As schools in Washington and across the U.S. (remotely) wind down for summer break, many families, teachers and state decision makers are hoping for an answer: Did closing school buildings ultimately help curb spread of the novel coronavirus?

New epidemiological and social science research hints that shuttering school buildings did indeed help slow the virus’s spread.

The findings, based mostly on unpublished data and a handful of peer-reviewed studies, don’t definitively prove that closing schools altered the epidemic’s course. But such emerging data may help policymakers decide if and how it’s safe to return to school come fall. Those officials will also have the advantage of watching how some other countries and states are reopening their schools — and the consequences of those decisions.

To understand the role of school closures, researchers are homing in on two angles: the biological ability of children to spread the coronavirus, and what we know so far about how school closures tracked with the disease’s trajectory.

Researchers have already moved beyond a big biological question that marked the early days of the coronavirus’s spread: whether children actually contract it. They do, several studies find. But children tend to be spared from its worst effects, with some exceptions.

Now, scientists are racing to unravel other important aspects of the condition in children. Epidemiological studies are attempting to trace infected children’s contacts to decipher how well kids transmit the coronavirus. Some researchers are examining children’s so-called “viral load” — the concentration of virus a person has inside them — compared to adults, which may also lend insight into this question. A major project funded by the National Institutes of Health aims to track incidence of the disease in children.

But studies on transmission and incidence are still in early days. They are hampered by the fact that studying the disease directly is relatively difficult in children, experts say. Because children’s symptoms tend to be mild, few show up at hospitals or clinics seeking testing or treatment and don’t get picked up by public health departments — an important source of data. Although researchers are working quickly, the relatively slow pace of the academic publishing world also means that many coronavirus studies in children aren’t yet peer-reviewed.


Researchers don’t need a definitive understanding of transmission in children before policymakers decide to open schools, said Simon Clarke, associate professor in cellular microbiology at the University of Reading in the United Kingdom. But schools should be cautious about how and when they resume.

“We don’t know and we probably shouldn’t try to find out what would happen if children were let loose to socialize and go to school exactly the same way as they were six months ago,” he said. “It will take time to get a fuller picture.”

Getting a “fuller picture” may happen faster through studies that draw on sociological and other types of nonbiological information.

For instance, preliminary results from an unpublished retrospective study suggest U.S. states that closed schools soon after local outbreaks began had fewer deaths per capita — and slower growth in death rates — than those that waited.

The findings fit with studies on school closures during past pandemics, such as the 1918 influenza outbreak, and are among the first to hint at the effectiveness of this specific strategy for curbing COVID-19.


The Brown University researcher who led the study, Emily Rauscher, paired COVID-19 case data collected by The New York Times with data on school closure dates from Education Week. For every day that schools were open following a state’s 100th confirmed COVID-19 case, deaths per capita grew on average by about 1% faster than in states that closed schools the same day they reached 100 cases. The study didn’t examine what happened after schools closed.

“In terms of individual children, families and communities and for children’s social, cognitive and psychological experiences, it’s creating massive costs,” said Rauscher, associate professor of sociology. “But [closing school] does seem to be related to slower increases in deaths.”

“Another future consideration is, if or when a second wave starts, school closures could be one piece of a containment measure,” she added.

But school closures are only one part, Rauscher said. For instance, Washington was one of the last states to close relative to when its 100th case was logged — but it didn’t see the same rise in deaths that other states witnessed after that point. This suggests that other preemptive social distancing measures, such as closing businesses and workplaces, were also effective.

Another recent study published in the journal Science used a mathematical model to predict how well school closures helped reduce incidence of the virus.

Drawing on data from China that estimates the number of typical contacts children have on school days, and their relative risk of contracting coronavirus, the scientists found that proactively closing schools cut peak coronavirus incidence rates by 40 to 60%.


The estimate isn’t perfect: The researchers’ calculation relies on the unproven assumption that children transmit the virus just as well as adults.

“What we get is the best estimate at the moment,” said Marco Ajelli, associate professor at Indiana University in Bloomington, who led the study while he worked at the Bruno Kessler Foundation in Trento, Italy.

Ajelli and his colleagues are now examining additional data on these children’s contacts — a method that should help the team refine their analysis and answer questions about transmission patterns in children.

Absent hard data on either question, schools in some countries have still chosen to reopen.

In Germany, Denmark and China, schools are phasing in reopening with a new set of social distancing measures. Desks are far apart. Hand-washing is constant. Some schools require masks.

Schools are beginning to open in pockets of the U.S., too. At an elementary school in Montana, where a handful of rural schools opened this month, teachers hold pool noodles to remind students to keep a distance from one another.


In Washington, state education officials are looking for lessons from places that have returned to school in-person.

“It is both interesting and useful to watch and learn how other governmental entities are approaching this unprecedented challenge,” said Katy Payne, spokesperson for Washington’s Office of Superintendent of Public Instruction.

Washington officials are just beginning to hash out what learning will look like next school year. Barring a vaccine and its widespread distribution, or a serious reduction in transmission, state education leaders say school as usual isn’t likely. A 123-person work group is studying a set of six other scenarios, which range from remote learning to rotating students through buildings so fewer are on any campus at a time.

As for evaluating emerging research, Payne said: “We see this as the role of our public health partners.”