Extreme weather conditions and dengue outbreak in Guangdong, China: Spatial heterogeneity based on climate variability

Background: Previous studies have shown associations between local weather factors and dengue incidence in tropical and subtropical regions. However, spatial variability in those associations remains unclear and evidence is scarce regarding the effects of weather extremes. Objectives: We examined spatial variability in the effects of various weather conditions on the unprecedented dengue outbreak in Guangdong province of China in 2014 and explored how city characteristics modify weatherrelated risk. Methods: A Bayesian spatial conditional autoregressive model was used to examine the overall and city-specific associations of dengue incidence with weather conditions including (1) average temperature, temperature variation, and average rainfall; and (2) weather extremes including numbers of days of extremely high temperature and high rainfall (both used 95th percentile as the cut-off). This model was run for cumulative dengue cases during five months from July to November (accounting for 99.8% of all dengue cases). A further analysis based on spatial variability was used to validate the modification effects by economic, demographic and environmental factors. Results: We found a positive association of dengue incidence with average temperature in seven cities (relative risk (RR) range: 1.032 to 1.153), a positive association with average rainfall in seven cities (RR range: 1.237 to 1.974), and a negative association with temperature variation in four cities (RR range: 0.315 to 0.593). There was an overall positive association of dengue incidence with extremely high temperature (RR:1.054, 95% credible interval (CI): 1.016 to 1.094), without evidence of variation across cities, and an overall positive association of dengue with extremely high rainfall (RR:1.505, 95% CI: 1.096 to 2.080), with seven regions having stronger associations (RR range: 1.237 to 1.418). Greater effects of weather conditions appeared to occur in cities with higher economic level, lower green space coverage and lower elevation. Conclusions: Spatially varied effects of weather conditions on dengue outbreaks necessitate area-specific dengue prevention and control measures. Extremes of temperature and rainfall have strong and positive associations with dengue outbreaks.