Spatio-temporal detection for dengue outbreaks in the Central Region of Malaysia using climatic drivers at mesoscale and synoptic scale

Abstract

AbstractThe disease dengue is associated with both mesoscale and synoptic scale meteorology. Previous studies for south-east Asia have found very limited association between synoptic variables and the reported dengue cases. It will immensely beneficial to establish more clear association with rate of cases and the most relevant meteorological variables in order to institute an early warning system.A rigorous Bayesian modelling framework is provided to identify the most important co-variates and their lagged effects for developing an early warning system in the Central Region of Malaysia.Along with other mesoscale environmental measurements, we also examine multiple synoptic scale Niño indices which are related to the phenomenon of El Niño Southern Oscillation and an unobserved variable derived from reanalysis data. A probabilistic early warning system is built based on a Bayesian spatio-temporal hierarchical model.Our study finds a 46.87% of increase in dengue cases due to one degree increase in the central equatorial Pacific sea surface temperature with a lag time of six weeks. We discover the existence of a mild association between the rate of cases and a distant lagged cooling effect related to a phenomenon called El Niño Modoki. These associations are assessed by using a Bayesian spatio-temporal model in terms of estimated out-of-sample predictive accuracy.With the novel early warning system presented, our results show that the synoptic meteorological drivers can enhance short-term detection of dengue outbreaks and these can also potentially be used to provide longer-term forecasts.Practical ImplicationsIn 2019, it was reported that there is a severe dengue upsurge in Malaysia. Reported cases rose over 60% from 80,615 in the 2018 to 130,101 with 182 deaths (Rahim et al., 2021). The disease has been described as a silent killer that the infection rate once surpassed that of COVID-19. There is a need of an early warning system to alert the authority in order to identify relevant risk factors and the forthcoming outbreak hot-spots. The Bayesian hierarchical spatial dynamic model componentises different aspects of dengue dynamics into one unified model. Its flexibility allows us to regularly review the disease dynamic under changing environment and transmission mechanism such as the implementation of the Movement Control Orders (MCO) during COVID-19. Practically, this prototype model should be run at least once a week to generate forecasts which is fed with the dengue cases data from weekly press release and meteorological information from publicly available sources. By assessing the probability estimates, the alert has its intrinsic meaning and the sensitivity can be adjusted effortlessly.HighlightsEl Niño Southern Oscillation is a crucial driver to dengue outbreaks in Malaysia.A few different climate oscillations affect the dengue transmission pattern.Bayesian spatial dynamic model helps the development of early warning system.The model components can be added or modified under the hierarchical Bayes framework.