| Objective:By analyzing the epidemiological characteristics and spatial distribution characteristics of bacterial dysentery in China from 2005 to 2016,a spatial regression model and a morbidity prediction model were established to explore the spatial aggregation of bacillary dysentery,in order to further develop effective population prevention measures and optimize public health resource allocation.Provide a scientific theoretical basis.Methods:1.Application Excel 2016 and SPSS 24.0 software for data entry,collation and analysis,describing the epidemic characteristics of bacterial dysentery.2.Bayesian smoothing of the incidence of bacterial dysentery from 2005 to 2016was performed using ArcGIS 10.2 and GeoDA software,and spatial autocorrelation analysis was performed on the incidence of smoothing to study the spatial aggregation of bacterial dysentery.Time-scan,spatial scan and spatio-temporal scan statistic analysis of the incidence of bacterial dysentery from 2005 to 2016 were performed using SaTScan9.6 software to analyze the spatial distribution characteristics of bacterial dysentery.3.GeoDA software was used to establish a spatial regression model for the incidence of bacterial dysentery in 2008-2016 and related climatic and socioeconomic factors,and to study the relationship between the incidence of bacillary dysentery and climate and socio-economic factors.4.Using MATLAB to establish a gray prediction model to fit the annual incidence of bacterial dysentery in 2005-2016,and predict the trend of incidence in 2017-2021.Results:1.In 2005-2016,3108293 cases of bacterial dysentery were reported nationwide,and the average annual incidence was 259,024.42 cases,with an average annual incidence rate of 19.27/100,000.The age group with the highest incidence rate is under5 years old,and the main epidemic month is from May to October.In 2005-2016,there were 31 provinces/autonomous regions/municipalities in the country,and the average annual incidence rate of the top five provinces/autonomous regions/municipalities were Beijing,Tianjin,Gansu,Tibet and Ningxia Hui Autonomous Region.2.The number of cases of bacterial dysentery in the country declined linearly from2005 to 2016.The regression coefficient is b=-2.97,F=252.24,P<0.001,and the regression equation is y=-2.97x+5988.86.The incidence of bacterial dysentery in the country declined linearly from 2005 to 2016.The regression coefficient is b=-2.31,F=237.63,P<0.001,and the regression equation is y=-2.31x+4661.71.3.Global spatial autocorrelation analysis showed that the Moran's I index of the incidence of bacterial dysentery in 2005-2016 was 0.3668,0.4091,0.3988,0.4139,0.4254,0.4211,0.4382,0.4518,0.4589,0.4709,0.4548 and 0.4463,respectively.The test was statistically significant(P<0.05).4.The results of local spatial autocorrelation analysis show that the distribution of“High-High”in 2005-2016 is Tianjin and Beijing,and the distribution of“Low-Low”is different,mainly in Guangdong,Jiangsu and Jiangxi provinces.Waiting for the eastern coastal areas.All tests were statistically significant(P<0.05).5.Time scan statistic analysis showed that the cumulative time of bacterial dysentery in the country from 2005 to 2016 was from May to October(RR>1,P=0.001).6.Spatial scan statistics analysis showed that the first-level accumulation areas of bacterial dysentery in China from 2005 to 2016 were Beijing and Tianjin,and the RR values were 6.22,6.49,7.57,5.93,6.12,5.87,4.99,4.42.4.50,4.91,5.30,and 5.84,all of which were statistically significant(P<0.05).The secondary gathering areas are concentrated in the western regions such as Xinjiang Uygur Autonomous Region,Qinghai Province,and Ningxia Hui Autonomous Region.7.Spatio-temporal scan statistic analysis,scanning in years as the time unit,the results show that the first-level gathering time is from January 2005 to December 2010,and the first-level gathering area is 15 provinces/autonomous regions/municipalities in western China(RR=2.66,P<0.001);secondary aggregation time was from January2005 to December 2016,and the secondary aggregation area was 7 provinces/autonomous regions/municipalities in the southeast coastal areas of China(RR=1.36,P<0.001).8.Spatio-temporal scan statistic analysis,scanning in months as the time unit,the aggregation time is from May to October.In 2007,the first-level gathering areas were Beijing and Tianjin.The first-level gathering areas in 2005-2006 and 2008-2016 were concentrated in the western regions of Xinjiang Uygur Autonomous Region,Qinghai Province,Tibet Autonomous Region,Gansu Province(RR>1,P<0.001);The secondary gathering areas in 2005 and 2006 are concentrated in the eastern coastal areas of Zhejiang Province and Shanghai;in 2007,the secondary gathering areas are concentrated in the southwestern part of China such as Tibet Autonomous Region and Yunnan Province;The district is concentrated in Jiangxi,Hubei,Hunan and Anhui provinces(RR>1,P<0.001).9.The Lagrange multiplier spatial dependence test showed that LM-lag was statistically significant(P<0.05)in 2008-2016,and both were less than the P value of LM-error.The data of bacterial dysentery in 2008-2016 was suitable for spatial lag model.The R~2 of the SLM model for the incidence of bacterial dysentery in 2008-2016was larger than that of the OLS model.The absolute values of AIC,SC and LIK were smaller than those of the OLS model,and the spatial lag model fitting effect was better than the ordinary least squares regression model.10.The results of influencing factors analysis of spatial lag model showed that the incidence of bacterial dysentery in China was positively correlated with the health technicians per 1,000 population in 2008-2011(?>0,P<0.05);in 2013,the incidence of bacterial dysentery in China was positively correlated with the average temperature(?=0.11,P=0.04),and negatively correlated with precipitation(?<0,P=0.02);in 2014and 2015,the incidence of bacterial dysentery in China was negatively correlated with the beds of medical and health institutions(?=-0.05,P=0.02),the incidence of bacterial dysentery in China was positively correlated with mean temperature(?=0.13,P=0.03)in 2016,and negatively correlated with health technicians per thousand population(?<0,P=0.04),and precipitation Negative correlation(?<0,P=0.05).11.The posterior difference of the gray GM(1,1)model is C=0.08,and the small error probability is P=1.The model prediction accuracy is excellent.Extrapolation predicts that the incidence of bacterial dysentery in the country from 2017 to 2021 are9.96/100,000,7.03/100,000,6.21/100,000,5.48/100,000 and 4.84/100,000 respectively.Conclusion:1.The incidence of bacillary dysentery decreased year by year in 2005-2016,showing a distinct seasonal distribution,with high incidence months from May to October;the age of onset was mainly distributed in children under 5 years old,followed by children aged 5-9 years.2.In 2005-2016,bacterial dysentery was concentrated in the country,rather than randomly distributed;the high-concentration areas were in Beijing,Tianjin the western regions,and the low-concentration areas were in the eastern coastal areas.3.The spatial lag model is better than the ordinary least squares regression model for fitting the incidence data of bacterial dysentery in China.4.The main factors affecting the incidence of bacterial diarrhea are average temperature,precipitation,health technicians per 1,000 population and beds in medical institutions.5.The gray GM(1,1)model predicts the incidence of bacterial dysentery better,and the incidence of bacterial dysentery in the country will continue to decline in 2017-2021. |
PDF Full Text Request