How do you reopen universities during COVID-19?

A network analysis showing “high clustering and short average path lengths” in a university setting as a representation of potential transmission risks of the novel coronavirus in a campus environment.

EDIT: I am adding additional relevant links to the list at the bottom of the story.

Deciding how and when to reopen colleges and universities will be as contentious and a difficult as the question of when to reopen the economy in general. And with a few additional complications:

• Medium/large schools with 20–60,000 students and 5,000–10,000 faculty and staff are like small cities. But most of the students come from other cities/state/countries (and travel back to those homes somewhat regularly) which dramatically diversifies the pool of potential transmission risks.
• Many students live in dorms or off-campus apartments that bring them into regular close contact. Or they eat in on-campus dining halls or gather in the student union or library or large lecture halls. Or bars, restaurants, grocery stores…
• Schools have a responsibility to provide both a safe environment and one as stable as possible and conducive to learning.

Those, among other reasons, make the problem amazingly complex. And if you do reopen but need to rapidly decrease social contact mid-semester you introduce more cost and disruption than if you remained online for the fall.

Clearly every school and university system has experts working on the question — Boston University’s recovery plan considers the impact of a January 2021 reopening. But how long can we wait this spring to make the decision?

It is like trying to call a snow day three months in advance. And we need those three months to build snow shovels. Because the one or two weeks we got to “move all of our classes online” in the spring was both unavoidable and something we should not repeat. To redesign every courses for the fall, three months seems like a minimum requirement. Call June 1, 2020 as a working deadline to consider.

So what are the criteria we should be thinking about? I would boil it down to:

1. A vaccine OR
2. Serology tests to check immunity AND quick-turnaround diagnostic tests AND the ability to effectively contact trace to isolate/quarantine.

There will be no vaccine before next year. And there is some evidence in Korea that exposure and recovery from the disease may not be enough to assure immunity. So as of now neither of those options are helpful for the fall. Faster diagnostics tests might be helpful. Potentially effective contact tracing methods are being developed, but how useful might those be on densely networked campus?

We have all become epidemiologists and statisticians and crisis managers in the last two months, with all of the caveats that implies. The above are my moderately-informed opinions, but I am reading what I can on the topic:

• The Small World Network of College Classes: Implications for Epidemic Spread on a University Campus (Twitter thread and preprint)
• Coronavirus: How should US higher education plan for an uncertain future? (McKinsey)
• Coronavirus and US education (McKinsey report for investors)
• Colleges are not light switches (Inside Higher Ed)
• National coronavirus response: A road map to reopening (AEI)
• Lesson From Singapore: Why We May Need to Think Bigger (NYT)
• WHO Sets 6 Conditions For Ending A Coronavirus Lockdown (NPR)
• WHO unsure antibodies protect against COVID, little sign of herd immunity (Reuters)