As the coronavirus disease 2019 (COVID-19) pandemic began to spread rapidly across the world in 2020, among the first measures taken by panic-stricken governments was to seal their borders to travelers returning or arriving from outside their country. Even today, restrictions on border crossings remain a big part of COVID-19 management policies in many parts of the world.
A recent study, released on the medRxiv* preprint server, explored the use of rapid antigen testing for COVID-19 at borders to reduce community transmission, and concluded that it would be effective only at lower volumes of passenger traffic.
Travel bans across borders
Border closures, mandatory quarantine following arrival, and travel bans have all been part of pandemic control interventions in many countries from the beginning. The reason is that once the virus begins to spread from person to person during ordinary interactions outside a healthcare-related setup, community transmission is said to have begun.
This is unlikely to be controlled by border restrictions and will drive the outbreak to high levels without stringent controls on mobility within the community. Countries like Australia and New Zealand have successfully contained the spread of the virus within the community, which has resulted in a reduction in the effective reproductive number (Rt) of the virus.
This has led to an emphasis on preventing the importation of new cases, especially as novel variants are emerging all around the world. These are often not only more infectious but may escape neutralization by antibodies elicited by natural infection and/or vaccines.
Testing for the virus
Testing is the cornerstone of all containment programs. Testing today depends on nasopharyngeal or oropharyngeal swabs, which undergo reverse transcriptase-polymerase chain reaction (RT PCR) testing to detect the presence of viral ribonucleic acid (RNA) or genetic material.
Testing should be accessible and widespread in order to be effective in containing viral spread. Currently, a negative RT PCR certificate obtained from a sample taken within 72 hours of travel is mandatory for entry into several countries.
The issue with this requirement is that many people in low- and medium-income countries (LMIC) cannot get such tests, either due to lack of testing kits, delays in reporting, or the need to cross the border frequently.
Rapid antigen diagnostic tests (Ag-RDTs) are a more affordable option to RT PCR, can be performed using less sophisticated tools, outside laboratory settings, and without laboratory staff, as long as proper training is provided. Moreover, they provide results in minutes. All these advantages have made it a go-to for speedy, widespread point-of-care testing.
Especially in Africa, global consortiums have provided World Health Organization (WHO)-approved Ag-RDTs for high-quality testing. Again, the Access to COVID-19 Tools (ACT) Accelerator reported in September 2020 that they would ensure 120 million Ag-RDTs reached LMICs.
The question remains, is their use appropriate at border crossings? The current study aimed to evolve an elegant algorithm that would estimate the fraction of travelers who must be tested with Ag-RDT in order to prevent imported cases from pushing up the number of community transmissions beyond 1% of the baseline value.
The algorithm includes the number of arrivals from outside the country per 100,000 population of the host country; the prevalence among the incoming travelers; the sensitivity of the test; and a 12% false-negative RT PCR rate.
The algorithm compares the number of undetected cases entering at any point with the allowable number to keep the Rt constant within 1% of baseline. It then predicts the least number of screenings that should be carried out on incoming travelers.
What were the findings?
The results showed that the minimum screened number will vary with the number of border crossings per 100,000 population of the recipient country, and the relative proportion of travelers arriving by land and sea, as well as the other two parameters of the algorithm.
With a low Rt in the destination country, more travelers will have to be screened. However, with very high numbers of crossings, this will fail to check the importation of cases. This is also the case with an Rt over 1, or with a relatively insensitive Ag-RDT.
In the case that the Rt is above 1, rapid antigen tests are irrelevant if the number of travelers is low, or if the prevalence is low. If the test sensitivity is high, these tests would be more useful over a wider range of parameters.
The test would also prove more efficient at screening for infected travelers if used only at land borders, or wherever a negative RT PCR is not always obtained before crossing into another country. This will not, obviously, hold good in a country like Australia, where most incoming travelers do so by air.
South Africa case study
The algorithm was applied to South Africa, where a 90% drop occurred in arrivals. Over 70% of travelers came by land. At the end of 2020, arrivals were at about 16 per 100,000 population. With an Rt below and above 1, the Ag-RDT would screen efficiently for low and high COVID-19 prevalence, respectively.
In Germany, there was an 85% drop in incoming travelers, at 40 per 100,000 population in the second half of the year. About 60% were by land, the rest by air. The test would be adequate for screening only at low prevalence and high or moderate Rt levels, if applied to land arrivals only, or even if both routes were included.
Only 1% of the previous year’s incoming traffic was seen in 2020, at 3 per 100,000 population, almost completely by air. The test would only suffice for screening at low prevalence, with the Rt below 1, because at all Rt levels, not even one infection could be allowed to be imported to keep community transmission within 1% of baseline.
For this reason, the country demands mandatory quarantine on arrival for all travelers.
Generally, in situations where there is no longer community transmission, Ag-RDT screening at borders must be combined with quarantine measures to prevent re-seeding of the epidemic.”
What are the implications?
The findings imply that in order to use Ag-RDT effectively as a border screening strategy, various parameters must be taken into account.
With a low Rt in the destination country, more arrivals, and a higher prevalence in the host countries, the fraction of tests required increases. With a large number of incoming travelers from countries with high COVID-19 rates, this test will be inadequate unless coupled with mandatory quarantine and further testing.
With a high Rt, and low numbers of travelers and cases in the host country, Ag-RDT testing will become unnecessary at the border. In such scenarios, this money could be better channeled elsewhere to prevent community spread.
Additional NPIs will be needed at borders where 100% screening by Ag-RDT will not prevent case importations enough to increase the cases beyond 1% from baseline over a month. It is also suggested that testing be done at land borders first, before air travelers are also included, in situations where the test is found to be sufficient to contain imported infections.
The algorithm presented here can be used as it is by any country, irrespective of its resources, to determine the best border policy and use of resources.
These studies suggest testing alone is unlikely to completely eliminate the risk of international arrivals seeding new outbreaks. The role of Ag-RDTs in such strategies ultimately depends on the goal of each country.”
medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.