A web-based geospatial toolkit for the monitoring of dengue fever

The rapid propagation of vector-borne diseases, such as dengue fever, poses a threat to vulnerable populations, especially those in tropical regions. Prompt space-time analyses are critical elements for accurate outbreak detection and mitigation purposes. Open access web-based geospatial tools are particularly critical in developing countries lacking GIS software and expertise. Currently, online geospatial tools for the monitoring of surveillance data are confined to the mapping of aggregated data. In this paper, we present a web-based geospatial toolkit with a user-friendly interactive interface for the monitoring of dengue fever outbreaks, in space and time. Our geospatial toolkit is designed around the integration of (1) a spatial data management module in which epidemiologists upload spatio-temporal explicit data, (2) an analytical module running an accelerated Kernel Density Estimation (KDE) to map the outbreaks of dengue fever, (3) a spatial database module to extract pairs of disease events close in space and time and (4) a GIS mapping module to visualize space-time linkages of pairs of disease events. We illustrate our approach on a set of dengue fever cases which occurred in Cali (659 geocoded cases), an urban environment in Colombia. Results indicate that dengue fever cases are significantly clustered, but the degree of intensity varies across the city. The design and implementation of the on-line toolkit underscores the benefits of the approach to monitor vector-borne disease outbreaks in a timely manner and at different scales, facilitating the appropriate allocation of resources. The toolkit is designed collaboratively with health epidemiologists and is portable for other surveillance data at the individual level such as crime or traffic accidents.