Pairwise-like models for non-Markovian epidemics on networks

TL;DR

This paper extends pairwise epidemic models to non-Markovian cases on networks, using delay differential equations and analytical methods to accurately predict epidemic thresholds and final sizes.

Contribution

It introduces a generalized pairwise model for non-Markovian epidemics, including delay differential equations and analytical expressions for thresholds and final sizes.

Findings

Excellent agreement with stochastic simulations

Derived a new R0-like threshold quantity

Provided a closed-form expression for final epidemic size

Abstract

In this letter, a generalization of pairwise models to non-Markovian epidemics on networks is presented. For the case of infectious periods of fixed length, the resulting pairwise model is a system of delay differential equations (DDEs), which shows excellent agreement with results based on explicit stochastic simulations of non-Markovian epidemics on networks. Furthermore, we analytically compute a new R0-like threshold quantity and an implicit analytical relation between this and the final epidemic size. In addition we show that the pairwise model and the analytic calculations can be generalized in terms of integro-differential equations to any distribution of the infectious period, and we illustrate this by presenting a closed form expression for the final epidemic size. By showing the rigorous mathematical link between non-Markovian network epidemics and pairwise DDEs, we provide the framework for a deeper and more rigorous understanding of the impact of non-Markovian dynamics with explicit results for final epidemic size and threshold quantities.