Study On The Effect Of Meteorological Factors On Bacillary Dysentery Incidence

Objective:The epidemic factors of bacillary dysentery are complex and the incidence of bacillary dysentery is always at a high level.The overuse of antibiotics leads to high resistance or even multiple resistance of shigella to commonly used antibiotics.High incidence of disease and antibiotic resistance make the therapeutic effect of bacillary dysentery decrease.Effectively controlling the epidemic of bacillary dysentery has become one of the public health problems that need to be solved urgently.Through the analysis of the monitoring data onto bacillary dysentery in China,this paper explores the areas with high incidence of bacillary dysentery,establishes a scientific and reasonable prediction model of bacillary dysentery,and provides scientific basis for the prediction and analysis of bacillary dysentery.Second,the relevant published literatures were searched,and different results were quantitatively and comprehensively studied by Meta analysis method to explore the relationship between meteorological factors and the incidence of bacillary dysentery.Lastly,the relationship between meteorological factors and bacterial dysentery in Chaoyang City,Liaoning Province was analyzed using structural equation model?SEM?method.This part aimed to explore the comprehensive influence of meteorological factors,such as sunshine,airflow and humidity,on the incidence of bacillary dysentery,so as to provide the public health departments with relevant suggestions in improving public health consciousness and shigellosis prevention.Methods:This paper collected 13-year data?2004-2016?of bacillary dysentery incidence from the database of public health science data center resource.Hierarchical clustering analysis was applied to identify high-incidence areas.Exponential smoothing method,grey prediction model GM?1,1?,autoregressive moving average model?ARIMA?and Weighted Combination Model were used to predict the cases of bacillary dysentery.And then the best forecast model was identified by comparing the fitting and predicting results of the four models.In the meta-analysis of the relationship between meteorological factors and the cases of bacillary dysentery,this study systematically collected all the literature published before September 1,2019 at home and abroad.A total of 29 papers were included,and the results were quantitatively synthesized.In the last part of this resesrch,the relationship between meteorological factors and the incidence of dysentery was studied by using structural equation modeling.Chaoyang City was the study site.The monthly data of bacillary dysentery incidence from 1981 to 2010 was collected from Chaoyang City Centre for Disease Control and Prevention,and the monthly meteorological data was obtained from the Chaoyang City Meteorological Bureau.After the incidence rate of bacillary dysentery was logarithmic conversion,pearson correlation analysis was used to explore the relationship between meteorological variables and monthly incidence of bacillary dysentery,and SEM was used to study the impact of meteorological variables on the incidence of bacillary dysentery.To establish the hypothetical model,there are four potential meteorological variables:temperature,humidity,sunshine and airflow.Temperature coefficient includes average temperature,maximum temperature,minimum temperature,average ground temperature,maximum ground temperature and minimum ground temperature;humidity coefficient includes average monthly evaporation,absolute humidity,relative humidity,maximum frozen soil depth,barometric pressure,non-precipitation days,maximum precipitation,maximum snow cover thickness and maximum snow cover days;sunshine factors include average monthly sunshine intensity,average day.Illumination and average sunshine hours;airflow factors include average monthly wind speed and direction.This study used the R software factoextra package for clustering analysis,forecast package build Exponential Smoothing Model,greyforecasting package to build the GM?1,1?model,Metrics package,tseries package and forecast package to build ARIMA model,and the meta package for meta-analysis,this study also uses SPSS13.0 software for meteorological factors in factor analysis,test structure validity,using Lisrel 8.5 to construct structural equation model.Results:The bacillary dysentery cases showed a general decreased in China.From 2004 to2016,a total of 3602,639 cases of bacillary dysentery were reported nationwide.The most reported cases were 91,559 cases reported in July 2004,with a incidence of7.0820/100,000,and the least were 5184 cases reported in February 2016,with a incidence of 0.3831/100,000.Bacillary dysentery incidence has an obvious seasonal,which was the highest in July and August.According to the hierarchical cluster analysis,the incidence of bacterial dysentery in 2016 in all provinces and cities can be divided into three groups:the first group with high incidence?Beijing and Tianjin?,the second group with medium incidence?Gansu,Chongqing,Ningxia and Tibet?,and the rest with low incidence.Among the four models Exponential Smoothing Model,GM?1,1?Model,ARIMA model and Weighted Combination Model,RMSE,MER and R² indicated that the ARIMA?1,1,1??2,1,1?₁₂ Model has the highest fitting effect,the Weighted Combination Model is the second,and the Exponential Smoothing Model has the lowest fitting effect.The relative error of ARIMA model is the smallest.ARIMA?1,1,1??2,1,1?₁₂ model were selected to predict the annual incidence of bacterial dysentery in China from 2017 to 2018,which were 106404 and 94530 respectively.In the meta-analysis,29 literatures were selected,including 8 in English and 21 in Chinese,published from 2003 to 2019,and the research time ranged from 1949 to 2016.The research sites were all in China,including temperate zone and subtropical zone.According to STROBE's statement,the 29 studies included in the study were scored ranged from 16 to 20,with a balanced quality.The correlation coefficient r of average temperature with bacterial dysentery incidence?95%CI?was 0.644?0.567,0.710?,the average minimum temperature of 0.557?0.416,0.672?,the average highest temperature,correlation coefficient r of average rainfall and average relative humidity?95%CI?were 0.550?0.407,0.667?,0.401?0.326,0.471?,0.239?0.168,0.307?,and combined effect value were statistically significant.The sensitivity analysis showed stable and reliable results.The structural equation model fitting results of the relationship between meteorological factors and bacterial dysentery was that RMSEA=0.08,GFI=0.84,CFI=0.88 and SRMR=0.06.The value of?² is 231.95?p<0.01?and a degree of freedom of 15,the data of the model fitted well.SEM results showed that all relevant meteorological indicators in this study were divided into three latent variables:temperature,humidity and sunshine.Temperature and humidity were positively correlated with the incidence of bacillary dysentery,and the factor loading values were 0.59 and 0.78,respectively.Sunshine was negatively correlated with the incidence of bacillary dysentery,and the factor load was-0.15.Conclusion:The incidence of bacillary dysentery is on the whole decreasing,and there are obvious seasonal differences.July and August of each year are the peak period of the incidence of bacillary dysentery,but there are obvious regional differences in the incidence of bacillary dysentery.Beijing and Tianjin are the areas of high incidence.Holt-Winters exponential smoothing model,GM?1,1?,ARIMA model and weighted combination model can be used to predict the incidence of bacterial dysentery,among which ARIMA?1,1,1??1,1,2?₁₂2 model has the best fitting and prediction effect.Meteorological factors such as temperature?mean temperature,mean maximum temperature,mean minimum temperature?,sunshine and humidity are associated with the onset of bacillary dysentery.The study included multiple meteorological indicators,which may be better to explore the effect of meterological factors on dysentery than using a single factor.Overall,this study used bacterial dysentery surveillance data to choose the appropriate forecast method to explore bacillary dysentery incidence trends.The Meta analysis quantitatively explored the effect of meteorological factors on bacillary desentery.And Chaoyang city was selected to conduct the research on the effects of meteorological factors of bacillary dysentery.The results of the relationship between meteorological factors and bacillary dysentery would help to evaluate current health policies and develop effective control strategies for bacillary dysentery control.