URL
https://www.medrxiv.org/content/10.1101/2020.06.13.20130625v2
Type d’article
Preprint
Thème
Stratégies de contrôle
Que retenir de cet article, en 1-2 phrases ?
The authors use a network model to describe epidemiological dynamics of the COVID-19 epidemic before, during, and after social distancing measures. They compare the results between ODE-based and network-based models and conclude that network models better describe the specific dynamics of this epidemic.
Objectifs de l’étude / Questions abordées
Investigating a network-based model of COVID-19 propagation in various states or cities through the USA, and its ability to account for dynamics before social distancing and the impact of social distancing measures.
Méthode
Three different networks are considered, representing extreme views of social networks. They are simulated, and their dynamics are compared to data from several places in the USA as well as results from ODE-based models.
Résultats principaux
Network-based models allow the description of power law growth of the infection spread (instead of exponential laws with ODE models). Network-based models correctly describe plateau phases observed when social distancing is applied. Conclusions based on network models are sometimes opposite to ODE-based models, for instance network-based models suggest to delay the end of social distancing measures in order to get a lower second wave.
Commentaire/brève évaluation
It is interesting to challenge the classical SIR models based on ODEs, as these models assume an important mix of individuals within the population, whereas this epidemic shows that local contacts may prevail. The paper is clear and the methodology easy to understand.