WellsRC/Optimizing-COVID19-Quarantine-and-Testing-Strategies: Optimal COVID-19 quarantine and testing strategies
| Main Authors: | Chad R. Wells, Jeffrey P. Townsend, Abhishek Pandey, Seyed M. Moghadas, Gary Krieger, Burton Singer, Robert H. McDonald, Meagan C. Fitzpatrick, Alison P. Galvani |
|---|---|
| Format: | info software Journal |
| Terbitan: |
, 2020
|
| Online Access: |
https://zenodo.org/record/4315462 |
Daftar Isi:
- This repository contains codes and data used to simulate and analyze COVID-19 quarantine and testing strategies in the scenarios of Entry into quarantine uniformly during the incubation period or the course of disease Entry into quarantine through contact tracing Note: The code is set up for homogeneous transmission for symptomatic and asymptomatic individuals (i.e. R0_S=R0_A) for the purpose of the manuscript, and only partially set-up for heterogeneous transmission. See https://github.com/WellsRC/Optimizing-COVID19-Quarantine-and-Testing-Strategies for any relevant updates. The model code is written in MATLAB and results are saved as MATLAB data files (extension .mat), with plots also being constructed in MATLAB. As MATLAB is not an open-source software/programming language, a compatible code that can be run using GNU Octave can be found in the directory named Octave in the repository. OS System requirements The codes developed here are tested on Windows operating system (Windows 10 Home: 64-bit). However as Matlab and Octave are available for most operating systems, codes should run on Mac OSX and Linux as well. Installation guide MATLAB Installation instruction for MATLAB can be found at https://www.mathworks.com/help/install/install-products.html. Typical install time for MATLAB on a "normal" desktop is around 30-40 minutes. The current codes were developed and tested on MATLAB R2018b. GNU Octave When MATLAB is not accessible due to lack of license or any other reason, the open-source GNU Octave can be used to test the code. We tested our code with GNU Octave version 4.2.2. Necessary adjustment to code was done to make it compatible with GNU Octave and it can be found in the directory named Octave in the repository. Installation instruction for GNU octave can be found at https://www.gnu.org/software/octave/. Typical install time for GNU Octave on a "normal" desktop is 15 minutes or less. As most data is saved in MATLAB data file format as well, they can be read directly in GNU Octave for speed. If testing the code that reads the raw data and formats them (format_data.m), it may be essential to install 'io' package in GNU Octave and can be installed using pkg install -forge io, which can then be loaded in work environment using pkg load io Demo Figure1 generates Figure 1 in the main text; Figure S2 and Figure S3 (Run Time: 22 seconds). All mat files to run this script are avaialble. This figure script requires the scripts IPQ,ITE,ITXD, and ITEXD to be run if parsmeters changed in the analysis in order to update 'ImpactQuarantine.mat'; 'TestingonEntry.mat'; 'TestingonExit_OneDayDelay.mat'; 'TestingonEntryExit_OneDayDelay.mat'. Estimated run times for the analysis scripts on 6 processor in parallel (IPQ ~ 8 hrs; ITE ~ 16 hrs; ITXD ~ 16 hrs; ITEXD ~ 24 hrs) (Note: These scripts run two scenarios. Run time can be cut in half by commenting out one of the scenarios) Instructions for use To generate the Figures and output of the calculations, select a script from Figures section to run in MATLAB and enter the name in the command line. All mat file are availble to generate figures and conduct the calculations. To run analysis on a different set of parameters, adjust the parameters in the script and enter the name of the script in the command line to run. Analysis scripts IS Generates the infectivity profile for R0=2.5 and R0=2.0, as well as pA=30.8% and pA=22.6%. IPQ Runs the analysis for uniform entry into quarantine with 30.8% infections being asymptomatic, and 22.6% being asymptomatic, for quarantine durations from 0 to 21 days with no testing and R0=2.5 ITE Runs the analysis for uniform entry into quarantine with 30.8% infections being asymptomatic, and 22.6% being asymptomatic, for quarantine durations from 0 to 21 days with testing on entry and R0=2.5 ITX Runs the analysis for uniform entry into quarantine with 30.8% infections being asymptomatic, and 22.6% being asymptomatic, for quarantine durations from 0 to 21 days with testing on exit (with negligible delay in test result) and R0=2.5 ITEX Runs the analysis for uniform entry into quarantine with 30.8% infections being asymptomatic, and 22.6% being asymptomatic, for quarantine durations from 0 to 21 days with testing on both entry and exit (with negligible delay in test result) and R0=2.5 ITXD Runs the analysis for uniform entry into quarantine with 30.8% infections being asymptomatic, and 22.6% being asymptomatic, for quarantine durations from 0 