| ObjectiveVaricella was one of the high contagious diseases caused by varicella-zoster virus.This study deeply analyzed epidemiologic characteristics of varicella in Qingdao from 2007 to 2016 and aimed to provide basic data for varicella control and prevention.Methods1.Data of varicella in Qingdao from 2007 to 2016 were collected through the China Information System for Disease Control and Prevention,then use descriptive study to analyze time distribution,regional distribution and populati on distribution of varicella cases in Qingdao.The statistical analysis was performed with Excel 2007 and SPSS 18.0 software,and the ratio of multiple groups was compared by ? test.2.The monthly incidence of varicella in 2007?2015 years was modeled by SPSS18.0 software,then the monthly incidence of varicella in Qingdao in 2016 was predicted based on the optimal model,and the actual incidence of varicella in 2016 was compared with the results predicted by the model.3.According to the annual average incidence of varicella in 2007?2016 years in Qingdao,the ArcGIS10.0 software was used to analyze the autocorrelation analysis and local autocorrelation analysis,so that it can explore spatial distribution mode for estimating high-incidence area and low-incidence area.Results1.The annual morbidity of varicella in Qingdao from 2007 to 2016 was 27.78/105,32.76/105,37.30/105,32.10/105,35.63/105,31.27/105,31.92/105,28.80/105,32.47/105 and 31.37/105,the incidence has remained stable(?rn =,P).Circular distribution analysis indicated that the high-incidence of varicella was from November 17th to June 14th next year,P.All districts have cases,mainly distributed in urban areas.The male/female ratio of the incidence of varicella was 1.26:1,the morbidity of males was higher than that of women each year(P).Varicella cases occurred in all age groups and the annual average morbidity was highest in 5?9 year age group(153.80/105).In all kinds of professions,students occupied the highest proportion(50.30%),preschool children were in the second(25.51%).2.The time sequence diagram of the monthly number of varicella in Qingdao from 2007 to 2015 became a stationary sequence after the 1 order seasonal difference,and then through a series of steps,such as white noise test,model identification,parameter estimation,BIC criterion and so on,it is find that SARIMA(1,0,0)×(1,1,0)12 was the best model.The model was used to predict the monthly number of varicella in 2016,it is find that the model has good prediction performance,and the average relative error was 11.51%.3.Moran's ? s a is ics of full-domain autocorrelation analysis indicated that the inci n c of varic 11a ha spa ial clus ring Moran's I,P<0.05).Getis-Ord General G statistics found that this clustering was high-value clustering(ZG=2.093459,P<Local Moran's ? P<0.05)and local hot spot(P<0.05)of local autocorrelation analysis revealed that Shibei district and Licang district were both high-high value area,Jimo district and Pingdu district were both low-low value area.Conclusions1.In the past 10 years,the incidence of varicella in Qingdao is generally stable,mainly in urban areas,and school-age children are high incidence groups.2.Time series analysis was suit for predicting the monthly incidence of varicella in Qingdao.Predicted incidence trend was the same as the truth.3.With the help of GIS,the spatial distribution characteristics of varicella in Qingdao can be analyzed thoroughly,so as to guide the epidemic prevention and control scientifically. |
PDF Full Text Request