URL

https://science.sciencemag.org/content/sci/370/6518/811.full.pdf

Type d’article

Article peer-reviewed

Thème

Stratégies de contrôle Immunité

Que retenir de cet article, en 1-2 phrases ?

In this work, the authors propose a SIR(S) model, which is a generalization of both the SIR and the SIRS (Susceptible-Infected-Recovered-Susceptible) models (assuming different immune and infection phenotypes), to see the effect of several parameters on the long time evolution of the COVID-19 pandemic. Parameters are for example protection level gained after first infection or vaccine, effect of seasonal changes in the transmission rate, the vaccination rate.

Objectifs de l’étude / Questions abordées

The main objective concerns the assessment of the influence of some parameters, such as

  • the protection level gained after the first infection or vaccine,
  • the effect of temporal changes in the transmission rate (due to seasons or nonpharmaceutical interventions),
  • the vaccination rate, on the prediction of the pandemic future. The authors study different scenarios. Moreover, the question of heterogeneity (due to age, or nonhomogeneous vaccine hesitancy) is discussed.

Méthode

The authors propose to use a generalization of SIR and SIRS models, called SIR(S) model, with different values for the chosen parameters. They consider different scenarios, first in a homogeneous case, without or with vaccine (perfect or not). They describe and discuss the estimations obtained by the model.

Résultats principaux

Through several sets of parameters, this work presents different scenarios for the evolution of the COVID-19 pandemic. First, reduction of transmissibility of secondary infections (parameter eps) may increase the period between first and second peaks of infection, as well as width of secondary peaks. When considering the availability of a vaccine, eps and the duration of immunity given by the vaccine have a large influence on the vaccination rate that is necessary to achieve disease-free state at equilibrium.
The authors also propose four different scenarios of possible futures: two without vaccine and different values of eps and duration of natural immunity, two with vaccine and different durations of vaccine-induced immunity (3 months and 1 year).
These four sets of parameters let them conclude to the important role of a vaccine, even an imperfect one, on the dynamics of the epidemic. Vaccination leads to reduced peaks.

Commentaire / brève évaluation, limites, ouvertures possibles

This work presents interesting simulations based on the SIR(S) model, with several parameters taken into account, especially the transmissibility of secondary infections and the availability of a vaccine. The influence on the epidemic dynamics of (even partial) immunity and temporal changes is shown in a clear way.
This study is based on seasonality assumption and simplified scenarios. It would be of course interesting, but difficult, to look deeper into the role of heterogeneities in the population.