Using Digital Surveillance Tools for Near Real-Time Mapping of the Risk of International Infectious Disease Spread: Ebola as a Case Study

Sangeeta Bhatia; Britta Lassmann; Emily Cohn; Malwina Carrion; Moritz U.G. Kraemer; Mark Herringer; John Brownstein; Larry Madoff; Anne Cori; Pierre Nouvellet

ods for details). The raw and processed data from ProMED and HealthMap for all countries included in the feed are available in the Github repository for this article. There were substantial differences between the incidence time series derived from the three data sources, particularly at the peak of the epidemic. There may be multiple reasons underpinning such discrepancy, including potential variability in digital surveillance reporting during the course of the epidemic. It is also worth highlighting that the WHO data used here are an extensively cleaned version of the data collected during the epidemic [17, 18] , published more than a year after the epidemic was declared to be over. Moreover, while ProMED and HealthMap did not record probable cases for this epidemic, the WHO data contained confirmed, probable and suspected cases. However, despite these discrepancies, the weekly incidence derived from ProMED and HealthMap was moderately to highly correlated with that reported by WHO later (Pearson's correlation coefficients aggregated across the three countries 0.44 and 0.74 respectively, p value < 0.001, also see Fig 2 for weekly trends).