A data-driven assessment of early travel restrictions related to the spreading of the novel COVID-19 within mainland China

Alberto Aleta; Qitong Hu; Jiachen Ye; Peng Ji; Yamir Moreno

Given the unprecedented characteristics of this outbreak, we here adopt a slightly different approach and study a data-driven metapopulation model that makes use of the actual flows of the population to properly measure the early effects of the travel reductions in China. Specifically, we built a basic metapopulation model [19, 20] of 31 regions in China (except HongKong, Macau and Taiwan). Inside each population, we considered that the individuals in the population interact following a homogeneous mixing scheme. This approach is similar to the ones proposed in [7] and [15] . However, in our case, we perform a data-driven simulation with the actual flows of individuals that have taken during the period of study. That is to say, we do not rely on transportation data (which is most of the time given as the maximum flow capacity between subpopulations), but on a real dataset gathered for this occasion (see Materials and Methods). Given that we restrict the analysis to mainland China, this dataset could be better suited for relatively small scale movements. As recently reported, our modeling framework also allows to make an early assessment of the impact that travel restrictions have had on the spreading of the epidemic in mainland China. The results align with previous findings and indicate that travel restrictions do not have a significant impact in containing the expansion of the disease, though reducing the flow of individuals could lead to a delay in the importation of new cases in other subpopulations. Importantly, ours is another study in which taking the mobility patterns into account in as realistic as possible way emerges as a key factor for modeling purposes. In this sense, our conclusions point out that despite the many advances in disease modeling during the last two decades, there still remain many open challenges, most of them related to how to sensibly incorporate human behavioral changes and response to the COVID-19 outbreak.