¹û¶³Ö±²¥ÊÓƵ

Abstract:

In this talk, I will present some of the work I’ve been involved in over the last few years to develop simple statistical tools for the analysis of outbreak data.

I will start by presenting the R package EpiEstim, that I developed a few years ago to estimate transmissibility in real-time during an epidemic. I will demonstrate how I used it, as part of the WHO Ebola Response Team, to help inform the response at different stages of the West African Ebola outbreak. I will show how such simple tools can be used to better understand drivers of local pathogen transmissibility, with applications to pandemic influenza as well as Ebola. Finally I will describe three recent/ongoing extensions of EpiEstim. The first one addresses issues which limit its practical use, namely the possible presence of ‘imported cases’ and the often poorly defined serial intervals. The second one, which was largely developed during the West African Ebola outbreak, is designed to provide, in addition to estimates of pathogen transmissibility, incidence forecasts; I will show how this extended model performed against simulated data during a recent Ebola forecasting challenge. The third extension was developed in combination with phylogenetic methods, to estimate transmissibility of HIV within large clusters of HIV infected individuals in the Netherlands.

In conclusion, I will present recent initiatives to bring together infectious disease modellers, software developers and public health specialists to design better tools in preparation for future outbreaks.