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

Although we did not observe a significant change in intercity movements from January 30 th to March 8 th , Facebook data have the potential to track how the volume and pattern of travel change through time as the outbreak progresses, and we can incorporate any change in human mobility into the metapopulation models in nearly real time to help fight COVID-19. 11,25 Moreover, our model utilizing human mobility data from Facebook is not limited to intercity or intracity level, or Taiwan. Facebook movement data are also calculated at finer geographic scales (such as towns) and for other countries, and our model can be easily applied in these settings to understand disease dynamics of COVID-19. Figure 1 . The probability of having 1000 infections. P1000,3 from simulations with initial infections in Taipei City (representing big cities) or Changhua County (representing small cities) using both contact and residence models are shown. The difference between big and small cities was more significant in the contact model than in the residence model. Intracity and intercity travel reduction reduced P1000,3, while the impact of intercity travel reduction was minor. Increased human mobility during Lunar New Year increased P1000,3.