Construction Of The Prediction Model Of HFMD And Screening Of The Risk Factors Of Severe HFMD

Hand,foot and mouth disease(HFMD)is an acute infectious disease mainly caused by enteroviruses.It is an important public health problem that threatens the health of children in China.Therefore,analysis of the prevalence and severe prevalence of HFMD,and prediction of the incidence,prevention and control measures can be taken at the peak of the disease according to the predicted situation.At the same time,screening for the risk factors of severe hand-foot-mouth disease and adopting effective measures for treatment can help reduce the incidence and mortality of critical illness and improve the cure rate.In this study,the disease data collected by Zhengzhou Center for Disease Control and Prevention from 2009 to 2017 were used to construct a predictive model;meanwhile,clinical data of HFMD from Zhengzhou Children’s Hospital from 2015 to 2018 were used to screen risk factors for severe hand,foot and mouth disease.Prediction of hand-foot-mouth disease and screening of severe risk factors can provide an effective basis for epidemic surveillance and early clinical intervention.Objective1.Construct HFMD epidemic prediction model,and provide effective basis for prevention and control of HFMD.2.Screening the risk factors of severe HFMD provides a basis for early prevention and treatment of clinicians.Methods1.The data of HFMD cases in Zhengzhou from 2009 to 2017 were collected,and the data were analyzed by SPSS and R software.After statistics on the number of monthly outbreaks,the data is divided into two parts.The seasonal autoregressive integrated moving average model(SARIMA)and SARIMA-support vector regression(Support Vector Regression(SVR)model to predict the incidence of HFMD in Zhengzhou in 2017.The data of 2017 is used to compare with the prediction data,evaluate the prediction effect of the model,and select the best prediction model.2.The clinical data of HFMD patients from Zhengzhou Children’s Hospital from 2015 to 2018 were collected.Data were processed using SPSS software.The multivariate logistic regression model was used to screen the risk factors of severe hand,foot and mouth disease,and a prediction model was constructed.The receiver operating characteristic curve(ROC)was used to evaluate the prediction effect of the model.Results1.The SARIMA(2,1,2)(2,1,1)12 model constructed from the incidence of HFMD in Zhengzhou from 2009 to 2016.This model predicts the incidence of HFMD in Zhengzhou in 2017.The actual values are compared,MAE is 327,RMSE is 451,and MAPE is 24.63%;the constructed SARIMA-SVR model is used to predict the number of HFMD in Zhengzhou in 2017,and the predicted value is compared with the actual value.MAE is 97,and RMSE is 164,MAPE is 9.13%.2.Age<3 years(OR=1.266,95CI:1.146～1.399),fever duration>3 days(OR=2.321,95CI:1.856～2.879),tremor(OR=6.503,95CI:5.259～8.042),Limb shaking(OR=3.660,95CI:2.538～5.279),neutrophil ratio(>75%)(OR=1.324,95CI:1.103～1.539)are important risk factors for severe HFMD.The area under the ROC curve(AUC)of the prediction model was 0.826(0.806～0.846),the sensitivity was 0.776,and the specificity was 0.778.Cross-validation of the model,the prediction accuracy of the model is 78.21%.Conclusions1.Constructed model SARIMA(2,1,2)(2,1,1)12.Both SARIMA-SVR models can be used to predict the incidence of HFMD in Zhengzhou.The prediction effect of SARIMA-SVR model is better than that of model SARIMA(2,1,2)(2,1,1)12.The SARIMA-SVR model is more suitable for predicting the incidence of HFMD in Zhengzhou.2.HFMD patients aged<3 years,fever duration>3 days,tremor,limb tremor,and neutrophil proportion(>75%)are prone to develop severe illness,and the severe HFMD model constructed has a better prediction effect.