Study On Spatial-Temporal Distribution And Risk Factors Of Hemorrhagic Fever With Renal Syndrome In Changsha Based On GIS

Hemorrhagic fever with renal syndrome (HFRS), caused by Hantaviruses (HV), is characterized by fever, acute renal dysfunction and hemorrhage manifestations. It is one of the key infectious diseases harming people’s health and safety with wide distribution, high incidence and complicated type of endemic areas in China’s mainland. Traditional HFRS studies were mostly based on classical statistical analysis, ignoring the spatial information. Using Geographic Information System in epidemiological studies, the temporal and spatial distribution of HFRS and spatial risk factors could be studied quantitatively. Thus, prevention and control measures could be formulated more scientifically.With the help of GIS and basing on the cases of HFRS of recent years in Changsha, this study attempts to characterize the temporal and spatial distribution of HFRS in Changsha, and to find the risk factor of HFRS from the following three aspects:(1) Temporal distribution analysis:To forecast of the incidence of HFRS from January to December of 2009 with the method of Time Series and the time series model of HFRS from the incidence of HFRS from 2005 to 2008. The result of the forecaster is very close to the actual data and the model fits.(2) Spatial distribution analysis:using spatial cluster analysis and spatial autocorrelation analysis, we investigated the spatial distribution of HFRS of each town in Changsha. The result of the spatial cluster analysis shows that HFRS clusters in Changsha. It involves 23 towns (relative risk, RR=15.39, P=0.001). Spatial autocorrelation analysis shows that HFRS cases in Changsha are spatially autocorrelated. That is, the distribution of the high incidence of HFRS and the high incidence are adjacent. The hot spot areas in Changsha City are defined. Studies show that the hot spots of HFRS and spatial cluster areas are consistent with each other. That is, the result of the spatial cluster analysis is the same with the scanning of spatial autocorrelation analysis.(3) Risk factor analysis:The logistic regression model for the study of the relationship between HFRS and risk factors is established by employing principal component analysis and removing the influence of multivariable multicollinearity. The results show that the risk factors of HFRS include the average elevation, average temperature, cultivated land, forest land, construction land and water land. HFRS is lower in areas in higher attitude, areas with higher average temperature, and areas have large areas of forest while it is easier for the HFRS to spread in Water, farming lands and places where there are constructing buildings.In short, with the accuracy prediction, the method to predict the future incidence of HFRS was provided. HFRS of Changsha was highly clustered in the space. The probability of occurrence of HFRS was affected by risk factors in varying degree. The results are important in the clear understanding of the prevalence of HFRS in Changsha City. The method and means was provided for future similar epidemics study by the study.