Face masks: simple and effective

From the Schiffer group, Vaccine and Infectious Disease Division

The use of face coverings is one tool to limit the spread of viruses such as SARS-CoV-2. Face masks serve as physical barriers that prevent a percentage of the viral particles released by a transmitter from reaching an exposed person, an effect that is doubled when both people are masked. Although an important tool for preventing virus spread, commonly available masks do not filter 100% of infectious virus and are not airtight, while more effective N95 masks are in short supply and difficult for prolonged wear. Additionally, mask effectiveness is lowered when masks are removed intermittently for eating and drinking, worn incorrectly, or when mask compliance is low. Although imperfect, masking is an important tool for preventing the spread of respiratory infections such as SARS-CoV-2, but incomplete data on the effectiveness of masking in real-world scenarios has led to inconsistent and varied masking regulations and recommendations.

Although masking effectiveness has been tested in controlled laboratory and hospital environments where compliance is 100%, the effectiveness of masking in real-world scenarios and in the prevention of super-spreader events has not been studied. To simulate the effects of masking on SARS-CoV-2 spread in real-world events, researchers from the Schiffer group in the Vaccine and Infectious Disease Division developed a mathematical model to estimate the effects of masking on viral transmission, published in Scientific Reports. First, to define the period of infectiousness, the group used previous models they had developed to estimate baseline SARS-CoV-2 viral load dynamics in untreated, unmasked infected people. Next, they modeled viral transmission between pairs, assuming that either the transmitter or exposed person was masked, followed by a scenario where both people in the transmission pair were masked. The model simulated transmission using a range of low to high viral loads to determine how the effects of masking differ with viral load.  

Using these predictions, the team next modeled the effective reproductive number (Re)—or number of people that can be infected by one infected transmitter—based on different masking scenarios. They simulated transmission from 3,000 potential transmitters among exposure networks at different levels of mask use, which is a product of the percentage of people wearing a mask and the percentage of time they spend wearing them. Unsurprisingly, increased mask use is correlated with lower Re values. Based on 2020 US surveys that estimated public mask use, a scenario in which 75% of people wear masks 75% of the time (resulting in roughly 50% mask use), is most representative of real-world public settings; this number is lower in private settings. Although face coverings do not filter 100% of infectious virus, the prediction showed that most transmission occurs from an unmasked person to another unmasked person, highlighting the imperfect but highly effective prevention that masks provide, as the likelihood of any person transmitting disease drastically decreases with mask use. Notably, no previous studies have investigated the impact of masking on super-spreader events, where an individual person transmits infection to at least 5 people. However, the model found that even modest mask use decreased Re from super spreaders, suggesting that masking could significantly decrease infections even from highly contagious people with many social contacts.

Effect of mask utilization and efficacy on proportion of masked transmissions contributing to total R0.
Effect of mask utilization and efficacy on proportion of masked transmissions contributing to total R0. Figure from publication.

Finally, the authors validated this model by feeding it data collected from the Seattle area in September 2020 to estimate if the reduction in Re during this month was a result of mask use.  Modeling this data, including mask use, mobility of people within the community (decreased with social distancing), they found that mask use was responsible for a 27% reduction in Re, demonstrating a real-world scenario where masking imperfectly but meaningfully decreased SARS-CoV-2 transmission. Together, these results demonstrate that masks, which are inexpensive, accessible, and easy to use, are a crucial tool for preventing viral transmission in both typical and super-spreader events.

This work expanded on previous masking efficacy studies by factoring in real-world, imperfect mask use scenarios. Importantly, this study found that even moderate masking compliance with moderately effective masks provides an effective intervention against transmission where vaccines are unavailable, unused, or incompletely effective. However, a critical threshold of masking must be obtained to meaningfully prevent spread, and the authors predict that increasing mask use from 50 to 80% would drastically reduce spread, and 100% mask use in those with more than 10 contacts per day could effectively stop super-spreader events. Although effective vaccines are now available, masking continues to provide a simple and effective option for preventing viral spread among those who can’t or won’t be vaccinated and likely decreasing breakthrough infections among the vaccinated.


Goyal A, Reeves DB, Thakkar N, Famulare M, Cardozo-Ojeda EF, Mayer BT, Schiffer JT. Slight reduction in SARS-CoV-2 exposure viral load due to masking results in a significant reduction in transmission with widespread implementation. Scientific Reports.2021 Jun 4;11(1):11838. doi: 10.1038/s41598-021-91338-5.

This work was supported by the National Institutes of Health, the Washington Research Foundation, and the University of Washington Center for AIDS Research.