Analysis of overdispersion in airborne transmission of COVID-19

Abstract

Superspreading events and overdispersion are hallmarks of the COVID-19 pandemic. However, the specific roles and influence of established viral and physical factors related to the mechanisms of transmission, on overdispersion, remain unresolved. We, therefore, conducted mechanistic modeling of SARS-CoV-2 point-source transmission by infectious aerosols using real-world occupancy data from more than 100 000 social contact settings in ten US metropolises. We found that 80% of secondary infections are predicted to arise from approximately 4% of index cases, which show up as a stretched tail in the probability density function of secondary infections per infectious case. Individual-level variability in viral load emerges as the dominant driver of overdispersion, followed by occupancy. We then derived an analytical function, which replicates the simulated overdispersion, and with which we demonstrate the effectiveness of potential mitigation strategies. Our analysis, connecting the mechanistic understanding of SARS-CoV-2 transmission by aerosols with observed large-scale epidemiological characteristics of COVID-19 outbreaks, adds an important dimension to the mounting body of evidence with regard to airborne transmission of SARS-CoV-2 and thereby emerges as a powerful tool toward assessing the probability of outbreaks and the potential impact of mitigation strategies on large scale disease dynamics.

Analysing the distribution of SARS-CoV-2 infections in schools: integrating model predictions with real world observations

“…Schijven et al [15] utilized the dose-response model to derive a risk assessment model for various expiratory events. Consolidating these findings, Chaudhuri et al [5] proposed a model for the Probability Density Function (PDF) of secondary infections due to long-range airborne transmission, which relied on observations based on certain numerical results generated from cell-phone based occupancy data [16]. In this study, we systematically develop a model for the PDF of secondary infections from the equation governing the virion concentration at an indoor location, and couple its solution to the dose-response model governing pathogen-host interactions.…”

“…Here, a well-mixed room is assumed to invoke i.e., the virion concentration is homogeneous in space. If the index case possessing a viral load ρ continuously ejects mucosalivary fluid at a rate of Q ( t ) within a room of volume V , then the virion concentration c is governed by where the effects of air change rate ACH , wall deposition parameter β 0 , and virus half-life t 1 / 2 appear through the loss parameter a = ACH/ 3600 + β 0 + ln(2) /t 1 / 2 [5]. The ventilation effect is assumed to be uniform here, though it is recognized that practically the aerosol flow can be anisotropic with a degree of directionality.…”

“…Viral load ρ is a dominant factor that decides the shape of the Z distribution, and hence its variability needs to be captured when providing it as an input parameter [5,7].…”

“…The review by Prather et al [4] highlights the ability of aerosols to remain airborne for an extended period of time. Chaudhuri et al [5, 6] showed that the viral load is a major contributor in determining the large variations in the number of secondary infections. These were complemented by Chen et al’s [7] observations of viral load heterogeneity being strongly tied to the overdispersed nature of SARS-CoV-2 infections, along with Bhavnani et al’s [8] findings of a direct correlation between higher index case viral loads and rising secondary infection counts.…”

Improving Indoor Air Ventilation by a Ceiling Fan to Mitigate Aerosols Transmission

“…Effective mitigation measures necessitate clear understanding of droplet and aerosol transport in air (Agrawal and Bhardwaj 2020 , 2021 ; Chaudhuri et al. 2022 ), droplet retention by a surface (Bhardwaj and Agrawal 2020 ; Chatterjee et al. 2021 ), droplet evaporation kinetics in different environments (Bhardwaj and Agrawal 2020 ; Chatterjee et al.…”