When panicked administrators shut down schools in the spring, little was known about how the coronavirus could spread among students and teachers. Could children fall critically ill and spread it to peers and teachers, like the flu? Would asymptomatic young people pass it on to their parents or educators?
More than nine months after schools closed, some of the answers to those questions are becoming clear. Emerging data on contact tracing – which illuminates the origins of infections – shows that the virus does not seem to spread much within schools when they require masks, urge social distancing, have good ventilation and when community spread is low.
But because of a lack of a cohesive federal response, huge gaps in the data remain, and many say new information about school transmission is not sufficient to make far-reaching conclusions. A dearth of data has plagued many aspects of pandemic response, and it has left governors, school superintendents, school board members and parents on their own to interpret the shifting body of knowledge as they make decisions that could affect the lives of everyone connected to school communities. This, coupled with soaring infection rates, has made these decisions especially fraught as school officials weigh whether to reopen their doors next month.
The results? School systems have responded to the virus in wildly different ways that often seemed more shaped by politics than data, according to an analysis conducted by the Center on Reinventing Public Education.
“It’s ridiculous. The districts are having to figure this out on their own and manage the local politics,” said Robin Lake, the director of the center, an education research group that has been tracking school closures. “It puts them in a pretty untenable position.”
Plenty of school systems report the number of infections that emerge among students and staff. But many experts and public health officials say data about where the infections originate is just as important. School transmission data, gleaned from contact tracing, can show whether students and teachers are getting sick from being at school, or because they picked up the virus at a child’s sleepover or a family birthday party.
“It’s incredibly important to have that second piece of data, that context, because it helps us understand, are these cases being acquired external to the school environment,” said Jennifer Kertanis, the director of the Farmington Valley Health District in Connecticut and the president of the National Association of County and City Health Officials. None of the students or staff who have contracted the virus in the 10 school districts she covers have caught it from school.
Researchers say that in some cases, closing schools, and leaving children in the care of adults who do not force them to wear masks or socially distance, may put them at higher risk of contracting and spreading the virus.
“If kids or teachers are not in schools, they may be in child-care centers or learning pods that are also causing community spread,” said co-author Dan Goldhaber, who directs the Center for Education Data and Research at University of Washington. “From a public policy standpoint … it’s not necessarily safer to have the schools closed.”
A study by the Centers for Disease Control and Prevention of Mississippi schoolchildren found those who tested positive for the virus were not more likely to have attended in-person classes than those who tested negative. But they were more likely to have attended family gatherings, play dates, parties and funerals.
Researchers, too, are getting more information about when and how schools can contribute to spreading the virus in the community. A study by researchers in Michigan and Washington state examined school and infection data, and found that when community infection rates were low, reopening schools did not seem to worsen the situation.
But that calculation changed when infection rates were high – like they are now in Washington, Michigan and wide swaths of the country. Then, schools did seem to contribute to community spread. According to the researcher’s metrics, both states now greatly exceed that threshold, which appeared to be around 5 to 15 new cases per 100,000 people. On Dec. 27, Michigan was reporting a 7-day average of 29 new cases per 100,000 people. That same day, Washington’s 7-day average was 28 cases per 100,000.
There is debate among public health experts and epidemiologists over whether the data is sufficient to draw any conclusions or make recommendations, especially because only a handful of jurisdictions are reporting it, and because there’s no uniform way to gather it. They worry that schools that decide to remain open as infections rise to unmanageable rates, especially those that do not mandate masks or limit activities, could end up significantly worsening the virus’s spread.
“So there are two issues: One is that we don’t have enough contact tracing all across the country,” said Leana Wen, a physician and public health professor at George Washington University who previously served as Baltimore’s health commissioner. “The second problem is that the community prevalence is just so high that it’s going to be very difficult to sort out where the infections are originating from.”
And not all the data points to low incidence of school spread. St. Charles County, Mo., one of the few jurisdictions to post detailed case information about school spread, linked more than 1,100 coronavirus cases to school transmission. There is little data on what happens in schools where students are not required to wear masks or socially distance – as is the case in many schools in Georgia and other southern states.
Even as some are urging schools to reopen, infection rates are skyrocketing in many parts of the country, feeding growing tensions between those who wish to reopen buildings, including many politicians and school officials, and those who caution that we still do not know enough to assure safety, including teacher unions. Parents stand on both sides of the debate, but Black and Latino families appear more likely to advocate to keep schools closed.
The Chicago Teachers Union on Dec. 7 filed a petition for an injunction from the Illinois Educational Labor Relations Board to stop the city from reopening schools. Teachers unions in New Jersey have criticized Gov. Phil Murphy, a Democrat typically allied with them, of playing down the risks of the virus.
The number of infections among children has also risen sharply in the United States, with more than 1.3 million testing positive since the onset of the pandemic. But many children appear to be catching the virus from peers off school grounds, like at sleepovers or parties, or from their parents.
More than nine months into the pandemic, parents, educators, policymakers and even scientists remain divided on whether it is prudent to reopen school buildings. Part of what has stumped them is the lack of data about transmission, which could help them understand what kinds of measures make a difference, if teachers with underlying health conditions should stay home, or whether schools should reopen at all.
The decision is among the most difficult and most consequential one governors, local school boards and superintendents have been forced to make during the pandemic, with experts warning that school closures could have lifelong consequences for the most vulnerable students.
