The risk from reopening international travel during a Disease X pandemic: a case study of SARS-CoV-2

During the COVID-19 pandemic, caused by SARS-CoV-2, many mathematical models and risk assessments have been created to inform policymakers on effective strategies for controlling the spread of the pathogen. These models were often developed rapidly for timely input to strategic decisions. It is prudent to evaluate the models created and learn from experience so that we can be adequately prepared for a future pandemic. One area of modeling developed during the COVID-19 pandemic was for international travel and how to safely reopen borders and to which countries. We developed an importation risk model for estimating the risk of SARS-CoV-2 infectious travelers entering any airport and parameterized it for UK airports. We ran the model using prevalence estimates from August 2020 and found that 895 (CI: 834–958) infectious travelers would arrive in a single week from the 25 countries considered. We simulated health measures on arrival to assess the efficacy of self-isolation, the policy at the time, in comparison to proposed alternatives. We found that the 14-day self-isolation is 78.0% effective (CI: 74.4–81.6), whereas a test at the airport plus an additional test 4 days later is 68.9% (CI: 64.9–73.0) effective, after accounting for 20% noncompliance. Rapidly implementing control measures for travelers from riskier countries is vital to protect public health. This methodology can be quickly updated to assess the impact of any further changes to international travel policy or disease occurrence. We assess whether our model results would be applicable for a future Disease X, the adaptability of our model, future work to ensure that the model is comprehensive, and the lessons learned from modeling during COVID-19. In particular, we highlight the importance of building flexible, transparent, and adaptable models due to the speed at which policy or the epidemiological situation can change and for use in any future pandemic.