Spatial Epidemiology Characteristics Of Pertussis In Hunan Province And Prediction Of Short-term Incidence Trend

Objective:To explore the spatial epidemiological characteristics of pertussis incidence in Hunan Province from 2009 to 2019,established a pertussis incidence prediction model,and predicted short-term incidence trends to provide scientific basis for the prevention and control of pertus-sis epidemic in Hunan Province.Methods:Collect the pertussis surveillance data from 2009 to 2019in the infectious disease surveillance system of the"China Disease Pre-vention and Control Information System",and use SPSS18.0 software to analyze the classic epidemiological characteristics of the incidence of pertussis in Hunan Province from 2009 to 2019.Arc GIS 10.3 and Sa TScan9.6 software were used to analyze the spatial epidemiological characteristics of pertussis in Hunan Province from 2009 to 2019.SPSS18.0 software was used to establish an autoregressive integrated moving average（ARIMA）model,and to predict the incidence of pertus-sis in 2020.Results1.From 2009 to 2019,Hunan Province reported a total of 6,796pertussis cases,with no deaths,and an average annual incidence rate of0.899/100,000.There was a small peak of incidence every 3 to 5 years.The range of fluctuation from 2009 to 2017 was relatively small.The in-cidence rate has been reported to rise sharply to 2.895 per 100,000 in2018,and the incidence rate continued to rise significantly to 6.184 per100,000 in 2019.2.From 2009 to 2019,Hunan Province reported a total of 6,796pertussis cases,with no deaths,and an average annual incidence rate of0.899/100,000.There was a small peak of incidence every 3 to 5 years.The range of fluctuation from 2009 to 2017 was relatively small.The in-cidence rate has been reported to rise sharply to 2.895 per 100,000 in2018,and the incidence rate continued to rise significantly to 6.184 per100,000 in 2019.3.The thematic map of the incidence of pertussis at the county and district level in Hunan Province from 2009 to 2019 showed that the inci-dence of pertussis in all counties and districts remained at a low level from 2009 to 2016,and the relatively high incidence areas were mainly in the parts of Yiyang City in the northeast.In the county.In 2017,it ex-panded from the northeast to the southwest.There was an outbreak of pertussis in 2018 and 2019.Almost all counties and districts had different degrees of incidence,and the incidence rate had increased significantly compared with before 2018.The Moran,s I index of the global autocor-relation analysis results was greater than 0 in each year.Except for 2013,2014 and 2016,the global spatial autocorrelation analysis of the inci-dence of pertussis in other years all had P<0.05.According to the annual LISA cluster map of pertussis incidence,the detected high-high clusters occurred mainly in Yiyang（Taojiang County,Heshan District,Ziyang District,and Yuanjiang City）,Yueyang（Miluo City,Xiangyin County and Yueyang County）,Changsha（Tianxin District,Changsha County,Ning-xiang City and Wangcheng District）and Shaoyang（Daxiang District,Shaodong City,Shaoyang County,Shuangqing District and Xinshao County）.Pure space scanning,a total of seven levels of gathering areas were detected.The first-level gathering areas were Tianxin District,Yuhua District,Yuelu District,Wangcheng District,Furong District,Kai-fu District and Ningxiang City in Changsha;Tianyuan District,Shifeng District,Hetang District and Lusong District in Zhuzhou;Xiangtan City Shaoshan City,Xiangxiang City,Yuhu District,Xiangtan County and Yuetang District;Louxing District and Shuangfeng County in Loudi;Hengshan County in Hengyang and Heshan District in Yiyang;LLR val-ue was 207.00,RR value was 1.74.The remaining six-level agglomera-tion areas were mainly in Huaihua City,Xiangxi Tujia and Miao Auton-omous Prefecture and some counties under the jurisdiction of Hengyang City.A total of four levels of clusters were detected by spatio-temporal scanning analysis.The gathering time of the first-level clusters is from March 2019 to September 2019.The gathering areas are Tianxin District,Yuhua District,Yuelu District,Wangcheng District,Furong District,Kai-fu District,Ningxiang City and Changsha County in Changsha.;Tian-yuan District,Shifeng District,Hetang District,Lusong District and Zhuzhou County in Zhuzhou;Shaoshan City,Xiangtan City,Yuhu Dis-trict,Xiangtan County and Yuetang District in Xiangtan;Heshan District,Taojiang County,Ziyang District and Yiyang City Xiangyin County;Louxing District and Shuangfeng County in Loudiand Hengshan County and Nanyue District in Hengyang;LLR value was 574.97,RR value was3.12.The gathering time of the remaining three-level clusters was rough-ly distributed from April 2018 to October 2019.The gathering areas were mainly in Chenzhou City,Xiangxi Tujia and Miao Autonomous Prefec-ture,and some counties under the jurisdiction of Yongzhou City.4.ARIMA（0,1,1）×（0,1,0）₁₂ was the optimal prediction model.The actual value of the monthly incidence of pertussis in Hunan Province in 2019 was basically the same as the predicted value,and the average relative error was 3.55%,The forecast effect was better.The model pre-dicts that the incidence rate in 2020 will be 7.642/100,000,which was higher than 2019（6.218/100,000）.Conclusions1.The incidence of pertussis from 2009 to 2019 had the characteris-tics of"high-low-high-low"irregular fluctuations.In 2018,the incidence of pertussis in Hunan Province increased significantly,with an outbreak,and the incidence had continued to increase,and there may be a“pertus-sis reappears”phenomenon.2.May-September were the months of high incidence of pertussis,and October-February of the following year were the months of low inci-dence;the high incidence areas were Changsha City,Yiyang City,Loudi City and Xiangxi Tujia and Miao Autonomous Prefecture;the average annual incidence of males is slightly higher female.The reported cases are mainly scattered children,children in kindergartens and students.3.The distribution of pertussis in Hunan Province had temporal and spatial agglomeration.The areas where pertussis epidemics were concentrated were mainly in the central and northeastern parts of Hunan Province,and the time dimension was from March to October.4.The ARIMA model can better predict the short-term incidence trend of pertussis,and the model predicts that the incidence level will continue to rise in 2020.