient home-hospital flow matrix is formulated from the 1977 patient survey undertaken by the Ibaraki Prefecture. An isometric image of the hierarchical structure is drawn for each of the out-patient flow matrix and in-patient flow matrix. This method is judged superior to the conventional cluster analysis in that there is no need for the preprocessing of the matrix data and that it has no problem in treating the zero entry cell, which could be a problem for the cluster analysis using chi-square values as proximity criteria.
社団法人日本建築学会の論文
1986-12-30