What are travel-related control measures?
To contain the spread of COVID-19, numerous countries have implemented control measures related to international travel. These include:
· complete closure of borders (i.e. a total ban on any border crossings);
· partial travel restrictions (e.g. restrictions on air travel only, or restrictions on travellers from certain countries);
· entry or exit screening (e.g. when travellers are asked about symptoms, examined physically, or tested for infection when leaving or entering a country);
· quarantine of travellers (e.g. when travellers have to stay at home or at a specific place for some time after crossing a border).
Some countries implemented similar travel-related control measures during the recent outbreaks of two related diseases, severe acute respiratory syndrome (SARS) and Middle East respiratory syndrome (MERS).
What did we want to find out?
We wanted to find out how effective travel-related control measures are in containing the COVID-19 pandemic. We also wanted to know about the costs of the measures and what effect they have on healthcare and other resource use, as well as potential negative effects, such as feeling isolated.
What we did
We searched for studies on the effects of travel-related control measures on the spread of COVID-19, as well as on SARS and MERS to provide supporting information. Studies had to report how many cases (people with infection) the measures prevented or detected, and whether the measures changed the course of the epidemic. Studies could include people of any age, anywhere. They could be of any design including studies that used ‘real-life’ data (observational studies) or hypothetical data from computer-generated simulations (modelling studies).
We included studies published up to 26 June 2020.
What we found
We found 25 studies on COVID-19, 10 on SARS and one on both SARS and MERS. Studies took place across the world except for Africa and the eastern Mediterranean.
Twelve studies (11 modelling studies, 1 observational study) on COVID-19 found that restricting cross-border travel at the beginning of an outbreak may reduce new cases by a minimum of 26% to a maximum of 90%, may reduce the number of deaths, may reduce the time to an outbreak by between 2 to 26 days, and may reduce the spread and risk of an outbreak. There was also a reduction in imported or exported cases and in growth of the epidemic.
We found 12 studies (6 modelling studies, 6 observational studies) on entry or exit screening, with and without quarantine, to contain the spread of COVID-19. Based on data from three modelling studies, there may be a delay in the time to an outbreak, and between 10% to 53% of infected travellers would be detected. However, the results from the observational studies varied considerably, and we are uncertain about the proportion of people identified accurately as having COVID-19 from these studies.
Only one modelling study examined quarantine measures for COVID-19. It found fewer new cases due to imported cases where 14-day quarantine was in place.
How reliable are these results?
Our confidence in these results is limited for several reasons. Most studies were not based on real-life epidemics but on mathematical predictions. Their results depended on the assumptions that they made, not on real-life data. Also, the studies were very different from each other and their results would probably vary according to the stage of the epidemic, the amount of cross-border travel, other measures undertaken locally, and the extent of implementation and enforcement. Results of entry and exit screening studies might vary according to the screening method used and the level of infection among travellers. Also, some studies were published as ‘preprints’, which means they did not undergo the rigorous checks of most peer-reviewed studies.
What this means
Overall, travel-related control measures may help to limit the spread of disease across national borders. Cross-border travel restrictions are probably more effective than entry and exit screening. Screening is likely to be more effective if combined with other measures, such as quarantine and observation. We found very little information on travel-related quarantine as a stand-alone measure and no information on costs or negative effects.
With much of the evidence deriving from modelling studies, notably for travel restrictions reducing cross-border travel and quarantine of travellers, there is a lack of 'real-life' evidence for many of these measures. The certainty of the evidence for most travel-related control measures is very low and the true effects may be substantially different from those reported here. Nevertheless, some travel-related control measures during the COVID-19 pandemic may have a positive impact on infectious disease outcomes. Broadly, travel restrictions may limit the spread of disease across national borders. Entry and exit symptom screening measures on their own are not likely to be effective in detecting a meaningful proportion of cases to prevent seeding new cases within the protected region; combined with subsequent quarantine, observation and PCR testing, the effectiveness is likely to improve. There was insufficient evidence to draw firm conclusions about the effectiveness of travel-related quarantine on its own. Some of the included studies suggest that effects are likely to depend on factors such as the stage of the epidemic, the interconnectedness of countries, local measures undertaken to contain community transmission, and the extent of implementation and adherence.