to 21 days with testing on exit (with one day delay to obtain test result) and R0=2.5 ITEXD Runs the analysis for uniform entry into quarantine with 30.8% infections being asymptomatic, and 22.6% being asymptomatic, for quarantine durations from 0 to 21 days with testing on both entry and exit (with one day delay to obtain test result) and R0=2.5 ITE4Q7 Runs the analysis for uniform entry into quarantine with 30.8% infections being asymptomatic for quarantine duration of 7 days with testing on both entry and 96 hrs after the first test and R0=2.5 CT Runs the analysis for contact tracing with 30.8% infections being asymptomatic, and 22.6% being asymptomatic, for quarantine durations from 0 to 21 days with no testing and R0=2.5 CTTE Runs the analysis for contact tracing with 30.8% infections being asymptomatic, and 22.6% being asymptomatic, for quarantine durations from 0 to 21 days with testing on entry and R0=2.5 CTTX Runs the analysis for contact tracing with 30.8% infections being asymptomatic, and 22.6% being asymptomatic, for quarantine durations from 0 to 21 days with testing on exit (with negligible delay in test result) and R0=2.5 CTTEX Runs the analysis for contact tracing with 30.8% infections being asymptomatic, and 22.6% being asymptomatic, for quarantine durations from 0 to 21 days with testing on both entry and exit (with negligible delay in test result) and R0=2.5 CTTXD Runs the analysis for contact tracing with 30.8% infections being asymptomatic, and 22.6% being asymptomatic, for quarantine durations from 0 to 21 days with testing on exit (with one day delay to obtain test result) and R0=2.5 CTTEXD Runs the analysis for contact tracing with 30.8% infections being asymptomatic, and 22.6% being asymptomatic, for quarantine durations from 0 to 21 days with testing on both entry and exit (with one day delay to obtain test result) and R0=2.5 VPAD Runs the analysis for uniform entry into quarantine for quarantine durations of five and seven days days with testing on both entry and 96 hrs after the first test and R0=2.5 for 0% to 100% (at 5% increase in incrementes) infections being asymptomatic. VPANT Runs the analysis for uniform entry into quarantine for quarantine durations of five and seven days days with no testing and R0=2.5 for 0% to 100% (at 5% increase in incrementes) infections being asymptomatic. Prob_Estimate_Data Returns the estimates for the benefit of extending the quarantine from three-day and test on entry to a five/seven day quarantine with testing on entry and testing 96 hrs follwoing first test. Also, calcualtes the probability of post-quarantine transmission for the cases detected on exit. DayTest Computes the number of secodnary infections for quarantine durations for 0 to 21 days, as well as evaluting the number of secondary infections if tested on day X of quanratine. AIC_St Runs the fitting and AIC comparison for the sensitivity function Figures FigureGen Generates all figures in the main text and the supplementary material Figure1A Constructs the line graph of the probability of post-quarantine transmission for quarantine for the scenarios of i) no testing, ii) texting on entry, iii) testing on exit, and iv) testing on both entry and exit for an incubation period of 8.29 days . This script also generates Figure S2 (Secondary number of cases post-quarantine) and Figure S3 (Probability of post-quarantine transmission). Figure2 Constructs the bar graph of the positivity for the testing data from the two site for the different strategies (testing only on entry versus testing on entry and exit). Figure =3A Constructs the line graph of the probability of post-quarantine transmission for contact tracing for the scenarios of i) no testing, ii) texting on entry, iii) testing on exit, and iv) testing on both entry and exit for an incubation period of 8.29 days This script also generates Figure S7 and Figure S8. Figure1B_3B Constructs the line graph of the optimal day to test for uniform entry (Figure 1B) and contact tracing (Figure 3B) under the assumption of one day delay to test result and a negligible delay to test result. Figure 4 Plots the optimal day to test for a known time of exposure under the assumption of one day delay to test result and a negligible delay to test result for quarantine periods of 14,7,5, and 3 days. FigureS1 Plots the infectivity profile for R_0=2.5 and 30.8% infections being asymptomatic, R_0=2.5 and 22.6% infections being asymptomatic, R_0=2.0 and 30.8% infections being asymptomatic, and R_0=2.0 and 22.6% infections being asymptomatic. FigureS4 Plots the optimal time to conduct the test for different quarantine durations FigureS5 Plots the probability of post-quarantine transmission for quarantine for a five-day quarantine and a seven day quarantine with no testing and testing on entry and 96 hrs after the initial test for various proportions of infections being asymptomatic