Routine testing strategies for airline travel during the COVID-19 pandemic: a simulation analysis

Introduction

Reproducible code for our The Lancet Infectious Diseases paper, Routine testing strategies for airline travel during the COVID-19 pandemic: a simulation study, in which we use microsimulation to estimate the effectiveness of different testing strategies on reducing SARS-CoV-2 transmission in a cohort of hypothetical airline travels.

The full citation is:

Kiang MV, Chin ET, Huynh BQ, Chapman LAC, Rodríguez-Barraquer I, Greenhouse B, Rutherford GW, Bibbins-Domingo K, Havlir D, Basu S, and Lo NC. Routine asymptomatic testing strategies for airline travel during the COVID-19 pandemic: a simulation study. The Lancet Infectious Diseases (March 2021). doi: 10.1016/S1473-3099(21)00134-1

Abstract

Background Routine viral testing strategies for SARS-CoV-2 infection might facilitate safe airline travel during the COVID-19 pandemic and mitigate global spread of the virus. However, the effectiveness of these test-and-travel strategies to reduce passenger risk of SARS-CoV-2 infection and population-level transmission remains unknown.

Methods In this simulation study, we developed a microsimulation of SARS-CoV-2 transmission in a cohort of 100 000 US domestic airline travellers using publicly available data on COVID-19 clinical cases and published natural history parameters to assign individuals one of five health states of susceptible to infection, latent period, early infection, late infection, or recovered. We estimated a per-day risk of infection with SARS-CoV-2 corresponding to a daily incidence of 150 infections per 100000 people. We assessed five testing strategies: (1) anterior nasal PCR test within 3 days of departure, (2) PCR within 3 days of departure and 5 days after arrival, (3) rapid antigen test on the day of travel (assuming 90% of the sensitivity of PCR during active infection), (4) rapid antigen test on the day of travel and PCR test 5 days after arrival, and (5) PCR test 5 days after arrival. Strategies 2 and 4 included a 5-day quarantine after arrival. The travel period was defined as 3 days before travel to 2 weeks after travel. Under each scenario, individuals who tested positive before travel were not permitted to travel. The primary study outcome was cumulative number of infectious days in the cohort over the travel period without isolation or quarantine (population-level transmission risk), and the key secondary outcome was the number of infectious people detected on the day of travel (passenger risk of infection).

Findings We estimated that in a cohort of 100 000 airline travellers, in a scenario with no testing or screening, there would be 8357 (95% uncertainty interval 6144–12831) infectious days with 649 (505–950) actively infectious passengers on the day of travel. The pre-travel PCR test reduced the number of infectious days from 8357 to 5401 (3917–8677), a reduction of 36% (29–41) compared with the base case, and identified 569 (88% [76–92]) of 649 actively infectious travellers on the day of flight; the addition of post-travel quarantine and PCR reduced the number of infectious days to 1474 (1087–2342), a reduction of 82% (80–84) compared with the base case. The rapid antigen test on the day of travel reduced the number of infectious days to 5674 (4126–9081), a reduction of 32% (26–38) compared with the base case, and identified 560 (86% [83–89]) actively infectious travellers; the addition of post-travel quarantine and PCR reduced the number of infectious days to 2518 (1935–3821), a reduction of 70% (67–72) compared with the base case. The post- travel PCR alone reduced the number of infectious days to 4851 (3714–7679), a reduction of 42% (35–49) compared with the base case.

Interpretation Routine asymptomatic testing for SARS-CoV-2 before travel can be an effective strategy to reduce passenger risk of infection during travel, although abbreviated quarantine with post-travel testing is probably needed to reduce population-level transmission due to importation of infection when travelling from a high to low incidence setting.

Issues

Please report issues via email or the issues page.

Changes reflecting comments from others (and our reply)

We thank Dr. Mohammad Shahid and Dr. Lee Altenberg for their letters [1, 2] and interest in our paper.

As outlined in our reply, Dr. Lee Altenberg suggested additional analyses looking at (a) only post-travel infectious days and (b) the number of infections per day. The results from these analyses are now presented in the updated Supplemental Materials as Tables S2 and S3, respectively. In addition, Dr. Altenberg suggested a correction to the accounting of quarantine days in our paper. We fully agree and have corrected the numbers accordingly in the main manuscript. This change affects the scenarios with quarantine, and improves their effectiveness — highlighting the importance of both testing and quarantine of airline passengers. Both the correction and the corrected manuscript are available online. The code and replication data have been been updated to reflect these changes.

