累积和模型在细菌性痢疾早期预警中的应用

目的探讨累积和(CUSUM)模型在细菌性痢疾早期预警监测中的应用。方法数据来源于2007年北京市法定传染病报告系统细菌性痢疾报告数据和北京市2007年细菌性痢疾暴发疫情处理记录。预警运算使用美国CDC早期异常报告系统软件。结果 330个街道或乡镇共发出预警信息数为3743条,平均每个街道或乡镇发出11.3条预警信息,其中最少的为0条,最多的为25条,中位数为12,四分位间距为7。及时预警了2007年疫情记录中的两起细菌性痢疾暴发。结论累积和模型运算过程简单,预警频次和因预警产生的审核、流调工作量均在可接受范围内,可以应用在日常细菌性痢疾监测和防控工作中。

Early warning for bacillary dysentery using cumulative sums model

Objective To explore application of cumulative sums model （CUSUM） in the early warning for bacillary dysentery. Methods Data for analysis were collected from notifiable disease networking report system of China and records of investigations of and responses to bacillary dysentery outbreaks in Beijing. CUSUM was fitted with the Early Aberration Reporting System （EARS） software form the Centers for Disease Control and Prevention in the United States. Results In total,3743 records of early warning messages reported from 330 sub-districts and townships （or towns） were used for fitting CUSUM,ranging 0 to 25 record items with an average of 11. 3 items per sub-district or township （ or town） and a median of 12 items and an interquartile of seven items. Two outbreaks of bacillary dysentery was timely detected in 2007by early warning with CUSUM. Conclusions Algorithm of CUSUM is simple and frequency of early warning,its check and workload of outbreak investigations all can be acceptable. Therefore,CUSUM can be employed in early warning for bacillary dysentery outbreak.