基于超几何分布的前瞻性时空扫描统计量在疟疾早期预警中的应用  

The Application of Prospective Space-time Scan Statistics Based on Hypergeometric Distribution Model in the Early Warning of Malaria

作　　者：陈飞[1] 李晓松[1] 冯子健[2] 赵星[1] 

机构地区：[1]四川大学华西公共卫生学院流行病学与卫生统计学系,610041 [2]中国疾病预防控制中心

出　　处：《中国卫生统计》2015年第2期186-189,共4页Chinese Journal of Health Statistics

基　　金：四川大学青年教师科研启动基金(2015SCU11012)

摘　　要：目的探索基于超几何分布概率模型的前瞻性时空扫描统计量在疟疾早期预警中的应用。方法利用R语言编程实现两种前瞻性时空扫描统计量,分别基于经典的泊松分布概率模型和较新的超几何分布概率模型,模拟实时监测系统对四川省2009年疟疾病例中6月21日-30日连续10天数据进行逐日前瞻性分析。结果基于超几何分布的前瞻性时空扫描统计量和基于泊松分布的前瞻性时空扫描统计量在疟疾早期预警中效果相当,都能早期探测到疟疾的爆发。结论基于超几何分布的前瞻性时空扫描统计量在疟疾早期预警中有应用价值。Objective To investigate the application of prospecti ve space-time scan statistic in the early warning of malaria. Methods We implement the R code for two prospective space-time scan statistic s, based on Poisson and hypergeometric models, respectively. They were applied to the d aily prospective analyses of malaria dat a of Sichuan province from June 21th to 30th, 2009 . Results The prospective space-time scan statistic based on Poisson and hypergeometric models both could timely detect the outbreaks of malaria. Conclusion The prospective space-time scan statistic based on hypergeometric distribution model has the potential value in the early warning of infectious diseases.

关 键 词：传染病预警 时空扫描统计量 超几何分布模型 泊松分布模型 

分 类 号：R531.3[医药卫生—内科学] R181.3[医药卫生—临床医学]