Thousands of students and teachers have become sick with COVID-19 since schools began opening last month, but so far, public health experts have found little evidence that the disease is spreading inside buildings, and the rates of infection are far below what is found in the surrounding communities.

This early evidence, experts say, suggests that opening school may not be as risky as many have feared and could guide administrators as they charter the rest of what is already an unprecedented school year.

“Everyone had a fear there would be explosive outbreaks of transmission in the schools. In colleges, there have been. We have to say that to date, we have not seen those in the younger kids and that is a really important observation,” said Michael Osterholm, director of the Center for Infectious Disease Research and Policy at the University of Minnesota.

This does not mean the risk of contracting COVID-19, the disease caused by the coronavirus, is zero. Poor and inconsistent reporting in many parts of the country means experts do not yet have a full view of the situation, and most schools have only been open for a matter of weeks. It’s also not yet clear how closely the incidence of COVID in schools is tied to policies inside schools such as mandatory mask-wearing.

Most of the nation’s largest districts opened with fully remote teaching, so the data to date is largely from smaller communities. And the pandemic may grow worse as flu season and winter approaches.

Students arrive for classes at the Immaculate Conception School in The Bronx on Sept. 9. Associated Press/John Minchillo

But the fact that large swaths of the country opened for in-person school while others did not offers the more cautious districts a chance to observe how things have gone elsewhere in charting their next steps.

On Wednesday, researchers at Brown University, working with school administrators, released their first set of data from a new National COVID-19 School Response Data Dashboard, created to track COVID cases. It found low levels of infection among both students and teachers.

Tracking infections over a two-week period beginning Aug. 31, it found 0.23% of students had a confirmed or suspected case of COVID-19. Among teachers, it was 0.49%. Looking at just confirmed cases, the rates were even lower: 0.078% for students and 0.15% for teachers.

“These numbers will be, for some people, reassuring and suggest that school openings may be less risky than they expected,” said Emily Oster, an economics professor at Brown University who helped create the tracker. She noted that the school COVID-19 rates are “much lower” than those in the surrounding community.

Still, she said: “I don’t think that these numbers say all places should open schools with no restrictions or anything that comes close to that. Ultimately school districts are going to have different attitudes toward risk.”

The information for her dashboard is voluntarily reported by schools and school districts, both public and private, including schools that offer in-person classes and those that are fully remote. As of Wednesday morning, the project had data from more than 550 schools, including more than 300 that have some in-person classes. Organizers are working to add more schools as they go.

Separately, early data from the state of Texas also shows low levels of infection. In Texas, about 2,350 students reported positive COVID tests – or about 0.21% of the 1.1 million students attending school in person, according to data released last week. Another 2,175 school employees tested positive, though a rate could not be calculated because it was not clear how many of the state’s more than 800,000 school staff were working in school buildings.

Teachers unions in Texas keeping track of infections say they have been surprised how low it was. In many parts of the country, teachers unions have resisted school systems’ efforts to return to classes, saying there were not sufficient safeguards in place.

“I am not seeing at this particular point the rate I had expected,” said Zeph Capo, president of the Texas branch of the American Federation of Teachers.

He said this is partly because parents in communities most affected by COVID are less willing to send their children back to school. And he predicted the numbers will rise as more students to return to buildings and if the pandemic worsens this winter, as experts have warned.

There’s also evidence from the Northeast. The Network for Public Education, an advocacy nonprofit organization that supports traditional public school districts, has been tracking 37 school districts in New York, Connecticut and Pennsylvania.

In the weeks since school has started, there have been 23 confirmed cases of COVID-19 across 20 different schools and no indication that disease was spread in schools, said Carol Burris, the network’s executive director.

The districts studied all were in counties with low COVID rates and all required wearing masks.

“So far, in the schools that we are following . . . outbreaks have not occurred, even when someone tests positive for COVID-19,” Burris said.

