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

is the (which was not peer-reviewed) The copyright holder for this preprint . https: //doi.org/10.1101/19011940 doi: medRxiv preprint estimates of the parameters over the course of the epidemic using the alternative priors. As the epidemic progressed, the parameter p stay assumed larger values suggesting a decreased probability of travel over time (Figures 34, 35) . As p stay assumes large values, the estimated flow is more strongly influenced by p stay than by γ. Furthermore, p stay is likely to depend on the spatial scale of the model. Our analyses were carried out at the national scale; we expect that γ will be more sensitive to p stay at a finer spatial resolution. Overall, the flow between locations using the parameters estimated using the two alternative priors did not vary much (Fig 36) . Figures 37 to 45 present the model forecasts using the alternative priors for γ and Figure 46 presents a comparison of model performance metrics using the two priors. Although the analysis was carried out for the three data sources (ProMED, HealthMap and WHO), for brevity we present results using ProMED data only. Here γ is allowed to vary between 1 and 10. p stay represents the probability of an individual to stay in a given location during their infectious period. The solid lines represents the median estimates obtained using ProMED (blue), HealthMap (green) and WHO (yellow) data. The shaded regions represent the 95% CrI.