High resolution proximity statistics as early warning for US universities reopening during COVID-19

Abstract

Reopening of colleges and universities for the Fall semester of 2020 across the United States has caused significant COVID-19 case spikes, requiring reactive responses such as temporary closures and switching to online learning. Until sufficient levels of immunity are reached through vaccination, Institutions of Higher Education will need to balance academic operations with COVID-19 spread risk within and outside the student community. In this work, we study the impact of proximity statistics obtained from high resolution mobility traces in predicting case rate surges in university counties. We focus on 50 land-grant university counties (LGUCs) across the country and show high correlation (PCC > 0.6) between proximity statistics and COVID-19 case rates for several LGUCs during the period around Fall 2020 reopenings. These observations provide a lead time of up to ∼3 weeks in preparing resources and planning containment efforts. We also show how features such as total population, population affiliated with university, median income and case rate intensity could explain some of the observed high correlation. We believe these easily explainable mobility metrics along with other disease surveillance indicators can help universities be better prepared for the Spring 2021 semester.

Informing University Covid-19 Decisions Using Simple Compartmental Models

“…One of these avenues was to use location-based mobility data. Studies of this have shown promising results for the early period (reopening) of a semester (Mehrab et al 2020), but the correlations did not hold for later portions of the semester.…”

“…There has been work completed that delves into the impact of such actions by schools on the health of the student population itself, often focusing on mental health (Conrad et al 2021;Zhai and Du 2020). Additionally, there has been work completed on the impact of large schools on the case prevalence of their surrounding community (Leidner et al 2021;Mehrab et al 2020;Andersen et al 2020), finding that schools with students returning to campus may have contributed to higher observed cases in the broader community. These findings heighten the importance for systems, like the one described in this article, that can inform college administration-level decision making.…”

Empirical networks for localized COVID-19 interventions using WiFi infrastructure at university campuses

“…Mobility generally has been used to dynamically model disease spread of influenza [47], rubella [60] and COVID-19 [3, 45]. For the latter, various studies show the effectiveness of mobility restrictions at a regional–, or city–level [66, 10, 6, 37, 30]. Previous studies that use mobility information to model disease spread and interventions typically rely on cell tower localization or aggregating GPS information from mobile phones.…”

“…Our methodology also promises other advantages. Mobility generally has been used to dynamically model disease spread of influenza [59], rubella [73] and COVID-19 [3,57] showing the effectiveness of mobility restrictions at a regional-, or city-level [79,11,7,45,34]. These studies typically rely on cell tower localization or aggregating GPS information from mobile phones [10].…”

Evaluating the utility of high-resolution proximity metrics in predicting the spread of COVID-19

“…Thus, despite successes in the use of mobility data for epidemic modeling and forecasting, it has been hard to pin down exactly how much mobility contributes to the spread of COVID-19. We have seen several scenarios, especially at universities, where a surge in interactions preceded a surge in cases, but there have been also cases where the correlation between interactions and cases is low [26].…”