Modelling preventive measures and their effect on generation times in emerging epidemics

TL;DR

This paper introduces a stochastic epidemic model to analyze how different preventive measures influence disease transmission dynamics, especially the reproduction number and generation times, with implications for COVID-19.

Contribution

The model uniquely captures the effects of various interventions on both reproduction numbers and generation times, highlighting potential biases in epidemic parameter estimation.

Findings

Uniform reduction and vaccination affect reproduction number but not generation time.

Isolation, screening, and contact tracing influence both reproduction number and generation time.

Generation time distribution variations can lead to biases in estimating epidemic parameters.

Abstract

We present a stochastic epidemic model to study the effect of various preventive measures, such as uniform reduction of contacts and transmission, vaccination, isolation, screening and contact tracing, on a disease outbreak in a homogeneously mixing community. The model is based on an infectivity process, which we define through stochastic contact and infectiousness processes, so that each individual has an independent infectivity profile. In particular, we monitor variations of the reproduction number and of the distribution of generation times. We show that some interventions, i.e. uniform reduction and vaccination, affect the former while leaving the latter unchanged, whereas other interventions, i.e. isolation, screening and contact tracing, affect both quantities. We provide a theoretical analysis of the variation of these quantities, and we show that, in practice, the variation of the generation time distribution can be significant and that it can cause biases in the estimation of reproduction numbers. The framework, because of its general nature, captures the properties of many infectious diseases, but particular emphasis is on COVID-19, for which numerical results are provided.