Effectiveness of Travel Restrictions on the Spread of the Novel 2019 Coronavirus (SARS-CoV-2)
Literature Watch Review
Chinazzi M, Davis JT, Ajelli M et al. The effect of travel restrictions on the spread of the 2019 novel coronavirus (COVID-19) outbreak. Science. 2020; doi: 10.1126/science.aba9757.
- The article uses a global meta-population disease transmission model to project the impact of travel limitations on the national and international spread of the 2019 novel coronavirus (SARS-CoV-2; the virus causing COVID-19).
- The model is calibrated based on internationally reported cases and shows that at the start of the travel ban from Wuhan on January 23, 2020, most Chinese cities had already received many infected travelers.
- The travel quarantine of Wuhan delayed the overall epidemic progression by only 3 to 5 days in mainland China but had a more marked effect at the international scale, where case importations were reduced by nearly 80% until mid-February.
- Modeling results indicate that sustained 90% travel restrictions to and from mainland China only modestly affect the epidemic trajectory unless combined with a 50% or higher reduction of transmission in the community.
As of March 3, 2020, a total of 80,151 cases of COVID-19 had been detected and confirmed in mainland China, with an additional 10,566 cases in 72 other countries. This article models the domestic and international spread of the COVID-19 epidemic, the effects of the travel ban implemented in Wuhan, and the international travel restrictions adopted by several countries in early February 2020.
The GLEAM Model
The international spread of the epidemic was modeled using the Global Epidemic and Mobility Model (GLEAM) in which the world is divided into over 3,200 subpopulations gathered around transportation hubs in approximately 200 different countries and territories. Airline transportation data consider daily origin-destination traffic flows, whereas ground mobility flows are derived from the statistics offices for 30 countries on 5 continents and Baidu Location Services for mainland China. Within each subpopulation, human-to-human transmission of SARS-CoV-2 is modeled using a compartmental representation of the disease in which individuals can occupy 1 of 4 sites/states: susceptible, latent, infectious, and removed. The sum of the mean latent and infectious periods defines the generation time. The model generates a range of possible epidemic scenarios. It assumes an epidemic start date between November 15 and December 1, 2019, with 40 cases caused by zoonotic exposure and an overall global detection rate of imported cases as low as 40%. A sensitivity analysis on all parameters is supplied in an appendix. This model analysis reports the results for generation time, Tg = 7.5 days, an average reproductive number R0 = 2.57 (90% CI: 2.37-2.78), and a doubling time measured at Td = 4.2 days (90% CI: 3.8-4.7).
Wuhan Travel Ban
On January 22, 2020, the day before travel restrictions from Wuhan were implemented, the majority of projected cases were in Wuhan, with a median number of 117,584 (90% CI: 62,468-199,581), compared with 7,474 (90% CI: 3,529-16,142), outside Wuhan. Initially, the model showed no change in the epidemic trajectory of Wuhan, although it showed a delay of about 3 days occurring for other locations in mainland China. The overall reduction of cases in mainland China (excluding Wuhan) was close to 10% by January 31, 2020, with a relative reduction of cases across specific locations varying in a range from 1% to 58%. With a doubling time of 4 to 5 days, this level of reduction corresponds to only a modest delay of the epidemic trajectory of 1 to 6 days in mainland China. Model predictions correlated closely with observed case numbers reported by WHO (Pearson's r = 0.74, P < .00001). The median ascertainment rate of detecting an infected individual in mainland China is equal to 24.4% (IQR: 12.7%, 35.8%). In other words, the modeling results suggest that in mainland China, only 1 of 4 cases are detected and confirmed.