The dearth of information – and abundance of political forces – has led different systems to make radically different choices.
The World Health Organization recommends communities shut down when positivity rates hit 5 percent. New York Mayor Bill de Blasio (D) closed all schools when the citywide positivity rate hit 3 percent, though he reopened schools for special-education students and the city’s youngest students on Dec. 7. By contrast, most European countries have kept schools fully or partially open while weathering huge spikes in infections that rose higher than in the United States. The city of Sioux Falls, S.D., has kept its schools open even as it experienced a rate of more than 100 new cases a day per 100,000 people.
Those who want to see schools remain closed point to the lack of understanding of how well children spread the coronavirus, and evidence they can be significant drivers inside households. Teachers unions in many places have pressed to keep schools closed, worried for their members’ safety. Black, Latino and Native American communities hard hit by the virus have been reluctant to send their children back into school buildings, having seen the wrath the infections can bring. Many epidemiologists side with doctors like Wen, who says it is too soon to understand what conditions a school can safely reopen in.
Stacy Davis Gates, vice president of the Chicago Teachers Union, said data that shows low transmission rates does little to assure her about the safety of the city’s schools, which have remained closed to schoolchildren since March. It cannot tell her how well the virus would spread in a school system with poorly ventilated classrooms or a dearth of school nurses, custodians and contact tracers – all conditions teachers might encounter when the city reopens schools.
“You don’t make high-stakes decisions with incomplete data,” Gates said.
President Trump, who peddles misinformation about the pandemic, has called unrelentingly to reopen schools since the summer, when there was little information on whether schools were safe. Schools in Republican strongholds that had few virus restrictions to begin with enthusiastically reopened, often without mask mandates.
But even as many Democratic-run states and cities keep schools closed, others in the party have shifted their views. At a Dec. 8 event, President-elect Joe Biden said he aimed to get 100 million Americans vaccinated and to have most schools open within 100 days of the start of his administration. Democratic governors have also joined the call to reopen schools, including Murphy in New Jersey and Govs. Tom Wolf of Pennsylvania and Gina Raimondo of Rhode Island.
So have parents across the ideological spectrum, and some prominent epidemiologists, such as Ashish Jha, a physician and the dean of the Brown University School of Public Health, who said in an interview that the data, while less than perfect, suggests that schools are not driving transmission.
Despite the Trump administration’s push to reopen schools, there is no federal effort to collect data on coronavirus cases that emerge on school campuses. The Centers for Disease Control and Prevention, whose director, Robert Redfield, has said schools can safely reopen, is relying on analysis from Emily Oster, a Brown University economist who solicited several school districts to submit data voluntarily. States and counties collect, tabulate and publish data in different ways, and jurisdictions overwhelmed by coronavirus cases often do not have the resources to track down the origin of an infection.
But those who have managed contact tracing have found surprisingly low levels of transmission.
Contact tracing works by piecing together all of the places and people an infected person might have come into contact with over the past few days, and who might have passed the virus on to them. The tracer also uses that information to warn anyone who had contact with that person that they should quarantine. In some schools, school nurses have reviewed surveillance video to determine who the infected student came into close contact with.
For example, out of all the schools in Illinois, an analysis by the Chicago Sun-Times found 16 schools where transmission probably occurred.
Daniel Benjamin, a professor of pediatrics at the Duke University School of Medicine, is part of a collective of scientists advising North Carolina schools. Last month, he examined data from 11 school districts that have hosted more than 90,000 students and school staff since the start of the school year and found that 805 of them contracted the virus. Of those, 32 of them contracted it while at school.
“To put that in context, in North Carolina right now, every North Carolinian who gets infected probably is transmitting somewhere between 1.0 and 1.2 other people,” said Benjamin. “But in North Carolina schools, it takes somewhere between 10 and 20 children to get one additional child infected.”
In New Jersey, Murphy allowed school districts to choose whether to reopen, angering the teachers union, which feared widespread school outbreaks. Three months in, the state’s health department has seen 459infections linked to schools. The data led Murphy to take a stronger stance on reopening schools, issuing a statement along with several other Northeastern governors urging districts to reopen buildings.
“We’re not batting a thousand – let there be no doubt about it,” Murphy said the first week of December. But, he said, “the in-school experience, if it can be done safely, and so far we believe the data suggests it can be in the overwhelming amount of circumstances . . . is the far preferred and the far richer experience.”
In Waukesha County, outside of Milwaukee, schools quarantine students and staff whenever they have close contact with a person who tested positive for the virus – even if they were wearing masks. The school system has ordered quarantines 10,000 times for students and staff, but testing has turned up 52 positive cases among those who were exposed.
Kertanis, in Connecticut, oversees a health district with over 100,000 people and several school systems, which started out the school year with hybrid instruction – having students come to school part-time and learn virtually the rest of the time. Once students had a chance to get used to the new protocols – and when officials saw no cases of school transmission – school systems brought students back full-time.
Several weeks into the school year, even though many students and staff members have contracted the coronavirus, not one of the infections came from inside the school, she said. It helps, too, that schools are eager to work with her and readily provide information about cases arising inside schools.
“One of the most important things that we can do right now during this pandemic is provide the kinds of data that allow our schools to make good decisions about in-person learning versus hybrid versus remote,” she said.
“Because the long-term implications of this pandemic on our youth are going to be significant.”