Spatiotemporal Patterns And Influencing Factors Of Dengue Fever In Typical Regions Of The World

Dengue fever has now become a major public health problem,causing high attention in all countries.Nowadays,GIS has been applied in the field of health,showing its superior performance in many fields such as disease risk estimation and disease feature mapping.In this paper,the data of dengue virus occurrence location in the world from2000 to 2014 were used as the research object.First,the global Moran’I and local Getis G* were used to summarize the statistical characteristics of dengue risk in each study area.Then,the Bayesian spatio-temporal hierarchical model(BSTHM)was used to analyze the spatiotemporal heterogeneity of dengue risk in the study area.At the same time,combined with meteorological data and social environmental data in the same period,the Geodetector method was used to quantify the explanatory power of the selected factors and their interactions on the risk heterogeneity of dengue fever in different study areas,and robust influencing factors were screened out.Finally,based on the results of BSTHM stratification in the study area,the generalized additive model(GAM)was used to analyze the differences in risk factors of dengue fever among different stratifications.The main conclusions of this study are as follows:(1)Over the past 15 years,dengue has shifted to a more clustered pattern in the Southeast Asian study region,with the highest concentration occurring in 2007.Dengue risk is concentrated around Thailand and Cambodia.The spatial relative risk of dengue fever in the study area showed an obvious pattern of low in the north and high in the south,with the highest value in Rayong Province,Thailand.Other areas(neither cold spot nor hot spot)accounted for the highest proportion in the risk category,and stable areas accounted for the highest proportion in the trend category.In addition,the overall temporal relative risk gap between the study areas was obvious,and the incidence risk was the highest in 2009.Overall mean maximum temperature explained the highest variation in dengue risk in the study area,at 41 percent.The interaction between the factors is nonlinear enhancement.In the attribute stratification of risk category and trend category,the significant influencing factors of the stratification region with the highest proportion are the total amount of regional precipitation and regional average precipitation.(2)Over the past 15 years,the incidence of dengue in the North American study region has changed to a more concentrated pattern,with the highest concentration in2012,and the risk centers of dengue were concentrated around Honduras and El Salvador.The spatial relative risk of dengue fever was high and concentrated in the southern part of the study area.In the risk category,hotspots accounted for the highest proportion,while stable areas accounted for the highest proportion in the trend category.The overall relative risk in the study area increased steadily during the 15 years,and the incidence risk was the highest in 2013.The explanatory power of the influencing factors to the change of dengue risk was low,and the explanation of the total precipitation was the highest,which was 1.7%.The interaction between the factors was of two types: nonlinear enhancement and double-factor enhancement.In the attribute stratification of risk category and trend category,the significant influencing factors in the stratification area with the highest proportion were average maximum temperature,wetland proportion and grassland proportion.(3)From 2000 to 2014,the incidence of dengue fever in the South American study area changed to a more concentrated pattern,with the highest concentration in2006,and the risk center of dengue fever was concentrated in the middle of the study area.The spatial relative risk of dengue fever in the middle of the study area was higher than that at both ends.In the risk category,hotspots accounted for the highest proportion,while stable areas accounted for the highest proportion in the trend category.The overall relative risk in the study area increased steadily during the 15 years,and the incidence risk was the highest in 2013.On the whole,population density had the highest explanatory power for the change of dengue fever in the study area,which was 10%.The interaction between the factors was of two types: nonlinear enhancement and double-factor enhancement.The average maximum temperature,regional average precipitation,the proportion of forest land,the proportion of agricultural land and the proportion of other land are the significant influencing factors of the main attribute stratification in the study area.