This week two Texas judges issued temporary restraining orders that allow public schools in Bexar and Dallas counties to require that staff and students wear face masks as a safeguard against COVID-19. The legal issue is whether Texas Gov. Greg Abbott's executive order banning such mandates fits within his authority under the Texas Disaster Act of 1975. But the wisdom of requiring masks in schools depends on whether the public health benefits of that precaution outweigh the burdens it imposes on students and employees. On that point, the evidence is not nearly as clear as mandate enthusiasts imply.
Two important facts should inform decisions about face masks in schools.
First, COVID-19 infections among children and teenagers are rarely life-threatening. According to the "current best estimate" from the Centers for Disease Control and Prevention (CDC), the infection fatality rate (IFR) for people younger than 18 is 0.002 percent. By contrast, the CDC estimates that the IFR for COVID-19 among people 65 or older is 9 percent, 4,500 times as high. The estimated IFRs for other age groups fall between those two extremes: 0.05 percent for 18-to-49-year-olds and 0.6 percent for 50-to-64-year-olds.
Second, COVID-19 vaccines are currently available to all Americans 12 or older, and the vaccination rate is especially high among older Americans, which helps explain why the recent surge in cases has not been accompanied by a commensurate increase in deaths. For teachers and other staff members who are concerned about catching COVID-19 in school, vaccination sharply reduces the risk of infection and is even more effective at preventing severe cases. The same goes for students 12 or older.
Keeping those facts in mind, what is the evidence that face masks play an important role in preventing school-related COVID-19 outbreaks? In a New York Times opinion piece published on Tuesday, Duke University pediatrician Kanecia Zimmerman and We Studied One Million Students. This Is What We Learned About Masking."
Here is how Zimmerman and Benjamin describe the results of their study, which was based on data from March through June 2021:
During that time, more than 7,000 children and adults acquired the coronavirus and attended school while infectious. Because of close contact with those cases, more than 40,000 people required quarantine. Through contact tracing and testing, however, we found only 363 additional children and adults acquired the coronavirus. We believe this low rate of transmission occurred because of the mask-on-mask school environment: Both the infected person and the close contact wore masks.
That belief is not actually supported by Zimmerman and Benjamin's study. Since all the North Carolina public schools they studied had universal masking, there was no control group of schools without that requirement. It is therefore impossible to say whether the low rate of secondary transmission can be attributed to the mask policy. "Because North Carolina had a mask mandate for all K-12 schools," Zimmerman and Benjamin concede, "we could not compare masked schools to unmasked schools."
In lieu of a control group, Zimmerman and Benjamin cite a few COVID-19 outbreaks that they attribute to a lack of universal masking. Here are the cautionary examples they mention:
• This month in North Carolina, Mooresville Graded Schools and the Union Academy Charter School decided to require masks after "both experienced outbreaks during the first days of the new school year," the ABC affiliate in Charlotte reports. An elementary school in the Mooresville system identified "nine positive cases," while the charter school saw "at least 14 confirmed COVID-19 cases."
• In Illinois last month, Springfield Public Schools began requiring masks during a summer session, citing "an increase in COVID-19 positive cases among SPS students and staff."
• In May 2020, an Israeli public school had an outbreak that involved "153 students and 25 staff members" who "were confirmed as COVID-19-positive." The outbreak was tied to an "extreme heatwave," during which a face mask requirement was suspended, windows were closed, and "air-conditioning functioned continuously in all classes." The authors of the report on the Israeli outbreak also note that "distancing among students and between students and teachers was not possible," because the classrooms were "crowded." The CDC cites "classroom crowding" and "poor ventilation" as factors in the outbreak.
These are all examples of outbreaks (or "an increase in COVID-19 positive cases") that occurred when masking was optional. But they do not show that a lack of masking was the main reason for virus transmission. And the Israeli outbreak, which is commonly cited by mask-mandate advocates, is striking because it was unusual, even though other Israeli schools also did not require masks during the three-day heat wave.
Meanwhile, Zimmerman and Benjamin overlook examples of school systems that did not require masks but nevertheless saw minimal COVID-19 transmission.
In Florida, where many school districts did not require masks, the CDC found that less than 1 percent of students were infected in schools during the first semester after they reopened in August 2020. The CDC did report that school districts without mask mandates had a higher school-related infection rate: 1,667 vs. 1,171 per 100,000 students. But the study notes that smaller districts were less likely to require masks, and they also "had a higher proportion of students attending in-person instruction," which likewise was "positively correlated with the student case rate." And even in districts without mandates, just 1.7 percent of students were infected at school.
When England reopened schools in August 2020, they did not require face masks. Public Health England identified 969 outbreaks, or about one for every 25 schools. The outbreaks affected 2 percent of primary schools and 10 percent of secondary schools. In response to a geographically representative survey, 100 primary schools and 79 secondary schools reported 2,314 cases, or an average of about 13 per school. In primary schools that experienced outbreaks, 0.84 percent of students were infected; the rate for secondary schools was 1.2 percent. Teachers were more likely to be infected and more likely to transmit the virus than students.
