| Objective:This study aimed to investigate the pattern of outbreaks,spreading trend and spatio-temporal clustering of human infections with H7N9 in Guangdong Province from 2013 to 2017.Additionally,we explored the environmental factors related to the epidemics,and quantify their impact.This will provide guidance on the prediction and warning and risk assessment of epidemics so as to provide a targeted basis for early prevention and control strategies to reduce the occurrence and spread of future epidemics.Method:The information of human H7N9 cases in Guangdong Province between June 2013 and May 2017 was collected through a national system for reporting of notifiable infectious diseases.Based on geographical information system to study the spatiotemporal distribution characteristic,the spatial distribution structure and spatiotemporal clustering of H7N9 outbreaks.Besides,the risk of A(H7N9)outbreaks from 2013 to 2017 during A(H7N9)seasonal epidemics in Guangdong whether would occur was considered as a binary outcome indicator,the environmental contamination data of H7 subtype avian influenza virus in the poultry market and the meteorological data prior to the weekly epidemic were selected to use as predictors,the receiver operating characteristic(ROC)curves were used to analyze the critical values of the predictor variables.In addition,taking the weekly number of human H7N9 cases as the response variable,the relationship between environmental factors and human H7N9 cases was analyzed using Spearman’s ranked correlation coefficient,and we further use boosted regression tree(BRT)model to quantitatively analyze the relative contribution of relevant environmental factors during the seasonal epidemics.Results:(1)As of May 2017,a total of 258 laboratory confirmed A(H7N9)cases were reported in Guangdong Province.According to the global autocorrelation analysis,the global Moran’s I index of the human infections with H7N9 in Guangdong Province from 2013 to 2017 was 0.236(Z = 4.089,P <0.001),suggesting that the overall epidemic situation was spatial aggregation.The results of standard deviation ellipse analysis showed that the spatial distribution structure of H7N9 outbreaks originated from the Pearl River Delta region,spread along the Pearl River Delta region to the northeast of Guangdong Province and finally expanded back to the Pearl River Delta region.The retrospective space-time permutation scan statistic also found two significant temporal and spatial aggregation areas(P <0.05).(2)The area under the ROC curves(AUC)of the environmental factors ranged from 0.666 to 0.817(P<0.001)in the week before the outbreak,meaning that both of them had a certain accuracy in distinguishing the risk of A(H7N9)outbreaks.While the market positive rate of H7 subtype avian influenza virus was 15.10%,the sample positive rate of H7 subtype avian influenza virus was 2.24%,the average precipitation was 79.98 mm,the average pressure was 1011.11 hPa,the average relative humidity was 75.65%,the average temperature was 17.30°C,and the average temperature difference was 8.22°C in the week before the epidemic,they were the critical values for the risk of A(H7N9)outbreaks in the coming week.The total compliance rate ranged from 65.63% to 76.56%.(3)The results of BRT analysis showed that the correlation of training data was 0.812,the correlation of cross validation was 0.672,the standard error was 0.051.In addition,the standard deviation of the simulated value and the measured value was 4.266%,and the standardization residual was 1.276%.AUC was 0.886,suggesting the model fitting effect was better.The contribution rate of environmental factors during the seasonal epidemics in Guangdong Province was respectively the market positive rate of H7 subtype avian influenza virus(28.27%),the mean temperature difference(27.09%),The mean temperature(22.07%),sample positive rate of H7 subtype avian influenza virus(11.01%),mean relative humidity(5.48%),mean pressure(4.17%)and mean cumulative precipitation(1.90%).Among the environmental factors contributing more than 5% to the weekly number of A(H7N9)cases,the weekly number of A(H7N9)cases was positively correlated with the market positive rate,the mean temperature difference and the sample positive rate,and negatively correlated with the mean temperature and the mean relative humidity.Conclusion:Market positive rate of H7 subtype avian influenza virus,samples positive rate of H7 subtype avian influenza virus,temperature and humidity and other environmental factors in the week play an important early warning role in the spread of H7N9 virus,which can be used for short-term forecasting of the risk and size of the outbreak in the coming week. |
PDF Full Text Request