On Monday, after just one week of in-person classes, the University of North Carolina at Chapel Hill reported 135 new coronavirus cases among students and staff over seven days, and announced it would go all-online for undergraduate instruction. The next day, the University of Notre Dame took instruction online for at least two weeks, after tallying 147 positive cases since the beginning of the month (that number has since increased to 222).
These early experiences suggest that learning in person this fall may be harder to pull off than some college leaders anticipated. Should other colleges take the experiences of UNC and Notre Dame, which started their semesters relatively early, as a warning?
Public-health experts interviewed by The Chronicle responded with a resounding “yes.”
I really do believe that if all the earlier assumptions would have played through, that we would have had a very good shot at making this work.
“For universities going back to more or less in-person operations, this is extremely likely without aggressive testing plans,” said Carl T. Bergstrom, a professor of biology at the University of Washington, who previously argued in The Chronicle that widespread testing of asymptomatic people on campuses is necessary. “I think it was entirely predictable.”
Spread could be happening at a greater scale than we know. “I think UNC has been forthcoming in the cases that they are having. We don’t know if other universities are letting us know what their numbers are,” said Reginald Fennell, a professor emeritus of kinesiology and health at Miami University, in Ohio, and a former president of the American College Health Association. “It’s unlikely that they are the only institution in the United States that is having this issue.”
At UNC, the county health department thinks that some number of students very likely arrived on campus already infected, although it doesn’t know that for sure, said Todd McGee, a health-department spokesman. Then, as students lived and interacted together in their first weeks back on campus, the virus may have spread.
After reporting four Covid-19 clusters in three days, the administration felt it couldn’t acceptably contain its outbreaks with students remaining in dorms and classes, as originally planned. “The current data presents an untenable situation,” Kevin M. Guskiewicz, the chancellor, and Robert A. Blouin, the executive vice chancellor, wrote in announcing that the university was going all online.
Guskiewicz and his administration defended their return-to-fall plan in a faculty meeting held immediately after they announced UNC’s flip to online. “I’m really proud of what we put in place within the infrastructure of the classrooms and the dining halls and the student union,” Guskiewicz said.
Blouin said that when the administration began planning for the fall, the prognosis looked different. “All the forecasts back then indicated that by the summer, we would have a relatively low density of virus and a high availability of testing. That didn’t turn out quite the way that we had expected,” he said. “I really do believe that if all the earlier assumptions would have played through, that we would have had a very good shot at making this work.”
The apparent silent spread of coronavirus at UNC underscores the need for residential colleges to test students as they return to campus, proponents say, and to test students, staff, and faculty widely and frequently, even if they aren’t feeling Covid-19 symptoms. UNC had planned only to offer tests to those who sought them, and people identified as close contacts of confirmed positive cases.
“Young people especially may be asymptomatic,” said Julia L. Marcus, an infectious-disease epidemiologist at Harvard Medical School. “Testing only based on symptoms is inadequate to prevent spread.”
In its guidance for institutions of higher education, the Centers for Disease Control and Prevention doesn’t recommend all students, staff, and faculty get tested before they return to campuses. The public-health agency also didn’t explicitly recommend frequent surveillance testing for colleges, and many aren’t doing either, citing the CDC’s guidance as the reason.
Outside experts The Chronicle spoke with disagreed with the CDC’s stance. A. David Paltiel, a professor of public health at the Yale School of Medicine, led a team that modeled various possible testing plans for residential colleges. None of the scenarios they modeled suggested testing only symptomatic students and their contacts would be enough to contain a coronavirus outbreak, Paltiel said: “I’m really distressed by the fact that so many schools are issuing themselves a free pass to forgo any kind of re-entry testing.”
Young people especially may be asymptomatic. Testing only based on symptoms is inadequate to prevent spread.
Notre Dame did mail its students coronavirus testing kits and required them to get a negative result before returning to campus. Before the start of the semester, 99.7 percent of them tested coronavirus free. So what happened? Students might have picked up the virus from community members in the South Bend region, then spread it. Or they might have been infected in the dayslong period between being tested and returning to campus. “Tests are not perfect, and human behavior is not perfect, so the combination of those two factors will likely result in some cases on campus,” Marcus said. “The question is: What is the scale, and how are those outbreaks contained?”
Together, North Carolina at Chapel Hill and Notre Dame show that the coronavirus has many ways to sneak onto campus, and that an in-person fall plan can fall short very quickly.
Experts acknowledge that frequent and widespread testing isn’t financially feasible for many colleges. In some communities, tests with a short enough turnaround time to be useful — 24 to 48 hours — may not be available, no matter how much money you have. In these cases, Bergstrom and Paltiel recommended colleges operate mostly online.
For those that do continue in person, Paltiel and Marcus recommended a step that has been uncommon among fall plans: positive alternatives to riskier student social gatherings. At both UNC and Notre Dame, leaders have blamed after-hours student socializing for their case spikes. But “just yelling at them for having parties” isn’t effective, Marcus said. Research in other arenas of public health, such as mental health and drug use, has found that stigma can prevent people from getting help, and that “Don’t do it” messages don’t work unless they’re paired with active steps people can take instead.
In an essay in The Atlantic, Marcus suggests “movies, yoga classes, and concerts — all outdoors with social distancing encouraged and school-branded masks provided.”
Paltiel had a more colorful idea. “The thing that I’m trying to avoid here is the frat-house basement, unventilated, unmasked, 150-students-in-a-crowded-space superspreader event,” he said. “If the price of averting that event is that I’ve got to let students play beer pong and do Jell-O shots outside, on the quad, in small groups, and turn a blind eye to the underage drinking, that’s fine! Grow up! It’s fine. I’m trying to avoid ‘basement party.’”
Although risky parties might be the proximate driver of on-campus infections, the experts pointed to larger, more systemic problems, including the U.S. government’s failure to ensure that coronavirus testing is cheap and easily available across the country; the CDC’s not recommending more widespread testing for colleges; and college administrators not requiring such tests anyway, despite the CDC’s omission.
“There’s blame-shifting all the way down from the top, down to the students. That really concerns me,” Marcus said. Student gatherings are just the end of the line. Prevention should have started at the beginning.
