Advertisement

< Back to Article

Anticipating epidemic transitions with imperfect data

Fig 1

Demonstration of aggregation effects in epidemiological data.

The bottom panel shows the progression of a simulated outbreak in a population, with cases ranked by their time of infection. Solid black lines indicate the duration of infectiousness, dots indicate time of recovery. The top panel shows three time series calculated from the simulated data: daily snapshots of the number of infected present in the population (black), weekly case reports (blue) and monthly case reports (red). For the purposes of this paper, the number of recovery events falling within an aggregation period serves as a proxy for the true number of cases in a case report. Aggregation periods are delimited by blue dots for weekly reports and red dashes for monthly reports. No reporting error is applied to the case reports shown in this figure. Transmission dynamics are modeled using the SIR model with birth and death with average population size N0 = 106, importation rate ΞΆ = 1 case per week, mean life expectancy of 70 years, and mean infectious period of 1 week. R0 increases linearly from 0 to 1 over 20 years. Simulations performed using the Gillespie algorithm [33].

Fig 1

doi: https://doi.org/10.1371/journal.pcbi.1006204.g001