中国主要呼吸道传染病分布模式及其社会经济影响因素

Distribution patterns of the main respiratory infectious diseases in China and their associated socio-economic factors

目的本研究旨在探讨影响我国大陆五种主要呼吸道传染病（麻疹、肺结核、流行性脑脊髓膜炎、百日咳、猩红热）流行的社会经济因素。方法通过聚类分析和多维尺度分析，阐明五种呼吸道传染病的分布模式；采用主成分分析从社会经济数据中提取主成分；然后通过MANTEL检验和匹配分析量化发病与影响因素之间的相关关系。结果聚类分析和多维尺度分析结果显示各省可按发病率聚类为6组。MANTEL检验结果显示疾病非相似性矩阵与影响因素主成分矩阵之间的相关系数范围为0．220～0．375。而BIO-ENV匹配分析结果显示较好，相关系数范围为0．402～0．545，最匹配影响因素包括0～14岁人口比重、平均受教育年限、文盲人口占15岁以上人口比例、城镇人口比重、人均日生活用水量和失业率。结论2009-2012年中国各省份主要呼吸道传染病的分布模式相对稳定，与同期的社会经济因素存在关联，可综合防治呼吸道传染病。

Objective The study was conducted to identify socio-economic predictors of five major respiratory in- fectious diseases （Measles, Tuberculosis, Epidemic Cerebrospinal Meningitis, Pertussis and Scarlet Fever） in China's Mainland. Methods Cluster analysis and multidimensional scaling analysis were applied to explore the distribution patterns of these diseases, with principal component analysis used for extracting principal components from the original socio-economic data. Then, mantel test and matching analysis were conducted to quantify the predictor-outcome relationships. Results The results of cluster analysis and multidimensional scaling analysis showed that provinces in China can be classified into 6 groups according to disease burdens. The results of mantel test showed that the correlation coefficient between the disease dissimilarity matrix and the principal component matrix of socio-economic predictors ranged from 0. 220 to 0. 375, whereas results of matching analysis tended to be better, with the correlation coefficient ranging from 0. 402 to 0. 545. The best matching predictors identified were ＂Proportion of children under 15 years old＂, ＂Average education years＂, ＂Illiteracy rate among people beyond 15 years old＂, ＂Urban residents proportion＂, ＂Water consumption per capita per day＂ and ＂Unemployment rate＂. Conclusions The distribution patterns of the main respiratory infectious diseases in China between 2009 to 2012 were relatively stable and closely related to the socio-economic predictors. And comprehensive control pro- grams should be implemented.