Spatial and spatiotemporal analysis of dengue virus transmission and Aedes aegypti abundance

Dengue fever and dengue hemorrhagic fever (DHF) are emerging as the most important mosquito borne ailments worldwide. These diseases occur in tropical regions of the world, where more than 2.5 billion people are at risk of infection. In Asia, DHF is a leading cause of hospitalization and death among children. A licensed vaccine for dengue is not yet available. Control of the primary vector, Aedes aegypti, is the only method currently available to prevent dengue virus epidemics. The goal of this research is to understand the ecology of Aedes aegypti and the relationship between vector populations and dengue virus transmission between humans. The dissertation is conducted in conjunction with a study being carried out in Kamphaeng Phet, Thailand. Provincial level DHF incidence and an outbreak in Puerto Rico are also examined. New techniques for spatial and space-time analysis are developed for this purpose. The Incremental Knox Test (IKT) is a method for examining the spatial clustering of temporally lagged cases of infectious disease. An algorithm for finding irregularly shaped clusters, called AMOEBA, is also introduced. The study of Aedes aegypti ecology indicates that containers used to store water for domestic use are important production sites. Results also indicate that mosquito density changes more than proportionally with a change in the density of these containers. Indoor spraying of insecticides is found to be quite effective at reducing mosquito densities, but these reductions are short-lived. Spatiotemporal analysis of a dengue outbreak at the individual level indicates that the minimum interval between successive cases of dengue was approximately 18 days. At this interval, cases cluster significantly within 100m of each other. Comparative clustering analysis of provincial level DHF incidence indicates that the Bangkok area is important in sustaining dengue transmission through the cool dry season when country wide transmission rates are low. This work reiterates that both disease transmission and species abundance are intrinsically spatial processes. Incorporating spatial approaches has generated new findings in the nature of dengue transmission and in Aedes aegypti ecology. The new spatial analysis techniques developed have potential applications in many areas of geographical research.