Identifying and Visualizing Space-Time Clusters of Vector-Borne Diseases

Annually, vector-borne diseases (VBDs) such as malaria, dengue fever, chikungunya, and Zika are responsible for over one billion infections and one million deaths around the world. Many VBDs are showing an expanded range due to climate change, overpopulation, globalization, urbanization, and other factors. As a result, novel outbreaks are occurring in many regions of the world, such as the Americas and the Caribbean. Due to the increased presence of VBDs, it is critical to identify significant disease clusters to facilitate surveillance and intervention strategies with the goal of mitigating future outbreaks. A variety of space-time methods exist to identify significant space-time clusters of VBDs, such as the space-time kernel density estimation and space-time scan statistics. However, the vast majority of the literature solely visualizes the results in two dimensions (2D), simply showing the spatial dimensions of the disease clusters. Visualizing space-time clusters in a three-dimensional (3D) environment can improve the understanding of the spatiotemporal dynamics of a disease outbreak, especially the size, duration, and movement of the clusters during the study period. Furthermore, 2D and 3D visualization methods can complement each other, further improving our understanding of VBD outbreaks. This chapter will discuss some popular space-time methods to identify significant space-time clusters of VBDs and describe popular 2D and 3D visualization techniques. We provide a case study of VBDs in Colombia during outbreaks in 2015 and 2016. The strengths and limitations of each method will also be discussed to shed light on the appropriateness of using each approach.