| Since the H7N7,H9N2,H7N3,H7N2 and H5N1 avian influenza virus affected human society,a new influenza A(H7N9)outbreak in human occurred in March 2013 became an emerging issue for China health and captured global attention.The H7N9 outbreaks in China spread quickly and the mortality rate is very high,which has caused great threat to the health,safety and stability of human society,and has caused great public panic because of repeated outbreaks and spreading.Given the source and the spreading pattern of H7N9 human infections are both uncertain,and considering the current outbreaks faced in China are urgent problems and the advantage of spatial epidemiology,spatial information technology such as GIS and RS,this paper took H7N9 outbreaks occurred in March 2013 to December 2014 in China as the research object,comprehensively used multi-disciplinary theories,methods and technical means such as statistical analysis,spatial information technology and spatial statistical analysis to make spatiotemporal analysis,carried out the research on spatial-temporal distribution,spatial-temporal autocorrelation,the space distribution pattern and space-time aggregation pattern,and to explore the spatial-temporal distribution characteristics and the change tendency of H7N9 human infections,then understand the spread regularity and epidemiological characteristics of the epidemic.And then,based on the propagation characteristics of H7N9,analysed the spatial-temporal autocorrelation and defined the corresponding spatial-temporal factor.Using correlation analysis to explore the environmental factors related to the outbreak.In the end,based on the binary logistic regression model,we constructed an risk prediction model considering both environmental factors,the effects of epidemic situation,also can reflecting the space-time risk propagation characteristics of H7N9 human infections,and used this model to forecast and validate the risk of the epidemic.The main research contents included in this paper are the following aspects:(1)Space-time analysis of H7N9 human infections.Comprehensively use of various spatial information analysis method to make space-time analysis on H7N9 outbreaks in China during the study period,and the study found that: The incidence of epidemic in spring and winter increased obviously than other time,and the peak of H7N9 appeared during March to April in 2013,while which appeared during January to May in 2014,and the east and southeast coastal areas of China are the high incidence regions of the epidemic.On the whole the spatial distribution pattern of the outbreak is spatial autocorrelation,and the results of hot spot analysis and clustering and outlier analysis showed that the high-risk areas in 2013 were located in Jiangsu,Zhejiang and Shanghai,and the high-risk areas in 2014 were located in Jiangsu,Zhejiang,Fujian and Guangdong.(2)The seasonal disease stage study of the epidemic.The results showed that the epidemic consisted of three distinct phases: phase I was from March 13,2013 to May 31,2013(n = 130);phase II was from June 1,2013 to May 31,2014(n = 294);phase III was from June 1,2014 to December 31,2014(n = 36).These results suggested that the epidemic in China has very obvious seasonal epidemic characteristics,and its seasonal fluctuations are evident.(3)The spatial distribution pattern research of H7N9 epidemic.Based on the three distinct epidemic phases confirmed in this paper,the spatial distribution pattern and the directional trend of H7N9 outbreaks was examined by using the standard deviational ellipse analysis in ArcGIS 10.1,and to study the dynamic evolution process of spatial distribution pattern of the epidemic.The results are: The spatial distribution pattern of H7N9 epidemic has experienced a process that which initialized in local areas of east China(phase I)-and then spread along eastern coastal areas to the southeast of China(phase II)-and spread along the southeast coastal areas to central and western regions(phase III).At the same time,the overall spatial distribution pattern of the outbreak has been changed,which shifted from "northwest-southeast" of phase I to "south by west-north by east" of phase II,and then "northwest-southeast" of phase III at last.(4)Study the spatio-temporal clusters of H7N9 epidemic.Based on the three distinct epidemic phases confirmed in this paper,retrospective space-time permutation scan statistics were applied by using SaTScan 9.4.2 to identify the spatio-temporal clusters of H7N9 human infections,and the results showed that,the regions with the most likely spatiotemporal cluster are: April 27,2013-May 11,2013,Fujian,Zhejiang,Hunan,Jiangxi and Taiwan;January 12,2014-January 31,2014,Zhejiang,Shanghai and Jiangsu;April 23,2013-April 30,2013,Fujian,Zhejiang,Jiangxi,Taiwan and Hunan;January 4,2014-January 15,2014,Guangdong;February 20,2014-February 27,2014,Guangdong;April 17,2014-April 24,2014,Jiangsu and Anhui.At the same time,the regions with secondary spatiotemporal clusters are: March 13,2013-April 11,2013,Jiangsu,Anhui and Shanghai;April 22,2014-May 31,2014,Anhui,Jiangsu,Shandong and Jiangxi;February 16,2014-March 2,2014,Guangdong;April 23,2013-April 30,2013,Jiangsu,Shandong and Anhui.And there was no statistically significant spatiotemporal cluster of H7N9 human infections(P < 0.05)identified in phase III.All of these provided evidences that there were significant spatiotemporal clusters during the first two phases.(5)Design and implement the risk prediction model of H7N9 based on environmental factors and spatial-temporal autocorrelation.Binary logistic regression analyses were utilized to investigate the risk factors associated with their distribution,and nine risk factors were significantly associated with the occurrence of A(H7N9)human infections: the spatial-temporal factor ?(OR=2546669.382,P<0.001),migration route(OR=0.993,P<0.01),river(OR=0.861,P<0.001),lake(OR=0.992,P<0.001),road(OR=0.906,P<0.001),railway(OR=0.980,P<0.001),temperature(OR=1.170,P<0.01),precipitation(OR=0.615,P<0.001)and relative humidity(OR=1.337,P<0.001).The improved model obtained a better prediction performance and a higher fitting accuracy than the traditional model.This study can provide a scientific basis for understanding the epidemiological characteristics and dynamic simulating H7N9 outbreaks,forecasting diffusion of H7N9 human infections,reasonably using of medical and health resources and defensing and monitoring the epidemic. |
PDF Full Text Request