These are files used in and as a complement to the paper Quarantine and the risk of COVID-19 importation, by Julien Arino, Nicolas Bajeux, Stephanie Portet and James Watmough, published in Epidemiology and Infection 148:e298 (2020), available here. We will progressively be adding files here.
We have set up a small OVH server on which the (R) Shiny apps run, so you can try these components without having to install R. It is located here. However, be mindful of the fact we purposefully limited capacities because of the limited capacity of the server. To run code as in the paper, you will have to download the code and run it yourself. Note that you can run the Shiny apps from R without cloning the Github repo by using a command of the form
shiny::runGitHub("julien-arino/covid-19-importation-risk", username = "some-github-user-name", subdir = "subdir_name")
where subdir_name is one of the subdirectories of Shiny apps: Q_calculator_shiny, single_stimulation_simulations_shiny or CAN_PT_HR_new_cases_shiny.
The (R) Shiny app Q_calculator_shiny is a quarantine efficacy calculator. It is also running online here. You can set disease progression paramaters and the duration of quarantine. It is based on the SLIAR model in the paper and provides two views: the efficacy of quarantine, which we define as the percentage of unobservable cases (L_1, L_2, A_1 and A_2) that are still unobservable at the end of the quarantine period; and the trajectories of individuals that start in one of these unobservable states.
The (R) Shiny app single_stimulation_simulations_shiny runs a number of single stimulation simulations. It is also running online here. It allows you to change most model parameters.
The (R) Shiny app CAN_PT_HR_new_cases_shiny (running online here) allows to produce variations on what is at present the first figure in the paper. When considering imporations or reimportations, it is useful to know how long a jurisdiction has been with/without cases. For now, we only show here jurisdictions in Canada. Code to update the epidemic information is provided (refresh_epi_data_CAN.R, process_epi_data_CAN_confirmed.R and process_epi_data_CAN_deaths.R).