Comparison of COVID-19 testing strategies and costs for professional sports teams: A case study of J. League clubs

Figures

• 

  Fig 1. Diagram of transitions among the states of individuals. The average duration of E, P1, and P2 was 5 days for the wild type and 3 days for the Omicron variant. The other epidemic parameters were common to the two variants. (See the Model Parameters section below).

• 

  Fig 2. Summary of the simulation and the cost calculation. (A) Simulation starts with one infected individual. (B) Infection spreads. If infected individuals are found, the individuals are quarantined. The revenue loss is counted as a cost. (C) Infected individuals remain. The simulation goes back to (B) and continues. (D) If the number of infected individuals is larger than a threshold number, games are cancelled. The cost of the cancelation is counted. (E) If there are no more infected individuals, the simulation ends. In addition to the specified costs, the cost of the testing is counted. The cost depends on the test frequency and the test type (antigen vs. PCR testing). More frequent testing leads to a high isolation rate of infected individuals but costs more.

• 

  Fig 3. Average values of four measurement items. The results for R0 = 2.5 are shown as circles and those for R0 = 5.0 as rectangles. Gray symbols represent PCR testing, black symbols antigen testing, and open symbols no testing.

• 

  Fig 4. Number of tests from the time an infected individual appears in the population until the cessation of infection. (A) PCR testing, (B) antigen testing. Plot symbols and labels are as in Fig 3. In the antigen-testing series (scenario codes W-A- and O-A), the number of PCR tests is not zero. This is because J.League conducted PCR tests to confirm cases that tested positive in the antigen tests. The cost-effectiveness calculation also incorporates the cost of these PCR tests.

• 

  Fig 5. Results of additional testing. Gray symbols show the results of PCR testing and black symbols the results of antigen testing; circles show the results for R0 = 2.5 and rectangles the results for R0 = 5.0. In each panel, the two results on the extreme right are for regular testing (from Figs 3 and 4, with open symbols showing the results for Omicron with no testing, and black symbols the results for Omicron with twice-weekly antigen testing at 50% relative PCR sensitivity).

• 

  Fig 6. Costs of testing scenarios. (A) regular-testing scenarios with P0 = 10–5, (B) regular-testing scenarios with P0 = 10–3, (C) regular and additional-testing scenarios with R0 = 2.5, and (D) regular and additional-testing scenarios with R0 = 5.0. In A and B, the circles represent the cost with R0 = 2.5, and the boxes represent the cost with R0 = 5.0. The two lines (for the results of R0 = 2.5 and 5.0) overlap in Panel A because the respective values are very similar. With the exception of scenarios O-N-N-N (no regular testing, fine black broken line) and O-A-2w-50 (twice-weekly antigen testing with 50% sensitivity, coarse black broken line), the order of the scenarios regarding costs remains the same, irrespective of P0.

• 

  Fig 7. Estimated effective reproductive number (Re) under various countermeasures. The dots are the results for R0 = 2.5 and the rectangles are the results for R0 = 5.0.

• 

  Fig 8. The final epidemic size.