COVID watchdawgs: SARS-CoV-2 at UW in the pandemic’s early days

September of 2020 at the University of Washington was very different from the start of most academic years. Due to the SARS-CoV-2 pandemic, stay-at-home orders were still in effect, more than 90% of classes were held online, and students were strongly encouraged to stay home. Far from a bustling campus welcoming ~40,000 students, UW Seattle was a ghost town.

We know that social distancing saved lives before vaccines were available, but it was only a matter of time before students and staff wanted to return to in-person activities. When the campus re-opened in September of 2021, there was excitement—but also concern and confusion about how safe it was to return to in-person learning.

To understand the transmission dynamics of SARS-CoV-2 in a college setting, a study published in Clinical Chemistry reports on insights from the Husky Coronavirus Testing program, a surveillance program specifically designed to track SARS-CoV-2 transmission at UW. This paper is a joint effort between the labs of Drs. Ana Weil and Helen Chu in the  UW Division of Allergy and Infectious Diseases. (You may recognize Dr. Chu’s name from her work with the Seattle Flu Study, which in 2020 quickly pivoted to track SARS-Cov-2).

The Husky Coronavirus Testing program conducted viral surveillance on the UW Seattle and 2 satellite campuses starting in September 2020. It enrolled over 40,000 UW students, faculty, and staff, inviting them to submit daily surveys on symptoms and providing COVID testing. Nasal swabs were collected either in the presence of study personnel, or they could be returned through droboxes. Over 234,000 swabs were collected.

The focus of this publication is how SARS-CoV-2 spread on UW campuses between September 2020 and September 2022, and how the diversity of viruses observed at UW compared to viral diversity in the state of Washington in the same period. “I am generally interested in examining the question of how respiratory viral diversity in certain subpopulations (such as a college campus) compares to diversity in larger geographic areas that include the subpopulations (such as a state),” lead author Dr. Amanda Casto says.

The authors sequenced 3,606 viral genomes from unique COVID events at UW and compared their sequences to publicly accessible SARS-CoV-2 genomes from Washington in an online database. They first wanted to understand generally how long it took to detect a viral lineage or clade at a UW campus after it was reported in Washington State. They found that, typically, a viral clade or lineage was picked up by the Washington surveillance about a month before it was detected by the Husky Coronavirus Study. There are some exceptions—for example, the Omicron subvariant BA.2 was detected January 3, 2022 on the UW main campus, a few weeks before it was reported in Washington State.

To understand transmission dynamics, the authors calculated what they refer to as ‘transmission clusters’. This method uses heterogeneity of viral sequences, number of identical sequences detected, and duration of time that a sequence was detected to understand viral transmission on campus. The number of genetically identical viruses that are observed in sequence databases can provide insight into the rate of viral transmission. A detailed analysis of this was performed by Trevor Bedford and co-authors at Fred Hutch (here's a more in-depth explanation of how it works).

Using these transmission clusters, it was found that most UW strains/lineages were closely related to at least 1 other specimen. They also were able to detect several possible spill-over events, where virus variants detected in King County or Washington State appeared to be descents from variants previously circulating exclusively at UW.

Their model suggests a back-and-forth between UW and King County: in the first year, viruses seemed to spread at a higher rate from King County to UW, while in the second year, there was a higher rate of spread from UW into King County. The authors point out that this conclusion may be confounded by the overall number of sequences collected, as the first year had far fewer data points than the second year when the students returned.

While the overall demographics of people testing positive for COVID was representative of the UW community, students tended to be over-represented in the UW-specific transmission clusters compared to faculty and staff. This makes sense, as “students frequently live in communal housing and attend university-related social events.” says Dr. Casto. Of these, sorority and fraternity members were over-represented in clusters relative to non-members. Students also were more likely to experience a second Sars-CoV-2 infection during the study period compared to non-students.

Overall, this study represents the largest and longest survey of SARS-CoV-2 cases on a college campus. Transmission dynamics appeared to be different between the years of online-only instruction versus the return to in-person learning. “This provides important information for public health agencies who formulate viral surveillance strategies,” Dr. Casto says, and “these findings are helpful to university administrators who have to make decisions around how limited infection control resources will be utilized.”

Fred Hutch/University of Washington/Seattle Children’s Cancer Consortium members Drs. Christina Lockwood, Jay Shendure, and Lea Starita contributed to this research.

The spotlighted research was funded by a Howard Hughes Medical Institute Covid Supplement Award and by the United States Senate and House of Representatives, Bill 748, Coronavirus Aid, Relief, and Economic Security Act.

Casto, AM, Paredes, MI, Bennett, JC, Luiten, KG, Han, PD, Gamboa, LS, McDermot, E, Gottlieb, GS, Acker, Z, Lo, NK, McDonald, D, McCaffrey, KM, Figgins, MD, Lockwood, CM, Shendure, J, Uyeki, TM, Starita, LM, Bedford, T, Chu, HY, & Weil, AA. 2025. SARS-CoV-2 Diversity and Transmission on a University Campus across Two Academic Years during the Pandemic. Clin Chem, 71(1), 192-202.