Modeling the impact of human mobility and travel restrictions on the potential spread of SARS-CoV-2 in Taiwan

Meng-Chun Chang; Rebecca Kahn; Yu-An Li; Cheng-Sheng Lee; Caroline O Buckee; Hsiao-Han Chang

By utilizing aggregated human movement data from Facebook, we built metapopulation models to understand the potential spread of SARS-CoV-2 in Taiwan given different initial conditions and to assess the potential impact of travel restrictions. We identified the top cities with the highest risk of infection as well as the top cities with the highest importation risk from other cities based on Facebook data and population sizes. We made a web-based interface showing the geographic distribution of infections at different time points (T100, T500 and T1000) in the initial stages of the outbreak given different locations of initial infections. This will help the public avoid high-risk areas, help public health professionals identify surveillance targets, and inform decisions on travel restrictions. Consistent with previous findings showing that international travel bans are less effective than social distancing within the country, 6, 19 we found that intracity travel reduction has a higher impact on disease dynamics than intercity travel reduction and increasing the length of intracity travel reduction increases the impact. Intercity travel reduction, however, influences the variation in infection numbers across cities and can reduce the number of cities that have infections at the initial stage of the outbreak. While intercity travel did not decrease the probability of having high numbers of infections, containing the infections to a few cities has important public health impacts, as this means surveillance system can focus on fewer cities and control efforts can be more targeted.