Relative Risk of Case Importation
The model also allows the estimation of the number of case importations in international locations from mainland China. It estimated a 77% reduction in cases imported from mainland China to other countries as a result of the Wuhan travel ban in early February 2020. Although the number of cases imported internationally initially showed a marked decrease, it began to increase again in the following weeks, with importation from locations in mainland China. The model indicates that after the travel restrictions in Wuhan were implemented on January 23, the top 5 ranked cities as the origin of international case importations were Shanghai, Beijing, Shenzhen, Guangzhou, and Kunming. Similarly, the model can rank countries across the world according to the relative risk of importing cases from mainland China. Prior to the travel ban, approximately 86% of the internationally imported cases originated from Wuhan. After the travel ban, the top 10 contributors to the relative risk are needed to account for at least 80% of the internationally imported cases, where the top 3 contributors are Shanghai (28.1%), Beijing (14%), and Shenzhen (12.8%). In terms of relative risk of importation, the countries at higher risk of importation after the implementation of the Wuhan travel ban are Japan (11% pretravel ban, 13.9% posttravel ban), Thailand (22.8% pretravel ban, 13% posttravel ban), Republic of Korea/South Korea (7.4% pretravel ban, 11.3% posttravel ban), Taiwan (9.5% pretravel ban, 10% posttravel ban), and the U.S. (4.7% pretravel ban, 5.7% posttravel ban).
International Travel Restrictions and Transmissibility Reduction
A relative reduction of transmissibility could be achieved through early detection and isolation of cases as well as behavioral changes and awareness of the disease in the population. Along with travel reductions, the article considers 3 scenarios concerning disease transmissibility: 1) a status quo situation with the same transmissibility found from the model calibration through January 23, 2020; 2) a moderate relative reduction of the original transmissibility (25%), corresponding to a transmissibility dampening factor of r = 0.75; and 3) a strong reduction (50%) of the original transmissibility (r = 0.50). The simulated scenarios show that even in the case of 90% travel reductions, if transmissibility is not reduced (r = 1), the epidemic peak is delayed for no more than 2 weeks (to the end of April to early May 2020 in mainland China and to the first week of March 2020 in Wuhan). Additionally, the number of cases imported in other countries is initially affected by a 10-fold reduction, but by March 1, 2020, if transmission is not reduced (r = 1), the number of importations is estimated to reach the levels of 170 and 35 cases detected per day for the 40% and 90% travel-restriction scenarios, respectively. In the moderate transmission reduction scenarios, the epidemic peak is delayed until late June 2020. Even larger travel limitations (> 90%) will extend the period during which the importation of cases is greatly reduced. However, by February 1, 2020, in the strong transmissibility scenario, the model estimates 101 (90% CI: 50-173) importation events with ≥ 1 potential case(s) that could be seeding multiple epidemic outbreaks across the world, thus potentially leading to the international expansion of the COVID-19 epidemic, consistent with the emergence of outbreaks in countries across the world (including Italy, South Korea, and Iran in the second half of February 2020). Figure 3 presents a flow chart of the cases projected to have been exported from major Chinese cities to the top 20 countries at greatest risk of importation. Figure 4 is a scatterplot illustrating how 0%, 40%, and 90% reductions in travel from mainland China would be effective in flattening the epidemic peak only when transmissibility is also greatly reduced (r = 0.5).
This article is a penetrating analysis (using GLEAM) of the COVID-19 outbreak, which arose from a zoonotic source in Wuhan, China at the end of November 2019 and spread rapidly within Wuhan and into surrounding mainland China. It was temporarily slowed by an internal travel ban imposed on Wuhan on January 23, 2020. The model predicted that only 1 in 4 early cases were identified and confirmed. Thus, the epidemic escaped containment measures at an early stage. Subsequent international restrictions and attempts at reduction of transmissibility have neither been quick nor rigorous enough to stop the spread or reduce transmission. In their discussion on the limitations of their analysis, the authors point out that many parameters were chosen based on early results of the outbreak. Although the model is stable to variations in these parameters, knowledge of key characteristics of the disease would reduce uncertainties. The model does not account for heterogeneities due to age differences in susceptibility and contact patterns and assumes that travel probabilities are homogeneous across all individuals at each transport hub. The model also assumes long-term enforcement of individual mobility restrictions until June 2020, which may not be feasible or sustainable for such a long period.