Examining SARS-CoV-2 Interventions in Residential Colleges Using an Empirical Network

Abstract

Objectives Universities have turned to SARS-CoV-2 models to examine campus reopening strategies. While these studies have explored a variety of modeling techniques, none have used empirical data. Methods In this study, we use an empirical proximity network of college freshmen obtained using smartphone Bluetooth to simulate the spread of the virus. We investigate the role of immunization, testing, isolation, mask wearing, and social distancing in the presence of implementation challenges and imperfect compliance. Results We show that frequent testing could drastically reduce the spread of the virus if levels of immunity are low, but its effects are limited if immunity is more ubiquitous. Furthermore, moderate levels of mask wearing and social distancing could lead to additional reductions in cumulative incidence, but their benefit decreases rapidly as immunity and testing frequency increase. However, if immunity from vaccination is imperfect or declines over time, scenarios not studied here, frequent testing and other interventions may play more central roles. Conclusions Our findings suggest that although regular testing and isolation are powerful tools, they have limited benefit if immunity is high or other interventions are widely adopted. If universities can attain even moderate levels of vaccination, masking, and social distancing, they may be able to relax the frequency of testing to once every four weeks.

Keywords: Bluetooth; COVID-19; Copenhagen Network Study; Proximity network; Repeat testing; SARS-CoV-2.

Fig. 1

Number of students infected over the course of a simulated 16-week semester for${\mathit{R}}_{\mathbf{0}}\approx \mathbf{3}.\mathbf{0}$and high levels of transmission from the community. Rows show different proportions of the population immunized, as annotated on the right. Columns show scenarios where scheduled testing was conducted every three, seven, 14, and 28 days, respectively, and no scheduled testing. Grey lines denote individual simulations. Blue lines indicate the point-wise average trajectory over all 100 replicates. Vertical red lines and text indicate the average times to reach $10\%$ of the population infected, which were computed by identifying the time required to reach $10\%$ infected for each realization and averaging those times.

Fig. 2

Cumulative percentage infected for various proportions of the population social distancing and/or wearing masks for${\mathit{R}}_{\mathbf{0}}\approx \mathbf{3}.\mathbf{0}$. Cell values indicate the proportion of the population infected by the end of a simulated 16-week semester. Rows show different proportions of the population immunized, as annotated on the right. Columns show scenarios where scheduled testing was conducted every three, seven, 14, and 28 days, respectively, and no scheduled testing.