Background. Masking serves an important role in reducing the transmission of respiratory viruses, including SARS-CoV-2. During the COVID-19 pandemic, several perspective and review articles have also argued that masking reduces the risk of developing severe disease by reducing the inoculum dose received by the contact. This hypothesis, known as the variolation hypothesis, has gained considerable traction since its development. Methods. To assess the plausibility of this hypothesis, we develop a quantitative framework for understanding the relationship between (i) inoculum dose and the risk of infection and (ii) inoculum dose and the risk of developing severe disease. We parameterize the mathematical models underlying this framework with parameters relevant for SARS-CoV-2 to quantify these relationships empirically and to gauge the range of inoculum doses in natural infections. We then identify and analyze relevant experimental studies of SARS-CoV-2 to ascertain the extent of empirical support for the proposed framework. Results. Mathematical models, when simulated under parameter values appropriate for SARS-CoV-2, indicate that the risk of infection and the risk of developing severe disease both increase with an increase in inoculum dose. However, the risk of infection increases from low to almost certain infection at low inoculum doses (with <1000 initially infected cells). In contrast, the risk of developing severe disease is only sensitive to dose at very high inoculum levels, above 106 initially infected cells. By drawing on studies that have estimated transmission bottleneck sizes of SARS-CoV-2, we find that inoculum doses are low in natural SARS-CoV-2 infections. As such, reductions in inoculum dose through masking or greater social distancing are expected to reduce the risk of infection but not the risk of developing severe disease conditional on infection. Our review of existing experimental studies support this finding. Conclusions. We find that masking and other measures such as distancing that act to reduce inoculum doses in natural infections are highly unlikely to impact the contact's risk of developing severe disease conditional on infection. However, in support of existing empirical studies, we find that masking and other mitigation measures that reduce inoculum dose are expected to reduce the risk of infection with SARS-CoV-2. Our findings therefore undermine the plausibility of the variolation hypothesis, underscoring the need to focus on other factors such as comorbidities and host age for understanding the heterogeneity in disease outcomes for SARS-CoV-2.

Animal models are frequently used to characterize the within-host dynamics of emerging zoonotic viruses. More recent studies have also deep-sequenced longitudinal viral samples originating from experimental challenges to gain a better understanding of how these viruses may evolve in vivo and between transmission events. These studies have often identified nucleotide variants that can replicate more efficiently within hosts and also transmit more effectively between hosts. Quantifying the degree to which a mutation impacts viral fitness within a host can improve identification of variants that are of particular epidemiological concern and our ability to anticipate viral adaptation at the population level. While methods have been developed to quantify the fitness effects of mutations using observed changes in allele frequencies over the course of a host’s infection, none of the existing methods account for the possibility of cellular coinfection. Here, we develop mathematical models to project variant allele frequency changes in the context of cellular coinfection and, further, integrate these models with statistical inference approaches to demonstrate how variant fitness can be estimated alongside cellular multiplicity of infection. We apply our approaches to empirical longitudinally sampled H5N1 sequence data from ferrets. Our results indicate that previous studies may have significantly underestimated the within-host fitness advantage of viral variants. These findings underscore the importance of considering the process of cellular coinfection when studying within-host viral evolutionary dynamics.

Animal models are frequently used to characterize the within-host dynamics of emerging zoonotic viruses. More recent studies have also deep-sequenced longitudinal viral samples originating from experimental challenges to gain a better understanding of how these viruses may evolve in vivo and between transmission events. These studies have often identified nucleotide variants that can replicate more efficiently within hosts and also transmit more effectively between hosts. Quantifying the degree to which a mutation impacts viral fitness within a host can improve identification of variants that are of particular epidemiological concern and our ability to anticipate viral adaptation at the population level. While methods have been developed to quantify the fitness effects of mutations using observed changes in allele frequencies over the course of a host’s infection, none of the existing methods account for the possibility of cellular coinfection. Here, we develop mathematical models to project variant allele frequency changes in the context of cellular coinfection and, further, integrate these models with statistical inference approaches to demonstrate how variant fitness can be estimated alongside cellular multiplicity of infection. We apply our approaches to empirical longitudinally-sampled H5N1 sequence data from ferrets. Our results indicate that previous studies may have significantly underestimated the within-host fitness advantage of viral variants. These findings underscore the importance of considering the process of cellular coinfection when studying within-host viral evolutionary dynamics.