Performance of China infectious disease automated-alert and response system in 2016
Honglong Zhang,Lingjia Zeng,Shengjie Lai,Liping Wang,Zhongjie Li
DOI: https://doi.org/10.3784/j.issn.1003-9961.2018.02.011
2018-01-01
Abstract:Objective To analyze the performance of China Infectious Diseases Automated-alert and Response System(CIDARS) established in 2008 by China CDC,and provide reference for the improvement of CIDARS. Methods CIDARS used fixed-value detection method and temporal aberration detection method for aberrant detection, with the work flowchart of signals generating, preliminary verification and field investigation. The fixed-value detection method was for 15 diseases, and the temporal aberration detection method was for 33 diseases, including 3 types of sub-methods,such as moving percentile method, cumulative sum control chart algorithm and cluster early-warning method. Data analysis also included the concerned priority diseases which were added by province according to the actual disease control needs. All the signals generated by CIDARS in 31 provinces in 2016 were collected, the response status,such as response rate and response timeliness, and preliminary verification outcome, such as the signals related to suspected outbreaks and its proportion, were deeply analyzed from the dimension of province and disease, respectively. Results A total of 325 208 signals were generated nationwide by the system, in which 323 271 (99.40％) were responded,300 614 (92.44％) were responded within 24 hours. The proportion of signals of suspected outbreaks was 4.91％ (15 964). The first 3 provinces with high number of signals were Shandong, Henan and Guangxi, and Shandong had highest proportion of signals of suspected outbreaks(22.33％). Among all the signals, 97 689 were generated by the fixed-value detection method, the overall response rate of the signals was 98.83％, and in 19 provinces it was 100％. The median interval of the response was 0.72 h (P25 = 0.16 h, P75 = 4.09 h),the response rate within 2 hours was 65.66％. In addition, 227 519 signals were generated by the temporal aberration detection method,in which 3 398 (1.49％) were verified as suspected outbreaks. The response rate of signals was 99.65％ (226 724),and the response rate within 24 hours was 93.23％. The median interval of signals was 0.99 h (P25 = 0.50 h, P75 = 2.44 h). The first 3 provinces with high response rates within 24 hours were Guizhou, Hunan and Shanghai, the first 3 diseases with highest proportion of signals of suspected outbreaks were dengue fever, leptospirosis and rubella. Conclusion In 2016,the performance of CIDARS was well, with response and timely response rates at high levels, but the proportion of signals of suspected outbreaks was low,so it is necessary to further improve CIDARS by identifying the areas and seasons with different disease incidence levels, exploring new data sources and narrowing the space detection coverage.