Effect Of Meteorological Factors On Hand-Foot-Mouth Disease And The Application Of Prediction Model In Fangshan District Of Beijing

ObjectiveTo describe the epidemiological status of Hand-Foot-Mouth Disease (HFMD).To examine the association between weekly HFMD cases and meteorological factors. To establish and essess a time-series analysis model to predict the transmission trend of HFMD in Fangshan District of Beijing. The aim of this study is to provide evidence for the prevention and control of HFMD.Methods1、The epidemiological data of HFMD and meteorological data in Fangshan from 2009 to 2013 were collected and analyzed using a descriptive epidemiological method.2、The association between weekly HFMD cases and meteorological factors, such as the average temperature, relative humidity, sunshine, and wind velocity, were analyzed using the generalized additive model(GAM) and time-series analysis, after controlling for long-term trend, seasonality and delayed effects.3、A time-series analysis model was established basing on the monthly incidence of HFMD from 2009 to 2013 in Fangshan District of Beijing. We evaluated the fitting results of actual and predicted values, and used this model to forecast and analysis the transmission trend of HFMD by the data from January to December 2014.4、Soft wares were used including Excel2010, SPSS 18.0 and the library mgcv and TSA in R3.0.1.Results1、A total of 9113 HFMD cases, including 70 serious cases and 1 death were reported during the study period. The average annual incidence was 159.42 per 100,000 population.2、The diseases were mainly reported among children aged 1-5 years old with a proportion 84.09%(7666/9113), and male-to-female ratio was 1.40:1 (5311/3802).3、The scattered and kindergarten’s children were the dominant. The proportion of scattered children was 51.21%(4667/9113) and the proportion of kindergarten’s children was 41.09%(3745/9113)4、The incidence of HFMD had obvious seasonality with large summer peaks observed from May to August, and second smaller peaks observed from October to December,2010-2013.5、423 gathered HFMD cases were reported in Fangshan District. The childcare facilities gathered was dominant, accounting for 90.31%(382/423), involved 2021 cases.6、Enterovirus 71 (EV71) was the most common pathogen in most years expect 2011 and 2013.7、The average temperature(℃), relative humidity(1%), sunshine(h), and wind velocity (m/s)with one week lag were significantly associated with weekly HFMD cases. We found that a 1℃ increase in temperature led to an increase of 1.454% (95%CI:0.508～2.410) in the weekly number of HFMD cases. A one percent increase in relative humidity may lead to an increase of 0.911% (95%CI:0.508-2.410) in the weekly number of HFMD cases. A one hour increase in sunshine may lead to an decrease of 5.517%(95%CI:-8.509--2.426) and a one meter per hour increase in wind velocity may lead to an increase of 19.118%(95%CI:3.909-36.553) in the weekly number of HFMD cases.8、Average temperature(℃), relative humidity(1%), sunshine(h), and wind velocity (m/s) with one week lag were significantly associated with weekly male HFMD cases, but only average temperature(℃) was significantly associated with weekly female HFMD cases. Except wind velocity, other meteorological factors were more significantly associated with weekly kindergarten’s HFMD cases.9、The residuals of SARIMA(0,0,1) (0,1,0)12 model were white noise sequence, which fitted well with the actual values. It was the best model to forecast the monthly incidence of HFMD in Fangshan District.10、The further prediction of the monthly incidences from January to December 2014 by the model indicated that the actual incidences were all in the confidence interal of the predicted incidence. The predicted annual incidence of HFMD in 2014 was 180.99 per 100,000 population. The actual annual incidence of HFMD in 2014 was 243.61 per 100,000 population. The relative error was 25.71%.Conclusion1、The prevention and control of HFMD should be strengthened according to its local epidemiological characteristics.2、Meteorological factors such as the average temperature, relative humidity, sunshine and wind velocity may be important predictors of HFMD transmission in Fangshan District of Beijing.3、Sex and occupation might be important effect modifiers of the effects.4、SARIMA(0,0,1) (0,1,0)12 model can be well applied to predict short-term incidence of HFMD in Fangshan District, which provide a scientific basis for the measures of prevention and control of HFMD.