Epidemiological Characteristics And Prediction Of Hand-Foot-Mouth Disease In Yongzhou City,Hunan Province From 2010 To 2020

Objective:(1)To analyze the epidemiological and etiological characteristics of hand-foot-mouth disease(HFMD)in Yongzhou city from 2010 to 2020.(2)To analyze the spatial distribution characteristics of HFMD incidence in Yongzhou city from 2010 to 2020,and to identify the high incidence areas;(3)Based on the time series data in the past decade,a prediction model with significant prediction effect was established to predict the incidence of hand-foot-mouth disease in 2022.Methods:(1)The incidence data and etiological data of HFMD in Yongzhou city from January 1,2010 to December 31,2020 were obtained through China Disease Control and Prevention Information Monitoring System.(2)Descriptive epidemiological methods were used to analyze the time,region and population distribution characteristics of HFMD in Yongzhou city from 2010 to 2020.And the overall detection status of hand,foot and mouth disease in Yongzhou city from 2010 to2020 were described.(3)Excel 2010 was used to sort out a database including administrative codes and morbidity was established.Geo Da1.20 was used to generate spatial weight matrix based on Queen adjacency space moment weight method,and global autocorrelation and local autocorrelation analysis were performed.(4)Using SPSS26.0,SARIMA(Seasonal Autoregressivity moving Average Mixed Model)model was constructed based on whether COVID-19 data in 2020 were included or not.By comparing the stationary,BIC and mean relative errors of the model,the model with better predictive effect was selected to predict the monthly incidence in 2022.Results:(1)From 2010 to 2020,the average annual incidence of HFMD in Yongzhou city was 159.95/100,000,the severe disease ratio was 0.95%,and the fatality rate was 0.05%.The incidence was double peak in one year,and the cases were mainly concentrated in the age group of 3 years old and below.The incidence rate of males was higher than that of females,and the highest annual incidence was found in Jiangyong county and Shuangpai County.The occupational distribution was mainly scattered children.From 2010 to 2020,the detection rate of positive samples in Yongzhou city was 59.28%.(2)From 2010 to 2020,the incidence of HFMD in Yongzhou city showed a positive spatial autocorrelation distribution,and the high-high spatial correlation showed a trend of shifting from north to south,and the concentration area was decreasing.After 2015,the high-high concentration area was mainly concentrated in the southeast.(3)SARIMA(1,0,1)(0,1,1)12 was 0.75,BIC was 3.62,and the mean relative error was 16.55%.The monthly incidence data from January 2010 to December 2019 were used to construct SARIMA(0,0,2)(0,1,0)12,with a stationary value of 0.53,BIC of 10.62,and an average relative error of 24.96%.The model SARIMA(1,0,1)(0,1,1)12 with better fitting effect was used to predict the monthly incidence in 2022,which was 3.82(March),22.92(June),14.72(September),6.48(December).Conclusions:(1)From 2010 to 2020,the incidence of HFMD in Yongzhou city showed a decreasing trend,with obvious seasonal characteristics,showing double peaks in one year,from April to July and from September to October,respectively.In terms of regional distribution,jiangyong county and Shuangpai County had the highest incidence rate.The incidence rate in males was higher than that in females,and the incidence was concentrated in children under 3 years old.The pathogens were mainly EV71 and Cox A16.(2)From 2010 to 2020,the incidence of HFMD in Yongzhou showed a positive spatial autocorrelation,and the high-height spatial correlation shifted from north to south,with the concentration area decreasing.After 2015,the high-altitude concentration area was mainly concentrated in the southeast.(3)SARIMA model can be used to predict the monthly incidence of HAND-foot-mouth disease,and the incidence in 2020 can be taken into account when constructing the infectious disease prediction model.At the same time,because the model is only suitable for short-term prediction,the model parameters need to be updated constantly in practical application.