“We’re not seeing schools as crucibles for onward transmission,” said Sara Johnson, associate professor of pediatrics at the Johns Hopkins University School of Medicine. “It’s reasonable to say that it looks promising at this point.”

She added that data suggest schools should bring students back “slowly and carefully” and with safeguards to protect teachers and staff. “These data are promising but COVID is still a very big threat to people,” she said.

The early data are emerging as school officials continue to evaluate their plans and consider whether they want to change course. Many districts that began with entirely remote education are considering whether they want to introduce in-person options for some students or on certain days.

The data are helpful but superintendents will want more detail, such as information about districts similar to their own, and what COVID mitigation strategies seem to have worked, said Noelle Ellerson Ng, an associate director at AASA, The School Superintendents Association, which is a co-sponsor of the new dashboard reporting tool.

She called the early data “potentially optimistic” but said it is premature for districts to change course and decide to open their buildings.

Much of the concern is focused on teachers and other adults in the building, since data shows that COVID is far more deadly to older patients.

According to the American Federation of Teachers, there have been 14 active teachers, principals and counselors who have died of COVID-19 since the school year started, though it is not possible to conclude that any one of them contracted the coronavirus at school.

Some experts, including teacher union officials, say that it appears COVID rates are lower in school districts where face coverings are required and polices are enforced to keep distance between students in the building, though there is scant data to prove the correlation.

American Federation of Teachers President Randi Weingarten said the “science worked” in districts that took adequate precautions.

“We don’t have all the information, that’s true, but I’m not surprised that things were okay,” she said. She posited that early pictures of Georgia schools that went viral showing students packed into hallways without masks “scared people enough” to institute mask and physical distancing policies.

Across the country, though these policies vary widely. The new dashboard asked about COVID mitigation strategies and found about 7 in 10 schools required staff and students to wear masks. About half limited gatherings to 25 people or fewer. About 4 in 10 kept students in one classroom, with about the same portion checking temperatures upon arrival.

In suburban Atlanta, the Cherokee County School District encourages but does not require masks for students in most circumstances and only requires them for teachers and staff when they are not able to social distance. Officials have tried to reduce crowds in the hallway by staggering bell schedules and eliminated large gatherings such as assemblies.

The schools opened for in-person learning in early August, when COVID caseloads in the area were high, and since then, at least 381 positive COVID cases had been identified, each one triggering a two-week quarantine of teachers and classmates who may have been exposed. At one point, more than 2,329 people were in quarantine and three of six high schools were temporarily shuttered.

District spokeswoman Barbara Jacoby said that none of the cases had been definitively linked to in-school exposures though there were maybe a dozen where it had not been ruled out.

“Our community is entirely open,” she said. “You can go to not only school but youth sports, houses of worship, scouts, swimming pools. You can go anywhere. So it’s hard for public health officials to determine where transmission happens. A lot of these students interact outside school together.”

Last week, the number of students and staff in the district in quarantine was down to about 400, with 67 positive cases identified. The district has about 42,000 students.

Jacoby said that overall, things had gone “better than expected.”

One problem in evaluating school programs is not every school and not every state is providing useful data. Some districts in the country, like Cherokee, are reporting every case to the public. But others report nothing.

In Florida, where Gov. Ron DeSantis, R, has pushed schools to open even in areas with high coronavirus positivity rates, some countieshave been pressured not to release school-specific COVID-19 information.

A promised state school COVID-19 dashboard has not yet been produced, though there is a “pediatric report” that shows some data, by county, of cases for Floridians under 18 years old. According to the latest report, as of Tuesday, there is 13.7% rate of positive cases among young people from age 1 to 17.

Andrew Spar, president of the Florida Education Association, the state’s teachers union, said it’s hard to discern the true situation.

“There is no real reporting of cases going on in the state,” he said. “There is no mechanism for consistent and fair reporting so there is no way to analyze what is really happening in our schools.”