Students are beginning to return to campus at many colleges and universities this fall. Typically, they are required to be fully vaccinated before they arrive, and they are tested for Covid on arrival. But what happens then? How can a campus stay relatively Covid-free and nip new outbreaks in the bud, especially given that people can spread Covid to others before developing any symptoms themselves?
Here at Kendal, we are beginning a shift into new territory: a period in which symptomatic infections are relatively rare, but any case that does occur can trigger significant lockdowns and quarantines in order to avoid an outbreak. It is becoming more important than ever to detect infections as early as possible.
I decided to research what strategies colleges and universities were using, in case there were options that might be considered by a retirement community.
Let me emphasize that I am not proposing that Kendal adopt any particular strategy. Those decisions are up to the medical staff. But I am hoping my description of what is happening in the world of higher education might provide some useful background for residents as well as staff.
Before I go further, it’s worth mentioning the distinction between “asymptomatic” cases and “pre-symptomatic” infection. People who never show symptoms are termed “asymptomatic”. It is not yet clear whether (or to what extent) vaccinated people who have asymptomatic infections can infect others.
However, there is now plenty of evidence that people who are “pre-symptomatic” (infected and not yet showing symptoms, but who will develop symptoms later) can be very infectious. That includes vaccinated people. In some studies, the majority of transmission was found to be by pre-symptomatic infected people. Because of pre-symptomatic transmission, the symptoms of an outbreak might not be evident until many people had already been infected.
How do you detect an outbreak? If you want to stop an outbreak, you need to know within a few days—before infections have spread—that someone is infected. Earlier in the pandemic, most outbreaks were discovered because someone experienced symptoms and got tested. Then, contact-tracing was used to find and test others who might have been infected.
Now, with widespread vaccination, asymptomatic infections are common and symptoms are no longer a reliable warning sign. Colleges and universities have found they can’t afford to wait for symptoms. Even if no one is symptomatic, there needs to be frequent campus-wide screening and results need to be known quickly enough to isolate a source of infection and clamp down on transmission.
In an ideal world, you might have campus medical staff perform a PCR swab test on every student twice a week. That would probably catch an impending outbreak early enough to stop it, even if no one had yet developed any symptoms. As far as I can tell, based on web searching, no college or university is using that approach. It is too expensive, requires too many testers, and the logistics of performing all the PCR tests are too hard. Instead, other approaches to “Covid surveillance” are being tried. The approaches fall into three categories: pooled testing, self-administered testing, and sewage testing.
Pooled testing. Pooled testing represents an easy way to cut down on the amount (and expense) of testing required. Instead of performing a test on every sample, you mix multiple samples together and perform a test on the mixture. If the whole mixture is negative, you know that all the samples were negative. If the mixture is positive, you know someone in the group is positive and you do more testing to determine who is infected.
According to a National Academy of Sciences report, this approach is widely used. Duke is using it to test groups of 4 students at once, the University of New Hampshire is testing 5 at once, and Western Washington University is testing 10 at once. A report in the Journal of Occupational and Environmental Medicine discusses the uses of pooling in workplace settings where 30 or even 64 samples are tested at once. Pooling cuts testing costs dramatically, especially when relatively expensive PCR testing is used.
Although the cost of the actual test is reduced, the staffing required for swabbing remains the same. That’s a limitation on how extensive and frequent pooled, standard PCR testing can be.
Self-administered testing. It is possible to do testing that can be self-administered. Shallow swab testing followed by PCR is one option. Instead of the very deep testing (3-4 inches) that is standard for PCR in most medical settings, a self-administered shallow swabbing of about an inch can provide a reliable result. (Jan and I experienced this using a CVS drive-through self-administered test.)
Testing of self-collected saliva is also an option. This is done at the University of Illinois. Also, saliva samples can be pooled in the same way swabbed samples can, and the University of Florida is using this approach. The University of Maine is using swabs on its main campus and saliva on its smaller satellite campuses, where testing appointments are harder to arrange because the students often commute.
Self-administered antigen tests can also be used. The University of Arizona uses them. These small kits use a test strip, similar to how a home pregnancy test works, and can easily be done at home. The results are immediately available. Antigen tests are less sensitive than PCR, but far cheaper. And although the lower sensitivity means that a few early cases could be missed, widespread and frequent testing would still detect an incipient outbreak.
Unfortunately, Abbott Labs, one of the main purveyors of antigen tests, shut down its manufacturing in June on the mistaken assumption that widespread vaccination would eliminate the need for testing. It may be a while before antigen tests catch back up with the demand.
Sewage testing. A very effective approach to early detection of an outbreak is sewage testing. Days before symptoms appear, the virus is detectable in sewage from an infected person. Kendal, because it has its own wastewater treatment facility, has experimented with this approach (more on that in a moment). Sewage testing can determine when an area is Covid-free. It can be used to localize an outbreak to a given building or group of buildings, but then personal testing is needed to determine the source. Sewage testing is used at the Universities of Arizona, Florida, Maine, and New Hampshire.
The University of California at San Diego has set up a highly automated system for sewage monitoring. It has installed automated sampling machines at 126 sites along its sewer system. Results of testing are posted daily on a public, on-line dashboard. The University reports that “wastewater surveillance can detect a case up to 3 to 4 days before individual testing can.”
Of course, sewage testing doesn’t require a network of automated samplers like the one at UCSD. If sampling at the sewage plant turns up a positive test, manual sampling at upstream locations can help locate the source.
Sewage testing is a great surveillance tool, but (unlike the methods listed earlier) it does require the availability of other testing methods to locate the infected individuals.
Kendal’s experiment with sewage testing. At Kendal, sewage testing turned out to be successful at detecting Covid: after a period with no cases on either campus, it successfully detected a case at Crosslands. But at the time we were using it, it was too costly and slow to be beneficial. The two testing facilities we used were in Massachusetts and Florida (which entailed overnighting refrigerated samples) and the testing sites were struggling to keep up with growing demand (which meant it could take close to a week to get results). The cost was high, so we sent out samples only once a week. Based on the high cost and the lag time for results, it was decided to end the experiment.
What lessons might we take from colleges and universities? Currently, we depend primarily on symptoms to detect an outbreak, followed by contact-tracing and, if needed, comprehensive screening of staff and our most vulnerable population. That is keeping us safe, but it is cumbersome to administer and it can be burdensome for staff and for quarantined residents. Using PCR tests for screening of dozens of people is also costly. Some of the options mentioned above might be good alternatives.
There are many options that might be considered. One aspect of university testing that must be kept in mind, though, is that in most cases they have the ability to do tests right on campus. That reduces the turnaround time and potentially reduces the cost as well. We don’t have those advantages. With that caveat, here are some ideas.
- Self-administered tests might be a good option for staff and for independent-living residents.
- Shallow swab tests or saliva tests could be self-administered and could potentially benefit from pooled testing.
- Rapid antigen tests (either self-administered by residents or done at the reception desk for staff and visitors) would detect serious infections immediately, rather than having to wait a day or more for PCR results.
- Sewage testing could be worth another look. The costs have come down, turnaround times have declined, and there are far more places doing the testing, including some that are within reasonable driving distance.
Perhaps one or another of these approaches could ease the burden of the coming months.