Importation models for travel-related SARS-CoV-2 cases reported in Newfoundland and Labrador during the COVID-19 pandemic

Abstract

AbstractDuring the COVID-19 pandemic there was substantial variation between countries in the severity of the travel restrictions implemented suggesting a need for better importation models. Data to evaluate the accuracy of importation models is available for the Canadian province of Newfoundland and Labrador (NL; September 2020 to June 2021) as arriving travelers were frequently tested for SARS-CoV-2 and travel-related cases were reported. Travel volume to NL was estimated from flight data, and travel declaration forms completed at entry to Canada, and at entry to NL during the pandemic. We found that during the pandemic travel to NL decreased by 82%, the percentage of travelers arriving from Québec decreased (from 14 to 4%), and from Alberta increased (from 7 to 17%). We derived and validated an epidemiological model predicting the number of travelers testing positive for SARS-CoV-2 after arrival in NL, but found that statistical models with less description of SARS-CoV-2 epidemiology, and with parameters fitted from the validation data more accurately predicted the daily number of travel-related cases reported in NL originating from Canada (R2= 0.55, ΔAICc = 137). Our results highlight the importance of testing travelers and reporting travel-related cases as these data are needed for importation models to support public health decisions.SignificanceImportation models consider epidemiology and inbound travel volumes to predict the arrival rate of infected travelers to a jurisdiction, and are used to guide travel restriction recommendations during outbreaks. A limitation of many importation models are that they are not validated with ‘real world’ data. Using data from the province of Newfoundland and Labrador (NL), Canada, we develop an importation model that considers travel volumes estimated from travel declaration forms completed by arriving travelers, and SARS-CoV-2 epidemiology. We validate the model using the number of travel-related cases reported from September 2020 to May 2021 in NL. Our results highlight the importance of reporting travel-related cases, and the need for detailed travel and epidemiological data to formulate reliable importation models.