In late 2019, first cases of coronavirus disease 2019, or COVID-19, caused by the novel coronavirus SARS-CoV-2, were reported in Wuhan, China. Subsequently COVID-19 spread rapidly around the world. To contain the ensuing pandemic, numerous countries have implemented control measures related to international travel, including border closures, partial travel restrictions, entry or exit screening, and quarantine of travellers.
To assess the effectiveness of travel-related control measures during the COVID-19 pandemic on infectious disease and screening-related outcomes.
We searched MEDLINE, Embase and COVID-19-specific databases, including the WHO Global Database on COVID-19 Research, the Cochrane COVID-19 Study Register, and the CDC COVID-19 Research Database on 26 June 2020. We also conducted backward-citation searches with existing reviews.
We considered experimental, quasi-experimental, observational and modelling studies assessing the effects of travel-related control measures affecting human travel across national borders during the COVID-19 pandemic. We also included studies concerned with severe acute respiratory syndrome (SARS) and Middle East respiratory syndrome (MERS) as indirect evidence. Primary outcomes were cases avoided, cases detected and a shift in epidemic development due to the measures. Secondary outcomes were other infectious disease transmission outcomes, healthcare utilisation, resource requirements and adverse effects if identified in studies assessing at least one primary outcome.
One review author screened titles and abstracts; all excluded abstracts were screened in duplicate. Two review authors independently screened full texts. One review author extracted data, assessed risk of bias and appraised study quality. At least one additional review author checked for correctness of all data reported in the 'Risk of bias' assessment, quality appraisal and data synthesis. For assessing the risk of bias and quality of included studies, we used the Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) tool for observational studies concerned with screening, ROBINS-I for observational ecological studies and a bespoke tool for modelling studies. We synthesised findings narratively. One review author assessed certainty of evidence with GRADE, and the review author team discussed ratings.
We included 40 records reporting on 36 unique studies. We found 17 modelling studies, 7 observational screening studies and one observational ecological study on COVID-19, four modelling and six observational studies on SARS, and one modelling study on SARS and MERS, covering a variety of settings and epidemic stages.
Most studies compared travel-related control measures against a counterfactual scenario in which the intervention measure was not implemented. However, some modelling studies described additional comparator scenarios, such as different levels of travel restrictions, or a combination of measures.
There were concerns with the quality of many modelling studies and the risk of bias of observational studies. Many modelling studies used potentially inappropriate assumptions about the structure and input parameters of models, and failed to adequately assess uncertainty. Concerns with observational screening studies commonly related to the reference test and the flow of the screening process.
Studies on COVID-19
Travel restrictions reducing cross-border travel
Eleven studies employed models to simulate a reduction in travel volume; one observational ecological study assessed travel restrictions in response to the COVID-19 pandemic. Very low-certainty evidence from modelling studies suggests that when implemented at the beginning of the outbreak, cross-border travel restrictions may lead to a reduction in the number of new cases of between 26% to 90% (4 studies), the number of deaths (1 study), the time to outbreak of between 2 and 26 days (2 studies), the risk of outbreak of between 1% to 37% (2 studies), and the effective reproduction number (1 modelling and 1 observational ecological study). Low-certainty evidence from modelling studies suggests a reduction in the number of imported or exported cases of between 70% to 81% (5 studies), and in the growth acceleration of epidemic progression (1 study).
Screening at borders with or without quarantine
Evidence from three modelling studies of entry and exit symptom screening without quarantine suggests delays in the time to outbreak of between 1 to 183 days (very low-certainty evidence) and a detection rate of infected travellers of between 10% to 53% (low-certainty evidence).
Six observational studies of entry and exit screening were conducted in specific settings such as evacuation flights and cruise ship outbreaks. Screening approaches varied but followed a similar structure, involving symptom screening of all individuals at departure or upon arrival, followed by quarantine, and different procedures for observation and PCR testing over a period of at least 14 days. The proportion of cases detected ranged from 0% to 91% (depending on the screening approach), and the positive predictive value ranged from 0% to 100% (very low-certainty evidence). The outcomes, however, should be interpreted in relation to both the screening approach used and the prevalence of infection among the travellers screened; for example, symptom-based screening alone generally performed worse than a combination of symptom-based and PCR screening with subsequent observation during quarantine.
Quarantine of travellers
Evidence from one modelling study simulating a 14-day quarantine suggests a reduction in the number of cases seeded by imported cases; larger reductions were seen with increasing levels of quarantine compliance ranging from 277 to 19 cases with rates of compliance modelled between 70% to 100% (very low-certainty evidence).