Uncertainty analysis of facemasks in mitigating SARS-CoV-2 transmission

In the context of global spread of coronavirus disease 2019 (COVID-19) caused by a novel coronavirus (SARS-CoV-2), there is a controversial issue on whether the use of facemasks is promising to control or mitigate the COVID-19 transmission. This study modeled the SARS-CoV-2 transmission process and analyzed the ability of surgical mask and N95 in reducing the infection risk with Sobol's analysis. Two documented outbreaks of COVID-19 with no involvers wearing face masks were reviewed in a restaurant in Guangzhou (China) and a choir rehearsal in Mount Vernon (USA), suggesting that the proposed model can be well validated when airborne transmission is assumed to dominate the virus transmission indoors. Subsequently, the uncertainty analysis of the protection efficiency of N95 and surgical mask were conducted with Monte Carlo simulations, with three main findings: (1) the uncertainty in infection risk is primarily apportioned by respiratory activities, virus dynamics, environment factors and individual exposures; (2) wearing masks can effectively reduce the SARS-CoV-2 infection risk to an acceptable level (< 10^-3) by at least two orders of magnitude; (3) faceseal leakage can reduce protection efficiency by approximately 4% when the infector is speaking or coughing, and by approximately 28% when the infector is sneezing. This work indicates the effectiveness of non-pharmaceutical interventions during the pandemic, and implies the importance of the synergistic studies of medicine, environment, social policies and strategies, etc., on reducing hazards and risks of the pandemic.