Both of these studies were conducted at a time when vaccines were not available to staff or students. Now that 70 percent of American adults have been at least partly vaccinated and vaccines are available to students 12 or older, we should see similar or lower rates of school-related infection, even allowing for the greater transmissibility of the delta variant.
"Although outbreaks in schools can occur," the CDC says, "multiple studies have shown that transmission within school settings is typically lower than—or at least similar to—levels of community transmission, when prevention strategies are in place in schools." Regarding mask requirements specifically, it says "most studies that have shown success in limiting transmission in schools have [involved schools that] required that staff only or staff and students wear masks as one of the school's prevention strategies." That gloss implies that some studies found schools had "success in limiting transmission" even without mask mandates or with mandates that did not apply to students.
The CDC, which is now urging "indoor masking for all individuals age 2 years and older" in schools and child care facilities, cites six studies to support its conclusion that mask mandates are crucial: a preliminary report on Zimmerman and Benjamin's North Carolina research; a study concluding that reopening schools in Italy, where students were required to wear masks, did not appear to drive the second COVID-19 wave in that country; a study of Chicago schools, which also required masks, finding "a lower attack rate for students and staff participating in in-person learning than for the community overall"; a CDC study that reported "limited secondary transmission" of COVID-19 in Rhode Island child care programs that required adults to wear masks; a CDC study that found 6 percent of child care facilities in Washington, D.C., with a similar requirement experienced an outbreak; and a CDC report of "minimal" COVID-19 transmission at a New Jersey school "after implementation of a comprehensive mitigation strategy" that included "universal masking."
In their Times piece, Zimmerman and Benjamin cite CDC data from Utah, Missouri, and Wisconsin. The Utah study found that "mask adherence was high" and that "COVID-19 incidence among students and staff members was lower than in the county overall." According to the Missouri study, COVID-19 transmission was "much lower" in schools than in the general community. The schools had adopted precautions that included "mandating use of face masks, physical distancing in classrooms, increasing ventilation with outdoor air, identification of close contacts, and following CDC isolation and quarantine guidance." In Wisconsin, "reported student mask-wearing was high," and "transmission risk within schools appeared low."
These studies show that mask mandates are consistent with low rates of COVID-19 transmission. But contrary to what the CDC implies, they do not show that mask mandates are necessary to keep infection rates low, which would require comparing outcomes in otherwise similar schools with and without mandates.
In May, based on data from Georgia, the CDC reported that "COVID-19 incidence was 37% lower in schools that required teachers and staff members to use masks," which was similar to the difference associated with "improved ventilation."* But while the incidence of infection was 21 percent lower in schools that also required students to wear masks, that difference was not statistically significant.
In a May 21 preprint study, Brown University economist Emily Oster and four other researchers analyze COVID-19 data from Florida, New York, and Massachusetts for the 2020–21 school year. "We do not find any correlations with mask mandates," Oster et al. report. But they note that "all rates are lower in the spring, after teacher vaccination is underway." The authors caution that their study "does not imply masks are ineffective, as these results focus only on masking in schools and do not take community behavior into consideration." They also note that they considered "mask mandates and not actual masking behavior."
Zimmerman and Benjamin say it is "now clear" that "universal masking is linked to lower spread" in schools; that "schools that do not require masks will have more coronavirus transmission"; that "if we send children to school without masks, we increase their risk of acquiring Covid-19"; that "masking helps prevent spread among unvaccinated people in schools"; that "universal masking is a close second" to vaccination as a way of preventing school outbreaks; and that "universal masking in schools can save lives." But the evidence they cite to support these assertions is inconclusive at best.
Assuming that universal masking in schools does make a difference (which it might!), it is by no means clear that the benefits outweigh the costs. The data from Florida and England indicate that COVID-19 transmission in schools was a minor problem even without mask mandates and even before vaccination was possible. Given the low risks that children face from COVID-19, the low infection rates even in schools that don't require masks, and the fact that vaccination is readily available to adults and teenagers, the benefits of forcing kids to cover their faces all day, whether or not they are vaccinated, are likely to be small.
The costs, meanwhile, are more substantial than mandate supporters typically acknowledge. The inconvenience and discomfort caused by mask requirements aggravate the unpleasantness of environments that were stressful, boring, and restrictive long before anyone had heard of COVID-19. Masks interfere with communication, learning, and social interaction. And they unfairly burden children with the responsibility of preventing infections that primarily threaten adults, who can better protect themselves by getting vaccinated. To justify those costs would require more evidence than mandate advocates have been able to muster.
[This post has been revised to note that the the delta variant, which accounts for the vast majority of recent COVID-19 cases in the U.S., is more contagious than the variants involved in school outbreaks last year and earlier this year.]
*CORRECTION: The publication date of the CDC's Georgia study was misreported in the original post.