Effect of time varying transmission rates on coupled dynamics of epidemic and awareness over multiplex network

TL;DR

This study models how time-varying transmission rates influence the coupled spread of epidemics and awareness over a multiplex network, revealing conditions for oscillatory or damped endemic states and the impact of awareness on disease transmission.

Contribution

It introduces a nonlinear stochastic model capturing the co-evolution of epidemic and awareness processes with time-varying transmission rates on multiplex networks.

Findings

Damped and oscillatory dynamics depend on network parameters.

Awareness significantly reduces disease transmission and epidemic size.

Endemic state infection probability depends on transmission rates and node degrees.

Abstract

In the present work, a non-linear stochastic model is presented to study the effect of time variation of transmission rates on the co-evolution of epidemics and its corresponding awareness over a two layered multiplex network. In this model, the infection transmission rate of a given node in the epidemic layer depends upon its awareness probability in the awareness layer. Similarly, the infection information transmission rate of a node in the awareness layer depends upon its infection probability in the epidemic layer. The spread of disease resulting from physical contacts is described in terms of SIS (Susceptible Infected Susceptible) process over the epidemic layer and the spread of information about the disease outbreak is described in terms of UAU (Unaware Aware Unaware) process over the virtual interaction mediated awareness layer. The time variation of the transmission rates and the resulting co-evolution of these mutually competing processes is studied in terms of a network topology depend parameter({\alpha}). Using a second order linear theory it has been shown that in the continuous time limit, the co-evolution of these processes can be described in terms of damped and driven harmonic oscillator equations. From the results of the Monte-Carlo simulation, it is shown that for the suitable choice of parameter({\alpha}), the two process can either exhibit sustained oscillatory or damped dynamics. The damped dynamics corresponds to the endemic state. Further, for the case of endemic state it is shown that the inclusion of awareness layer significantly lowers the disease transmission rate and reduces the size of epidemic. The endemic state infection probability of a given node corresponding to the damped dynamics is found to have dependence upon both the transmission rates as well as on both absolute intra-layer and relative inter-layer degree of the individual nodes.