Projections of epidemic transmission and estimation of vaccination impact during an ongoing Ebola virus disease outbreak in Northeastern Democratic Republic of Congo, as of Feb. 25, 2019

Worden, Lee; Wannier, Rae; Hoff, Nicole A.; Musene, Kamy; Selo, Bernice; Mossoko, Mathias; Okitolonda-Wemakoy, Emile; Muyembe Tamfum, Jean Jacques; Rutherford, George W.; Lietman, Thomas M.; Rimoin, Anne W.; Porco, Travis C.; Kelly, J. Daniel

When we started performing our short-term forecasts on August 20, 2018, there were 102 reported EVD cases in North Kivu and Ituri provinces, DRC. We used our stochastic and auto-regression models to project one-week, two-week, four-week, and eight-week forecasts of outbreak size. As time lapsed, we compared predicted and known outbreak sizes, and found a higher probability of accurate forecasts at one week than at eight weeks (Figs 2 and 3) . Loglikelihood scores typically declined as projection times extended further into the future. These lower-scoring longer-term projections tended to include wider prediction intervals, reflecting less certain outcomes in which less probability was assigned to any one value. The epidemic curve accelerated in early October, and stochastic model forecasts occurring just before that change were especially low scoring, as they failed to anticipate the coming rise in case counts. Subsequent projections took into account the reported acceleration and their performance recovered.