Analysis parameters

Additional parameters can be customized in the *Sim inputs* tab.

Results

Plot



Table

Customize simulation parameters and then press "re-run simulation" below.


Options

Quarantine durations (days)

Probabilities

Test sensitivity




Tool summary

This tool is intended to estimate the impact of policies to manage the risk of COVID-19 introduced by arriving travelers. The policies compared are mandatory quarantine of varying length with or without testing. The user can customize several simulation parameters related to compliance with quarantine and isolation, test sensitivity, the relative prevalence of asymptomatic vs. infection. Uncertainty in outcomes reflect uncertainty in the distribution of the duration of the pre-symptomatic-infectious and symptomatic-infectious periods for those with symptomatic infections, the duration of the asymptomatic-infectious period for those with asymptomatic infections, and the incubation period for both infection types

Users can toggle between five metrics. The metric "Days at-risk per infected traveler" refers to the average time an infected traveler would be at risk of infecting members of the community because they are infectious and not in quarantine or isolation. The metric "Adjusted days at-risk per infected travelers" multiplies the infectious days in the community for those with asymptomatic infections by the relative transmission risk for asymptomatic infections, which is less than 1 in the base case. This reflects the belief that symptomatic infections have a higher community transmission risk. The primary metric "Percent risk reduced" is calculated by comparing the adjusted days at-risk for each policy to that of a 0-day quarantine without testing policy. While these first 3 metrics require only the parameters from the simulation, two others require additional parameters. The outcome "Person-days at risk per 10,000 travelers" has both infected and uninfected travelers in the denominator. For this outcome, the user must indicate an estimated or assumed prevalence of active infection among travelers. The outcome "Secondary cases" estimates how many community members you would expect to be infected by travelers. Calculating this outcome requires an estimate of the rate of secondary infections per person-day that an traveler is infectious and at-risk in the community.

About the authors

This tool was created by Alton Russell, postdoctoral research fellow at the Mass General Hospital Institute for Technology Assessment and Harvard Medical School and visitor at McGill Clinical & Health Informatics (MCHI), in collaboration with David Buckeridge, Professor of Epidemiology and Biostatistics at McGill University and director of the surveillance lab at MCHI.

For questions, feel free to email altonr <at> stanford <dot> edu.

A huge thanks to Aman Verma and Maxime Lavigne for technical assistance with hosting the app.

Links

McGill Clinical Health and Informatics (MCHI)
Alton's website
Github with code for this project