The effect of inter-city travel restrictions on geographical spread of COVID-19: Evidence from Wuhan, China

Billy J Quilty; Charlie Diamond; Yang Liu; Hamish Gibbs; Timothy W Russell; Christopher I Jarvis; Kiesha Prem; Carl A B Pearson; Samuel J Clifford; Stefan Flasche; Petra Klepac; Rosalind M Eggo; Mark Jit

We calculated the total number of daily travellers leaving Wuhan and entering each city by taking the product of the scaling factor, the total daily outflow index from Wuhan, and the daily proportion of travellers from Wuhan entering the four cities. From the daily number of arrivals, we then simulated the daily number of infected travellers by drawing from a Poisson process based on COVID-19 prevalence among Wuhan residents (Supplementary Appendix 2). Daily estimated COVID-19 prevalence in Wuhan was retrieved from outputs of a published model on the early dynamics of COVID-19 transmission in Wuhan (8) . We assumed that individuals would travel regardless of their infection status, and Wuhan was the sole source of infected individuals. We examined four travel scenarios (Table 1) , Scenario 1 is based on the observed travel pattern in 2020, and represents the Chunyun period with cordon sanitaire introduced on 23 January. Scenario 2 represents a counterfactual travel pattern used to evaluate how the COVID-19 outbreak would spread if no cordon sanitaire was implemented. This was based upon the actual travel from Wuhan for the equivalent Chunyun period in 2019. In Scenario 3, we synthesized a hypothetical travel pattern to represent a typical non-Chunyun period with cordon sanitaire introduced on 23 January, using outward travel flow on representative non-Chunyun days in 2019. Scenario 4 is a variation on Scenario 3 in which no cordon sanitaire was implemented.