Regional response of dengue fever epidemics to interannual variation and related climate variability

Dengue is a major international public health concern and one of the most important vector-borne diseases. The purpose of this article is to investigate the association among temperature, rainfall, relative humidity, and dengue fever by incorporating the lag effect and examining the dominant interannual model of the modern climate, the El Niño Southern Oscillation (ENSO), in the southern region of Taiwan. We built a linear Poisson regression model by including linear time treads and statistical indicators, verified with disease data in the 2004–2013 period. Here we showed that regional climatic factors in association with the interannual climate variability expressed by the ENSO phenomenon had a significant influence on the dynamics of urban dengue fever in southern Taiwan. The 2–4-month lag of statistical indicators of regional climate factors together with the 4-month lagged Pacific surface temperature (SST) anomaly in the proposed Poisson regression model could capture the regional dengue incidence patterns well. The statistical indicators of mean and coefficient of variation of temperature showed the greatest impact on the dengue incidence rate. We also found that the dengue incidence rate increased significantly with the lag effect of the warmer SST. The ability to forecast regional dengue incidence in southern Taiwan could permit pretreatment of mosquito habitats adjacent to human habitations with highly effective insecticides that would be released at the time of the high-temperature season.