Schools present a tradeoff. When they’re open, society benefits from the essential services schools provide: education, child abuse reporting, hot meals, and socio-emotional development. Closing them may theoretically slow viral spread. How should a society decide?
Nearly every discussion or debate around COVID-19 has become protracted, political, and bitter, and what to do about schools is no exception. Over the last few months, we have reviewed the emerging evidence and reached the same conclusion: Unless the local healthcare system is approaching overload or collapse, schools should remain open. In this op-ed, we hope to explain why.
We approach this question from two very different perspectives — one of us is a physician, epidemiologist, and public health scholar, and the other is a political scientist with a long-standing focus on education policy. Collectively, our expertise has allowed us to examine the tradeoffs of schools during a pandemic.
Keeping schools open: risks for different groups
Data last week from Sweden are reassuring. Nearly 2 million kids attended school from March to June, as the outbreak surged. Kids were not asked to wear masks. Despite this, only 15 kids developed severe COVID-19 or multi-inflammatory syndrome requiring ICU stay, of whom four had underlying medical problems. No children died.
Another study from North Carolina examined teacher and student outcomes when schools were open for hybrid in-person learning from August to October 2020. Kids used masks, hand hygiene, and social distancing — all reasonable precautions. During this period, the epidemic raged in the state, and the rate of infection in the population was 1 to 2 per 1,000 residents. If transmission occurred in school at the same rate as in the community, we would expect 800-900 cases that could be linked to schools. Yet, the study found just 32 cases of in-school transmission among students and teachers. In short, schools are remarkably safe for school-age kids.
Now consider the benefits of in-person school. Education is essential for building knowledge and skills that will allow today’s children to lead productive and rewarding lives, and education is a vehicle for upward mobility. But aside from academic knowledge, schools also help develop non-cognitive skills — such as self-control and executive function — with downstream consequences for things like risk of teenage pregnancy and involvement in the criminal justice system. Schools benefit kids.
Teachers are justifiably worried about their own safety. Again, both Sweden and North Carolina provide reassurance. In the same Swedish data, there was no significant increase for the risk of severe COVID among preschool teachers when compared to people in the general population, and there was no increased risk among regular school teachers (RR 0.43, 95% CI 0.28 to 0.68).
It is important to acknowledge that in-person teaching may not be entirely risk-free. There are a number of high-profile anecdotes of school employees becoming infected from fellow co-workers. And an earlier study on Sweden, which compared teachers in lower secondary schools (which remained open last spring) to upper secondary schools (which closed) did find higher rates of infections among those working in person. But the magnitudes of these risks need to be kept in mind. Swedish teachers were no more likely to become infected than the parents of the children they taught and at far lower risk than other workers such as taxi drivers.
In other words, total isolation is undoubtedly safer than working at a school — but that is true of almost every other activity and essential occupation. The risks to teachers are generally low, and can be mitigated with masks, distancing, and opening windows. The benefit they provide is tremendous. For this reason, we do not find risk to teachers as a compelling argument for keeping schools closed to students.
Parents and communities
The role of schools in broader community spread is difficult to estimate because most places closed schools at the same time that they adopted other non-pharmacological interventions last spring when fear about the virus was growing. Thus, it is almost impossible to separate the effect of school closure from other policies and the behavior changes that occurred when people voluntary stayed home.
Yet, one particularly well-done study from Germany took advantage of a pre-existing staggered summer vacation schedule to credibly quantify the impact of school openings on community transmission. Opening schools appeared to have absolutely no impact on the dynamics of the epidemic — not just among kids or teachers, but all members of the community. That’s worth repeating: this elegant study failed to find any increase in cases due to schools being open. The authors conclude, “the benefits of their closures do not outweigh the costs.”
Another recent study, which used examined COVID-related hospitalizations using health insurance claims in the U.S., found no increase in hospitalizations in most parts of the country that did reopen schools this fall. Although the results were more ambiguous for areas that already had unusually high rates of COVID spread, the impact of schools was fairly small. In short, it seems clear that schools have not been — and are unlikely to become — the primary drivers of community spread, as many feared last spring. (The European Centre for Disease Prevention and Control also has a good review of the topic.)
Now, let’s turn our attention to the alternative policy.
