Study On Epidemiological Characteristics And Prediction Of Severe Hand-foot-and-mouth Disease In Hunan Province From 2011 To 2020

Objectives: To analyze the epidemiological characteristics of severe hand-foot-and-mouth disease in Hunan Province from 2011 to2020,to identify the spatial distribution pattern and hot areas of severe cases,and to explore the influencing factors and to construct a prediction model for severe cases.Methods: Based on the surveillance data of hand-foot-and-mouth disease and the database of severe cases,mild cases and demographic data in Hunan Province from 2011 to 2020;Descriptive epidemiology study was used to describe the epidemiological characteristics of severe cases;The spatial autocorrelation analysis was performed to analyze the spatial distribution pattern of severe cases and to find hot spots in the district and county level;Univariate analysis and multivariate logistic regression analysis were used to explore the influencing factors,and to develop a prediction model for severe cases,and a nomogram was draw to show the prediction model.Results:(1)From 2011 to 2020,a total of 1462732 cases with hand-foot-and-mouth disease,and 8469 severe cases were reported in Hunan Province,with a case-severity rate of 0.08/100 000 to 4.72/100000;Severe cases showed a high incidence rate every other year from2011 to 2016,and decreased year by year since 2016,with the peak time of incidence mainly from April to July and from October to November;Male(5453 cases,64.4%),children under 3 years old(6989 cases,82.5%),and children living at home(7843 cases,92.6%)were more easier to be infected;The areas with a large number of severe cases were Xiangxi Tujia and Miao Autonomous Prefecture,Shaoyang City and Loudi City,respectively,with a total of 5026 cases,accounting for 59.3%;(2)The pathogen constituent was dominated by EV-A71,accounting for65.17% of laboratory confirmed cases,but the proportion of other enterovirus was increasing;(3)Global autocorrelation analysis revealed a spatially clustered distribution of severe cases,the results of the local spatial autocorrelation analysis were consistent with the spatial distribution,and the majority of the hot spots of severe cases were the districts and counties of Xiangxi Tujia and Miao Autonomous Prefecture and Shaoyang City.(4)A prediction model was developed by the independent variables such as age,gender,occupation,address type,feeding type,guardian’s educational level,infected in the month before onset,interval time between onset and first visit,fever,rash/herpes for severe cases,the AUC of the prediction model was 0.752(95%CI:0.737-0.767)with an accuracy of 69.0%,according to the values of the corresponding variables,the predictive value of the probability of severe cases could be obtained in the nomogram.Conclusion:(1)The number of severe cases and severe rate of hand-foot-and-mouth disease in Hunan Province has totally decreased year by year since 2016,the long-term trend is a decreased trend.The key groups are male,children under 3 years old,and children living at home.The key areas are Xiangxi Tujia and Miao Autonomous Prefecture,Shaoyang City and Loudi City,and high concentration areas are mainly located in Xiangxi Tujia and Miao Autonomous Prefecture and Shaoyang City.The main pathogen of severe cases is EV-A71,and while the proportion of other enterovirus was increasing.The changes of dominant pathogens of hand-foot-and-mouth disease should be closely monitored by relevant institutions,The measures have to be taken in health education for key groups and areas,and in improving the vaccination rate of EV-A71,so as in reducing the occurrence of hand-foot-and-mouth disease.(2)A prediction model for severe diagnosis of patients with hand-foot-and-mouth disease was constructed with age,gender,occupation,address type,feeding type,guardian’s educational level,infected in the month before onset,interval time between onset and first visit,fever,rash/herpes as prediction variables,which has certain prediction and differentiation ability,it can be used as an auxiliary tool for early identification of severe cases.The nomogram shows the prediction value of the probability of severe cases in Hunan Province estimated by the model,which has good practical value.