A new study in Illinois shows that, while in-person learning in schools might lead to the spread of coronavirus cases, hybrid learning plans might not be worse for community spread than full remote learning.
Gary Reinbold, associate professor and faculty associate at the University of Illinois Springfield, said he “wouldn’t bet his life” on his research alone. But a study in Spain and contact tracing data in Utah suggest that schools are not increasing coronavirus cases in the community.
“When you use a few different types of research and get the same results, you start to get more confident,” he said, comparing his work to other studies.
Reinbold’s study doesn’t go quite so far as the others: In-person learning looks like it has had some negative effect in Illinois, but hybrid learning and completely remote learning were statistically very similar, suggesting that a mix of digital lessons with some in-person school attendance might be as safe as keeping students at home on their computers.
The Illinois Department of Public Health released contact tracing data Friday, including outbreaks and potential exposures within schools. In the 30 days before the data was released, IDPH reported about 10.5% of positive cases said they had been in a school in the 14 days before symptoms started or a test was taken.
How the study works
Reinbold classified Illinois counties by whether the majority of students were learning in-person, remotely, or through a hybrid learning plan.
In hybrid plans, students typically go into school for half the time and work from home the other half, so buildings are never at more than half capacity on a given day, which allows for social distancing.
Reinbold matched counties with similar demographics, including poverty rate, population density and a population breakdown by race. This method allows for comparison between counties with different plans but similar demographics in order to account for other contributing factors to the number of COVID-19 cases, he said.
Reinbold then looked at the number of cases and deaths in the weeks before and after most schools started in the fall.
In-person learning was shown to have the greatest effect on the number of cases, increasing the number of cases on a seven-day rolling average per 100,000 people by up to 15 cases per day over counties with online learning.
The study shows slight differences between majority hybrid and majority online-only counties. Reinbold estimated that, compared to hybrid instruction, online-only might result in one to four fewer new daily cases per 100,000 people.
But that doesn’t mean every school is equipped to do hybrid learning safely, Reinbold said.
“In particular, there may be important differences in facilities or resources between districts that started the school year with hybrid instruction and districts that started with online-only instruction, such that the online-only districts would not have been able to offer hybrid instruction as safely as my results suggest that the hybrid districts did,” Reinbold wrote in his report.
No learning plan had a statistically significant impact on the deaths per 100,000 in the county, according to the study.
Data limitations
There are limitations to the analysis, Reinbold said. Both the study in Spain and the contact tracing in Utah also have their limitations, he said, but as all three were conducted differently, they have different limitations.
Many COVID-19 cases are still undetected and unreported, especially if the individual isn’t showing symptoms. The data looks at cases by county even though school staff, in particular, might live in a different county than where they work.
In some regions, including the metro-east, COVID-19 mitigations have been more restrictive in the past months, which also influences case numbers.
Some school districts are changing plans, based on conditions in the buildings and community. At various points this semester, Freeburg Community High School 77, Collinsville Community School District 10 and Dupo School District 196 have all changed from hybrid learning to remote lesson plans because of positive cases within the schools, among either staff or students.
Districts have moved the other way, too, from remote learning to hybrid learning.
Nearly all of the school districts in St. Clair County started the year remotely at the behest of the health department. In the months since, hybrid learning has started in Belleville Township High School 201 and O’Fallon Township High School 203, and Central School District 104 started fully in-person learning at the end of October.
Practically, school districts need to be flexible and make changes, but Reinbold said that makes it harder to see the effects return-to-learn plans might have on community spread of the virus. You can’t lock a school district into its decision for the sake of research, he said.
“As school boards, you’d be in a lot of trouble if you didn’t change plans when things were going south,” Reinbold said.
New contact tracing data isn’t as helpful as expected
Before IDPH released the contract tracing and school outbreak data Friday, Reinbold said that information should have helped give context as to whether schools are contributing to the rise in COVID-19 numbers.
In actuality, Reinbold didn’t think it was detailed enough to help school districts make informed decisions, even if it did corroborate conclusions drawn from his own study.
“It isn’t really telling us a lot,” Reinbold said.
At least two of the schools IDPH reported as having outbreaks in the last 30 days refuted the report. Of the remaining eight, six have conducted fully in-person learning at some point since the start of the school year. Administration at those schools either did not return phone calls or refused to comment.
One, Galesburg Christian School, touts its in-person learning on their website’s homepage.
In general, Reinbold said people could make better decisions and focus on the right places for COVID-19 spread if there was more thorough contact tracing.
“The first couple of weeks, when we first started opening up, I’d go to a restaurant or a barber shop and they’d take my name and number,” Reinbold said. “I thought it made a lot of sense, but it was over in like two weeks.”
IDPH reported a breakdown of contact tracing data from the 30 days before it was released by location, but noted that one positive case could be marked down for more than one location, depending on what the person disclosed to the contact tracer. But even if only one person corresponded with each of the 35,912 locations, that’s still data for only 22% of the more than 160,000 new cases reported in Illinois in the same time period.
Without substantial and robust contact tracing, Reinbold said, there’s also a question of sourcing: Are schools causing community spread, or mirroring what’s happening with the number of coronavirus cases outside of the building?
Since Belleville Township High School 201 began hybrid learning, the district has been tracking both the number of cases and the transmissions within the school by using contact tracing to see if an individual might have spread the coronavirus to anyone else in the building.
Contact tracing is a formalized attempt at identifying people who might have come into close contact with an infected person and where that contact was made. The Centers for Disease Control and Prevention define close contact as being within six feet of each other for more than 15 minutes within a 24-hour period.
District 201 Superintendent Brian Mentzer said that, as of Thursday, individual cases of COVID-19 have been documented but that there have been no transmissions, even as county, state and national numbers indicate a fast and wider spread of the virus. More than 10,000 Illinois residents have died of COVID-19, as of Thursday.
Region 4, which includes St. Clair County, where Belleville is, has had 14 consecutive days of a seven-day rolling average positivity rate above 8%, as of Nov. 2 — the data reported by the state is delayed.
“I think right now, obviously, with the community, the increase in the numbers nationwide, we are looking very closely at what’s going on, to make sure we don’t get in a situation where we’re vulnerable,” Mentzer said. “ … Any time there’s more of the virus in the community, there’s a chance of more of the virus in the school.”
John Wagner, director of the Monroe County Health Department said that it was difficult to link youth cases to school unless there was something large in a single classroom.
“We’re not seeing a whole lot of spread in the classroom,” he said. “ … Not to say that’s not happening at all. There are some — it’s just not where most of the spread is coming from.”
Looking at weekly youth cases
The most useful information IDPH released Friday isn’t about contact tracing data at all, Reinbold said: It’s weekly casa data for youths, broken down by groups ages 5–11, 12–17 and 18–22.
Between Aug. 15 and Sept. 19, there was a spike in the number of cases among 18–22 year olds, who were returning to college around that time. There were 1,693 new cases reported the week of Aug. 15, but by Sept. 5, new weekly cases were peaking at 3,252 per week before trailing back down.
Since Oct. 3, the number of cases for all three groups have trended upward, but children aged 5–17 never saw a peak when school started.
“You never see anything like that with the school-aged kids,” Reinbold said. “ … Lately, the numbers have been drifting up, but they’ve been drifting up for all age groups. It doesn’t look like they’re specific to schools.
Megan Valley covers education for the Belleville News-Democrat, a news partner of St. Louis Public Radio.
