A discrete epidemic model and a zigzag strategy for curbing the Covid-19 outbreak and for lifting the lockdown

¹ Université Paris-Saclay, UVSQ, Laboratoire de Mathématiques de Versailles, 78000 Versailles, France.
² Pacific Institute for the Mathematical Sciences (PIMS), Department of Mathematics and Statistics, University of Victoria, Victoria, British Columbia, Canada.

^* Corresponding author: tahar.boulmezaoud@uvsq.fr

Received: 3 May 2020
Accepted: 29 October 2020

Abstract

This study looks at the dynamics of a Covid-19 type epidemic with a dual purpose. The first objective is to propose a reliable temporal mathematical model, based on real data and integrating the course of illness. It is a daily discrete model with different delay times, and whose parameters are calibrated from the main indicators of the epidemic. The model can be broken down in two decoupled versions: a mortality-mortality version, which can be used with the data on the number of deaths, and an infection-infection version to be used when reliable estimates of infection rate are available. The model allows to describe realistically the evolution of the main markers of the epidemic. In addition, in terms of deaths and occupied ICU beds, the model is not very sensitive to the current uncertainties about IFR. The second objective is to study several original scenarios for the epidemic’s evolution, especially after the period of strict lockdown. A coherent strategy is therefore proposed to contain the outbreak and exit lockdown, without going into the risky herd immunity approach. This strategy, called zigzag strategy, is based on a classification of the interventions into four lanes, distinguished by a marker called the daily reproduction number. The model and strategy in question are flexible and easily adaptable to new developments such as mass screenings or infection surveys. They can also be used at different geographical scales (local, regional or national).