Optimal use of COVID-19 Ag-RDT screening at border crossings to prevent community transmission: A modeling analysis

PLOS Glob Public Health. 2022 May 16;2(5):e0000086. doi: 10.1371/journal.pgph.0000086. eCollection 2022.

Abstract

Countries around the world have implemented restrictions on mobility, especially cross-border travel to reduce or prevent SARS-CoV-2 community transmission. Rapid antigen testing (Ag-RDT), with on-site administration and rapid turnaround time may provide a valuable screening measure to ease cross-border travel while minimizing risk of local transmission. To maximize impact, we developed an optimal Ag-RDT screening algorithm for cross-border entry. Using a previously developed mathematical model, we determined the daily number of imported COVID-19 cases that would generate no more than a relative 1% increase in cases over one month for different effective reproductive numbers (Rt) and COVID-19 prevalence within the recipient country. We then developed an algorithm-for differing levels of Rt, arrivals per day, mode of travel, and SARS-CoV-2 prevalence amongst travelers-to determine the minimum proportion of people that would need Ag-RDT testing at border crossings to ensure no greater than the relative 1% community spread increase. When daily international arrivals and/or COVID-19 prevalence amongst arrivals increases, the proportion of arrivals required to test using Ag-RDT increases. At very high numbers of international arrivals/COVID-19 prevalence, Ag-RDT testing is not sufficient to prevent increased community spread, especially when recipient country prevalence and Rt are low. In these cases, Ag-RDT screening would need to be supplemented with other measures to prevent an increase in community transmission. An efficient Ag-RDT algorithm for SARS-CoV-2 testing depends strongly on the epidemic status within the recipient country, volume of travel, proportion of land and air arrivals, test sensitivity, and COVID-19 prevalence among travelers.

Grants and funding

This work was supported by the United States Agency for International Development through a grant awarded to KTLS and BEN (72067419CA00004), and the Government of The Netherlands through an award given to JMC, KTLS, SJG, and BEN. The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The authors’ views expressed in this publication do not necessarily reflect the views of USAID, or the Government of The Netherlands.