Weather as an effective predictor for occurrence of dengue fever in Taiwan

Abstract

We evaluated the impacts of weather variability on the occurrence of dengue fever in a major metropolitan city, Kaohsiung, in southern Taiwan using time-series analysis. Autoregressive integrated moving average (ARIMA) models showed that the incidence of dengue fever was negatively associated with monthly temperature deviation (beta=-0.126, p=0.044), and a reverse association was also found with relative humidity (beta=-0.025, p=0.048). Both factors were observed to present their most prominent effects at a time lag of 2 months. Meanwhile, vector density record, a conventional approach often applied as a predictor for outbreak, did not appear to be a good one for diseases occurrence. Weather variability was identified as a meaningful and significant indicator for the increasing occurrence of dengue fever in this study, and it might be feasible to be adopted for predicting the influences of rising average temperature on the occurrence of infectious diseases of such kind at a city level. Further studies should take into account variations of socio-ecological changes and disease transmission patterns to better propose the increasing risk for infectious disease outbreak by applying the conveniently accumulated information of weather variability.