Do Masks Work?

What is the evidence?

Filtering through the lies and chaos over masks need not be a daunting process.

We have tools in medical education that do that. UPTODATE is the best online clinical decision support tool providing current best evidence.

What does the data actually show?

“Multiple observational studies support the use of masks to provide source control and reduce transmission in the community [251,259-270]. In epidemiologic studies, government-issued mask mandates and high rates of self-reported mask wearing have each been associated with decreased community incidence rates and, in some cases, decreased COVID-19 hospitalization rates [266,271-273]; lifting of universal mask mandates has conversely been associated with increased case rates [274]. In a meta-analysis of six observational studies, mask-wearing was associated with a 53 percent reduction in the incidence of COVID-19 [268]. Modeling studies have also suggested that high adoption of mask-wearing by the general public can reduce transmission, even if masks are only moderately effective in containing infectious respiratory secretions [275,276].”

“Nevertheless, efficacy of masks has been difficult to demonstrate consistently in clinical trials. In a meta-analysis of six trials that did not demonstrate reductions in laboratory-confirmed influenza or SARS-CoV-2 infection with wearing medical masks in the community (risk ratio [RR] 1.01, 95% CI 0.72 to 1.42), only two of those trials evaluated SARS-CoV-2 transmission [277]. One of those was a cluster-randomized trial in Bangladesh, in which villages that received free masks as well as behavioral and social interventions to promote masks had increased mask use (40 versus 14 percent in control villages) and, among those who received medical masks, an associated 11 percent relative reduction in SARS-CoV-2 seroprevalence that was not statistically significant (adjusted RR 0.89, 95% CI 0.78-1.01) [278]. The other trial, from Denmark, is discussed below.

Prevent exposure – Mask-wearing in the community may protect the wearer; in several observational studies, consistent mask wearing, particularly with medical masks or respirators, has been associated with a lower risk of infection [279-282]. In a report of 382 service members who were surveyed about personal preventive strategies in the setting of a SARS-CoV-2 outbreak on a United States Navy aircraft carrier, self-report of wearing a face cover was independently associated with a lower likelihood of infection (odds ratio [OR] 0.3), as were avoiding common areas (OR 0.6) and observing social distancing (OR 0.5) [279]. In a retrospective analysis of 1060 individuals identified by contact tracing following clusters of infections in Thailand, wearing a mask all the time was associated with a lower odds of infection compared with not wearing a mask; there was no significant association between wearing a mask some of the time and infection rate [280]. In contrast, a randomized trial from Denmark did not identify a decreased rate of infection among individuals who were provided with surgical masks and advised to wear them when outside of the house for a month (1.8 versus 2.1 percent among individuals who were not given masks or the recommendation) [283]. However, the low rate of community transmission (as reflected by the low overall infection rate) may have made it difficult to detect a meaningful difference.”

So what's the bottom line? The lack of data from prospective studies leaves a lot to be desired but clearly masks do provide some benefit.

There are no prospective randomized controlled trials on parachutes either.

Masks help.