Back

Human mobility impacts the transmission of common respiratory viruses: A modeling study of the Seattle metropolitan area

Perofsky, A. C.; Hansen, C.; Burstein, R.; Boyle, S.; Prentice, R.; Marshall, C.; Reinhart, D.; Capodanno, B.; Truong, M.; Schwabe-Fry, K.; Kuchta, K.; Pfau, B.; Acker, Z.; Lee, J.; Sibley, T. R.; McDermot, E.; Rodriguez-Salas, L.; Stone, J.; Gamboa, L.; Han, P. D.; Adler, A.; Waghmare, A.; Jackson, M. L.; Famulare, M.; Shendure, J.; Bedford, T.; Chu, H. Y.; Englund, J. A.; Starita, L. M.; Viboud, C.

2023-11-01 epidemiology
10.1101/2023.10.31.23297868 medRxiv
Show abstract

Many studies have used mobile device location data to model SARS-CoV-2 dynamics, yet relationships between mobility behavior and endemic respiratory pathogens are less understood. We studied the impacts of human mobility on the transmission of SARS-CoV-2 and 16 endemic viruses in Seattle over a 4-year period, 2018-2022. Before 2020, school-related foot traffic and large-scale population movements preceded seasonal outbreaks of endemic viruses. Pathogen circulation dropped substantially after the initiation of stay-at-home orders in March 2020. During this period, mobility was a positive, leading indicator of transmission of all endemic viruses and lagged SARS-CoV-2 activity. Mobility was briefly predictive of SARS-CoV-2 transmission when restrictions relaxed in summer 2020 but associations weakened in subsequent waves. The rebound of endemic viruses was heterogeneously timed but exhibited stronger relationships with mobility than SARS-CoV-2. Mobility is most predictive of respiratory virus transmission during periods of dramatic behavioral change, and, to a lesser extent, at the beginning of epidemic waves. Teaser: Human mobility patterns predict the transmission dynamics of common respiratory viruses in pre- and post-pandemic years.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.