Keeping schools closed: risks for different groups
Although many teachers have been working tirelessly to adapt to the online environment, virtual instruction simply cannot replace in-person learning. Even before SARS-CoV-2, we had overwhelming evidence that online education was a poor substitute for the real thing.
While this is true for all kids, it is especially worrying for some of our most at-risk populations — homeless families, English learners, and students with disabilities. Nor can virtual learning be easily adapted for high school students enrolled in career and technical training programs. (Imagine trying to learn how to weld via Zoom!)
We also need to consider broader impacts, in terms of student mental and physical health. In many parts of the country that kept schools closed this fall, reports of suspected child abuse plummeted, suggesting that thousands of cases are going unreported, and children isolated with abusers. New evidence from Japan also shows that school closures have caused young children to gain significant weight, and some U.S. cities are seeing a disturbing surge in juvenile gun violence.
Teachers and school workers
We should not underestimate the impact of school closures on the reputational damage to public education and the impact on school segregation. Public school enrollment dropped sharply this fall. While part of this decline is likely due to parents delaying the start of kindergarten, some is also due to families leaving for other options, such as private schools and homeschooling. In addition to the direct impacts of lower enrollment on funding, it is important for educators to consider the long-term impact on popular support for public education, including the willingness of voters to approve future tax increases.
Schools have served as a vehicle in America to lift people out of poverty, to move closer toward equality of opportunity. If wealthier and upper middle-class families abandon public schools, the system that remains will be weakened and less equipped to bring about the transformation we still desperately need in this country. One famous economist noted pre-pandemic that you have a better shot at the American dream in Canada than the U.S. Loss of the American dream has implications for wealth and health of the citizenry, and democracy itself.
Parents and communities
As research from other countries shows, even short-term disruption to student learning can create long-term economic harms. For example, teacher strikes in Argentina reduced future earnings of the students impacted, and projections suggest that current school closures will depress economic growth for decades to come. While it’s easy to claim that saving lives trumps earning income, the truth is money, and what it does for human beings, often translates into health. The association between income and life expectancy is strong, and of large magnitude.
But we don’t have to wait that long to see some of the negative effects on society — especially on women, who still bear the brunt of childcare responsibilities in our society. Since last March, more than two million women have left the labor force and many others have reduced their hours. In December alone, women accounted for all of the reported job losses, and there is strong evidence showing that school closures have themselves increased anxiety and stress among mothers.
For these reasons, our conclusion is that schools provide a vital service to American society. Schools may even be more important in the U.S. than in Western Europe, which bent over backward in the fall to reopen them and keep them open. The schools that are open in the U.S. often cater to the affluent, and globally there is no relationship between metrics of pandemic virus — cases, hospitalizations, or deaths — and whether schools are open and closed.
We believe: schools must remain open unless the local healthcare system is facing collapse, due to capacity constraints. Short of such a scenario, the tradeoffs favor doing everything possible to keep schools open. This is not an easy decision to reach, but deciding which course is best when facing hard tradeoffs is the hallmark of policy.
Some argue that a new variant of SARS-CoV-2 — the B.1.1.7 strain — changes the calculus. B.1.1.7 has raised concern that it may be more likely to infect kids, more likely to spread in schools, and change the cost-benefit calculations regarding school openings. Early data from the United Kingdom does suggest the virus is more transmissible: 15% of household contacts acquire B.1.1.7 vs 11% for other strains. However, there is no evidence that its increased transmissibility is occurring preferentially in younger ages. As such, this strain must be watched cautiously and might increase overall cases of COVID-19, but there is no reason it changes the calculus we laid out here. Schools should remain open unless there is pending health systems collapse — with the strain or without. That point may come slightly sooner in some regions, but if it does not come, schools ought to remain open.
In short, for the reasons we summarize here, we are in favor of schools. We don’t take this view because we are callous or indifferent to loss of life, we take this view because we believe in our hearts and minds that it will result in the least amount of human harm and suffering over the long haul. When it comes to pandemics, there are no winners. Opening schools means we lose less. Sometimes losing less is the best we can do.
Vladimir Kogan, PhD, is an associate professor in the department of political science at Ohio State University. Vinay Prasad, MD, MPH, is a hematologist-oncologist and associate professor of medicine at the University of California San Francisco, and author of Malignant: How Bad Policy and Bad Evidence Harm People With Cancer.
Last Updated January 12, 2021