NEW YORK – A new modeling analysis shows colleges and universities could open safely for the fall semester with a strict COVID-19 testing regimen. However, the research team cautioned that the logistics, cost, and associated behavioral interventions needed may make opening impractical for some schools.
While some US colleges and universities have announced their fall semesters will be held virtually, others are weighing whether they can safely open to students in the midst of the COVID-19 pandemic.
In their models, David Paltiel and his colleagues of the Yale School of Public Health explored various testing frequencies, test sensitivities and specificities, and viral reproductive numbers for a hypothetical cohort of 5,000 students with 10 initial cases. As they reported in JAMA Network Open on Friday, they found that screening students every other day, even with a test with low sensitivity, in conjunction with behavioral interventions, could keep the number of COVID-19 cases on campus at a manageable level. However, implementing a testing program may be challenging.
"The safe return of students to residential colleges demands an effective SARS-CoV-2 monitoring strategy," Paltiel and his colleagues wrote in their paper.
For their model, the researchers envisioned a scenario in which a shortened, 80-day semester began with 4,990 uninfected students and 10 students with undetected SARS-CoV-2 infections. They modeled situations in which students underwent testing every day, every other day, every three days, and once a week, using tests with sensitivities between 70 percent and 90 percent and specificities between 98 percent and 99.7 percent.
They assumed anyone who tested positive for SARS-CoV-2 would be able to enter an isolation dorm within eight hours and undergo confirmatory testing. Those with false positives would be isolated for 24 hours and those with true positives for 14 days.
In their base scenario, which assumed a viral reproductive number (Rt) of 2.5 — indicating that every infected person would infect about 2.5 others — and 10 outside infections each week, the researchers calculated that testing every day with a test with 70 percent sensitivity and 98 percent specificity would lead to 162 infections over the course of the semester. They further estimated that this would lead to a mean daily occupancy of the isolation dorm of 116 people, 18 percent of whom would have true positive tests.
Testing every other day, meanwhile, would lead to fewer people in isolation, but a greater total number of infections, 243. Weekly testing would lead to a daily average of 121 people in the isolation dorm, 90 percent of whom would be true positives, and a total number of 108 infections.
The number of infections varied in other scenarios, with higher or lower viral reproductive numbers and different numbers of outside infections coming in. Screening based only on symptoms was not enough to contain the outbreak, they added, as it would lead to nearly the entire student population being infected by semester's end.
The researchers concluded that screening students every two days with a rapid test of more than 70 percent sensitivity, in combination with behavioral efforts, such as wearing masks indoors, frequent hand-washing, and limited bathroom sharing, could contain the number of COVID-19 infections.
Through a cost-effectiveness analysis, they added that screening using a test with 70 percent sensitivity every other day, every day, or weekly, would be the preferred approach in scenarios with Rts of 2.5, 3.5, or 1.5, and cost $470, $910, or $120 per student for the semester.
However, their analysis has a number of limitations. For instance, it elides the effect of screening faculty and staff and the effect of student infections on the wider community.
Additionally, putting this scenario into place would be a logistical challenge. It would require a large number of testing kits — about 195,000 kits if testing is done every two days — to implement large-scale screening with a fast return of results, a means of monitoring adherence, and the maintenance of an isolation dorm, among other challenges, the researchers noted.
In an editorial accompanying the paper, Vassar College's Elizabeth Bradley and her colleagues noted that whether it is wise to reopen depends on each school's particular circumstances. They applied Paltiel and colleagues' model to their own school — which has a smaller population size — and made slightly different assumptions about the number of students arriving with undetected infections and about the viral spread. With those tweaks, they found testing might only be needed every two weeks to reach no more than 50 total infections.
Still, they wrote that the findings of this study are valuable, as it shows that commitment is needed to keep case numbers down. "If a college is unable to reduce the influx of COVID-19 from outside and limit Rt on campus, far more testing is needed than current guidelines require," they noted.