Unreported cases in the 2014-2016 Ebola epidemic: Spatiotemporal variation, and implications for estimating transmission

摘要

Author summary Epidemics are defined by a surge of cases of a disease, yet often a significant number of cases in an epidemic are never reported, for example because not all infected individuals have access to medical care. This underreporting can introduce bias into analyses of disease spread, by distorting patterns in where and when the most cases are observed. Conversely, quantifying underreporting can improve epidemic forecasts and containment strategies. In this study, we analyze data from the recent Ebola epidemic in West Africa, including the time, location and Ebola status of 6491 individual community burials, conducted over 25 weeks in four districts in Sierra Leone. We quantify how reporting rates varied over space and time, and show that estimates of transmission rates that are corrected for dynamic underreporting diverge significantly from uncorrected estimates, particularly earlier in the epidemic and outside the capital.