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

In this supplement, we present the details of the pre-processing of ProMED and HealthMap feeds (Section 1), the model results using ProMED, HealthMap and WHO data and the impact of the datasources and model parameters on the performance of the model. We varied the length of the time window used for model fitting (see Methods for details). SI Sections 2, SI Section 3 and SI Section 4 present the forecasts using ProMED, HealthMap and WHO data respectively using different calibration windows (2, 4 and 6 weeks) and forecast horizons (4, 6 and 8 weeks). The model performance was moderately better with shorter calibration windows ( Fig 32) and with WHO data (Fig 33) . We explored the impact of alternate priors for the gravity model parameter on the results from the model. The results of this sensitivity analysis are presented in Section 8. We present additional analysis on the predicting the spatial spread of the epidemic in Section 9.