Understanding infection progression under strong control measures through universal COVID-19 growth signatures

TL;DR

This paper introduces a novel analytical framework that identifies universal growth signatures in COVID-19 case data, revealing dynamical regimes and key infection parameters to better understand disease progression under control measures.

Contribution

It develops a physics-inspired approach to analyze COVID-19 growth signatures, providing new insights into infection dynamics and parameter inference under strong control measures.

Findings

Identification of three dynamical regimes: exponential, superlinear, sublinear

Derivation of scaling laws for COVID-19 growth patterns

Framework applicable to other infectious diseases

Abstract

Widespread growth signatures in COVID-19 confirmed case counts are reported, with sharp transitions between three distinct dynamical regimes (exponential, superlinear and sublinear). Through analytical and numerical analysis, a novel framework is developed that exploits information in these signatures. An approach well known to physics is applied, where one looks for common dynamical features, independently from differences in other factors. These features and associated scaling laws are used as a powerful tool to pinpoint regions where analytical derivations are effective, get an insight into qualitative changes of the disease progression, and infer the key infection parameters. The developed framework for joint analytical and numerical analysis of empirically observed COVID-19 growth patterns can lead to a fundamental understanding of infection progression under strong control measures, applicable to outbursts of both COVID-19 and other infectious diseases.