Geospatial-temporal Analysis of Dengue Fever Based on the Bayesian Spatiotemporal Model

Dengue fever (DF) is one of the most rapidly spreading mosquito-borne viral diseases in the world.It can impose an enormous socioeconomic and disease burden on the world's population.The spatial and temporal distribution patterns of DF cases in China are heterogeneous.It is evident that further research is necessary to identify the high-risk areas of dengue occurrence and associated risk factors at a fine spatiotemporal scale.This will facilitate the prevention and control of DF transmission.With the rapid development of remote sensing (RS) and geographic information system (GIS) technology, RS and GIS technology have played an important role in monitoring, forecasting, and identifying influencing factors and formulating prevention and control strategies for DF.In China, the majority of dengue cases were clustered in Guangdong and Yunnan Provinces.In this study, Bayesian spatiotemporal models were fitted at the county level in order to quantify the relationships between environmental and socioeconomic factors and DF.Our findings indicate that both environmental and socioeconomic factors can affect the transmission of DF in Guangdong and Yunnan Provinces.However, the underlying drivers and the spatially clustered patterns observed in the two provinces appear to be different.The results indicate that elevated temperatures facilitate the transmission of DF, but it was found that the risk of DF begins to decrease when the temperature exceeds 27.6 o C in Guangdong Province.There was a positive correlation between temperature and the incidence of DF, although no statistical significance was found in Yunnan Province.In Guangdong Province, the amount of precipitation did not significantly affect the incidence of DF.In Yunnan Province, the relationship between precipitation and DF is nonlinear.As the amount of precipitation increases, the risk of DF epidemics increases.In addition, vegetation cover can significantly affect the DF transmission, but the nonlinear relationships between vegetation cover and DF are diverse and complex in these two provinces.The incidence of DF was found to be positively correlated with the level of urbanization at the county level.The findings may prove beneficial to the governments of both provinces in the development of targeted strategies for the control of DF outbreaks at the county level.