Quantifying the transmission potential of pandemic influenza

Abstract

This article reviews quantitative methods to estimate the basic reproduction number of pandemic influenza, a key threshold quantity to help determine the intensity of interventions required to control the disease. Although it is difficult to assess the transmission potential of a probable future pandemic, historical epidemiologic data is readily available from previous pandemics, and as a reference quantity for future pandemic planning, mathematical and statistical analyses of historical data are crucial. In particular, because many historical records tend to document only the temporal distribution of cases or deaths (i.e. epidemic curve), our review focuses on methods to maximize the utility of time-evolution data and to clarify the detailed mechanisms of the spread of influenza.First, we highlight structured epidemic models and their parameter estimation method which can quantify the detailed disease dynamics including those we cannot observe directly. Duration-structured epidemic systems are subsequently presented, offering firm understanding of the definition of the basic and effective reproduction numbers. When the initial growth phase of an epidemic is investigated, the distribution of the generation time is key statistical information to appropriately estimate the transmission potential using the intrinsic growth rate. Applications of stochastic processes are also highlighted to estimate the transmission potential using similar data. Critically important characteristics of influenza data are subsequently summarized, followed by our conclusions to suggest potential future methodological improvements.2

The role of population heterogeneity and human mobility in the spread of pandemic influenza

“…By looking at past pandemics, we can only make assumptions on its transmissibility potential, which can be summarized by the reproductive number R 0 (essentially, the number of secondary infections that results from a single infectious individual in a fully susceptible population (Anderson & May 1992)). Therefore, according to the recent estimates of the reproductive number for recent influenza pandemics (Mills et al 2004;Ferguson et al 2005;Chowell & Nishiura 2008), plausible transmissibility scenarios on R 0 are drawn: the investigated values range from 1.6 to 2.4.…”

Effective reproduction numbers are commonly overestimated early in a disease outbreak

“…Various methods have been developed for estimating the effective reproduction number from daily incidence data, which are reviewed in Chowell and Nishiura 6. These range from methods based on the exponential growth rate of incidence in the early stages of a disease 7–9 (to cite but a few of the many), fitting a SIR (Susceptible, Infective, Recovered) ordinary differential equation model curve 10, likelihood‐based estimation methods 11–13 and stochastic SIR models with Bayesian estimation 10, 14–16.…”

Modelling the initial phase of an epidemic using incidence and infection network data: 2009 H1N1 pandemic in Israel as a case study

“…Modelling the initial phase of epidemics G. Katriel et al 859 distribution and stochasticity, it offers a very attractive and more direct alternative to the widely applied approach for estimating R e based on estimating the exponential growth rate r of the epidemic curve and relating r to R e using the generation-interval distribution [16].…”