Structure

This project is structured as follows:

• code: Contains all code used for this project. Designed to be run sequentially. A brief overview of the code files is provided below.
• data: Contains all summarized data necessary to reproduce our tables and plots.
• data_raw: Contains processed data from intermediate files but before summarizing. (Not on Github.)
• intermediate_files: Contains temporary files not necessary to reproduce our tables and plots. You can download our original intermediate files on OSF and unzip it in the project root.
• output: Contains numerical representations (in csv form) of all figures.
• plots: Contains all figures used the manuscript in jpg and pdf.
• rmds: Contains useful rmarkdown files such as reproducing our tables and providing session information.

Note, we use the config.yml file to modify high-level project-wide parameters.

Reproducibility

All information necessary for full reproducibility, including package version numbers, is available in ./rmds/02_session_info.html. This project uses the renv package for package version control. To use this, open the project in RStudio and run renv::restore().

config.yml

Simulation parameters such as the length of quarantine, number of burn-in days, number of passengers, proportion of subclinical infections, daily probability of infection, etc. can be changed in the config.yml file. In addition, the config.yml file lets you specify different sensitivity analysis configurations and allows you to vary one (or more) parameters across the default parameter sweep. See the config.yml file for details.

./code

The analytic pipeline in the ./code/ folder is designed to have each file run sequentially. That is, each file performs a discrete task and some tasks may be dependent on the output of previous tasks.

• utils.R: Contains almost all helper functions and functions necessary to run the simulations or create the plots. Most functions within utils.R use the roxygen2 documentation template. See this documentation for expected input, output, and notes about each function. Minor or self-explanatory functions may not include roxygen2 docs.
• 01_create_simulated_populations.R: Creates and performs the burn-in for all simulated passenger cohorts. In addition, each simulation state is saved with the random seed used to create the simulation as well as the seed necessary to resume the simulation. Each testing scenario calls one of the simulation states created here (using unpack_simulation_state()).
• 02_run_main_simulations.R: Run all testing scenarios used for the main results. Each simulation result is saved in ./intermediate_files to be collected later using 04_summarize_infection_quantities.R and 05_summarize_testing_quantites.R and then saved in ./data.
• 03_run_sensitivity_analyses.R: Run all testing scenarios used for the main results. Each simulation result is saved in intermediate_files to be collected later using 04_summarize_infection_quantities.R and 05_summarize_testing_quantites.R and then saved in ./data.
• 04_gather_and_munge_intermediate_files.R: Processes all the raw simulations in intermediate_files and saves them in data_raw.
• 05_summarize_infection_quantities.R: Processes the raw files created in 04, summarized them, and then saves the results in ./data. This file saves the time-varying quantities of interest and our primary end points.
• 06_summarize_testing_quantites.R: Processes the raw files created in 04, summarized them, and then saves the results in ./data. This file saves the time-invariant quantities (e.g., testing results).

Other files can be run in any order after 05 and 06 have successfully run. These files create plots used in the main manuscript or associated appendix.

./code/testing_scenarios

Testing scenarios are written as functions that take in all necessary parameters and return no output, instead they save their intermediate results in a temporary location. The structure each function is the same:

1. Load a simulated population using unpack_simulation_state()
2. Increment pre-flight time steps
3. Increment the day-of-flight time step
4. Increment post-flight time steps
5. Repeat these steps n_reps number of times
6. Save all n_reps realizations into a temporary file in ./intermediate_files

The scenarios themselves change in what happens within each stage above. For example, pre-flight testing strategies would occur in Step 2. The day-of-flight risk multiplier only impacts Step 3. Post-flight testing strategies occur in Step 4. See the documentation within each scenario file in ./code/testing_scenarios for details.

Authors (alphabetical):

• Sanjay Basu (: sanjaybasu | : @sanjaybmdphd)
• Kirsten Bibbins-Domingo (: @KBibbinsDomingo)
• Lloyd Chapman (: @lloyd_chapman_)
• Elizabeth Chin (: etchin | : @elizabethtchin)
• Bryan Greenhouse (: @bgreenhouse1)
• Diane Havlir
• Benjamin Huynh (: @benqhuynh)
• Mathew Kiang (: mkiang | : @mathewkiang)
• Nathan Lo (: NathanLo3 | : @NathanLo3579)
• Isabel Rodríguez-Barraquer (: isabelrodbar | : @isabelrodbar)
• George Rutherford (: @rutherford_ucsf)

Attribution

This simulation code is heavily based on the paper, Frequency of Routine Testing for Coronavirus Disease 2019 (COVID-19) in High-risk Healthcare Environments to Reduce Outbreaks, by Chin et al. and the accompanying Github repository.

Pre-print version

An earlier, pre-print version of our paper is available on medRxiv. Several large changes were made in response to reviewers; however, the state of the code when the pre-print was submitted is available in the pre-print release.