We are a doctor and a mathematician and a data scientist. N95s won’t work for kids
The Berkeley Unified School District (BUSD) in the Bay Area of SF where we live, announcement Tuesday that he planned to “switch all students and staff” to KN95 respirators. If worn correctly, these respirators filter out 95% of particles the size of those carrying the SARS-CoV-2 virus. BUSD proposed this measure as a way to slow the spread of COVID-19 and keep schools open. These respirators would be needed for the entire school day, including outdoors during the gymnasium and recess.
Unfortunately, the effectiveness of ventilators is vastly overestimated and there is little evidence that they stop community transmission. Additionally, NIOSH-approved respirators are tight, uncomfortable, and can interfere with breathing. OSHA requires both a fit test and a medical evaluation before workers can wear them. We’ve all seen images of healthcare workers with faces bruised by properly worn respirators.
The truth is, the burdens of these masks outweigh their benefits for children. We must consider more effective and less harmful interventions as we come together to keep schools open and safe.
We know that masks have become a very controversial issue. But as a doctor with a doctorate in computational mathematics and a professor of data science, we have read the studies on this subject. We both have children and want them to be safe. But we also don’t want them to be burdened with measures that won’t help protect them or their peers. And mandating ventilators for children is out of step with CDC guidelines and international standards.
What is the evidence that respirators stop the spread of covid19? Influenza studies provide some guidance. Although respirators provide better filtration under perfect laboratory conditions, people who wear them are just as likely to catch the flu whether they are wearing a surgical mask or a respirator. Although respirators have higher filtration capabilities, a Cochrane Review And one independent meta-analysis both found there were no clear differences between the effectiveness of surgical masks and respirators in preventing infections like the flu. The Influenza virus and the SARS-CoV-2 virus are large and infection transmission rate between close contacts are similar.
Meanwhile, countries that have mandated respirators have failed to stop the spread of COVID-19, even during power surges. Parts of Germany—bavaria and Berlin– and all Austria have both had long-serving terms since early 2021 and are still facing huge power surges this winter that exceeds all previous waves.
If N95s filter so well, why are respirators an ineffective intervention? Because masking is a behavioral intervention as much as a physical one. For respirators to work, they must fit snugly, must be OSHA tested, and should only be used for short periods of time because their effectiveness decreases as they get wet from breathing.
The fit requirements and comfort issues are untenable in children who have small faces and must wear masks for six or more hours each day. For these reasons, NIOSH specifically states that children should not use respirators, and no respirator is approved for children. These views are shared by the California Department of Public Health. Concerns about impaired breathing and improper use outweigh the potential benefits. There are no studies on the effectiveness of respirators on children because they are not approved for pediatric use.
Leaving respirators aside for a moment, many organizations concerned about the health of children do not recommend masking the youngest at all. The World Health Organization recommended against the wearing of masks for children 5 and under and recommends a “risk-based approach” in their use by children 6-11 years that takes into account “a child’s ability to comply with the ‘appropriate use of masks and the availability of appropriate adult supervision’. The European Center for Disease Control categorically recommended against the wearing of masks by children under 12.
Respirators are not necessary to protect children from COVID-19 due to the surprisingly low risk COVID-19 poses to them. It’s always scary as a parent to face an invisible threat, but we have over two years of global data on this. The risks children face from COVID-19 continue to be overshadowed by other infectious risks. Some parents may consider the different risks and decide that they may want to find good respiratory protection for their children. But given the lack of evidence of benefits, and given the well-documented harms that exist, the precautionary principle tells us that we cannot require ventilators for all children.
While we shouldn’t panic about the danger, there are still helpful things we can do at this stage of the pandemic. We should focus our time and energy on interventions that can help during this current Omicron surge. Because California is blessed with pleasant weather, even in winter, we may open windows and move some school activities outdoors during times of high community transmission. We have excellent vaccines that can protect people from serious illness, including hospitalization and death. We must ensure that all vulnerable teachers are vaccinated and strengthened, as well as all elders and others in our communities. And we can use the state-declared state of emergency to allow parents to volunteer to be substitute teachers to address staffing shortages.
These interventions can be made today, and by focusing our efforts there, we can keep our schools open and thriving. But forcing children to wear respirators designed and intended for specialist professionals is not the appropriate route to protecting children or the community.
Update: After this editorial was published, BUSD sent an email to the community retracting the claim that they plan to “switch all students and staff” to KN95 respirators. They have confirmed that KN95 respirators will be optional.
Ram Duriseti is a Clinical Associate Professor of Emergency Medicine at Stanford. Benjamin Recht is a professor of electrical engineering and computer science at the University of California, Berkeley, where he researches and teaches data science and machine learning.
The opinions expressed in this article are those of the authors.