FigureS6 Plots the probability of post-quarantine transmission for quarantine for a three-day, five-day, seven-day, and 14-day quarnatine, with testing on entry, testing on exit, and testing on entry and exit when accounting for underlying prevalence in the community and a crew size of 40. FigureS9_S10 This script generates the figures for the secondary number of cases post-quarantine and probability of post-quarantine transmission when entering quarantine through contact tracing. FigureS11 This script generates the infectivity profile for the 8.29 day incubation period and a 2.9, 1.9 and 3.9 latent period FigureS12_S13_S16_S17 This script generates the figures for the secondary number of cases post-quarantine and probability of post-quarantine transmission when entering quarantine uniformly and through contact tracing for a latent period of 1.9 days. FigureS14_S15_S18_S19 This script generates the figures for the secondary number of cases post-quarantine and probability of post-quarantine transmission when entering quarantine uniformly and through contact tracing for a latent period of 3.9 days. FigureS20_S21 This script generates the figures for the secondary number of cases post-quarantine and probability of post-quarantine transmission when entering quarantine not through contact tracing, with no delay between the test sample and test result (i.e. delay is negligible in receiveing test result) FigureS22_S23 This script generates the figures for the secondary number of cases post-quarantine and probability of post-quarantine transmission when entering quarantine through contact tracing, with no delay between the test sample and test result (i.e. delay is negligible in receiveing test result FigureS24_S25 This script generates the figures for the secondary number of cases post-quarantine and probability of post-quarantine transmission when entering quarantine not through contact tracing, with 22.6% infections being asymptomatic FigureS26_S27 This script generates the figures for the secondary number of cases post-quarantine and probability of post-quarantine transmission when entering quarantine through contact tracing, with 22.6% infections being asymptomatic FigureS28 Generates the plot for the temporal diagnostic sensitivity FigureS24_S25 This script generates the figures for the probability of post-quarantine transmission when entering quarantine not through contact tracing, as well as through contact tracing, assuming transmisison post-quarantine is Poisson distributed FigureS30 Generates the plot for pdf in which individuals enter quarantine uniformly versus contact tracing Functions BaselineParameters Returns the baseline parameters for the analysis DurationInfected The duration in which a contact of a symptomatic index case has been infected for FigureChart Generates the Figure chart for the Secondary infections post quarantine with and without testing, assuming delay from sample to result is irrelevant FigureChartDelay Generates the Figure chart for the Secondary infections post quarantine with and without testing, assuming one day delay from sample to result InfectiousnessfromInfection The temporal number of secondary infections post-quarantine given no testing InfectiousnessfromInfectionTestingEntry The temporal number of secondary infections post-quarantine given single test time (does both entry/exit). Entry and exit can both be done using this function as it only requires the time in quarantine in which the test is done InfectiousnessfromInfectionTestingEntryExit The temporal number of secondary infections post-quarantine given timing of two tests Probability_Onward The probability of psot-quarantine transmission given how secondafy cases are distributed RiskChart Generates the Figure chart for the probabaility of post quarantine trasnmisison, assuming delay from sample to result is negligible. It also generates the the equivialent duration of quaratnine with testing to a duration of quarantine with no testing RiskChartDelay Generates the Figure chart for the probabaility of post quarantine trasnmisison, assuming one day delay from sample to result. . It also generates the the equivialent duration of quaratnine with testing to a duration of quarantine with no testing SensitivityTimeSamp Temporal sensitiviity of the RT-PCR assay with respect to time (dependent on the incubation period) SensitivityvsViralLoad The sensitiviity of the RT-PCR assay given the viral load TimeInfection The time in which a contact was infected from a symptomatic index case ViralShedding_Asymptomatic The infectity profile of an asymptomatic individual (independent of R0) ViralShedding_Symptomatic The infectity profile of an symptomatic individual (independent of R0) MAT files TableData_A The data for region A TableData_B The data for region B
- These authors contributed equally to the work on the manuscript: Chad R. Wells and Jeffrey P. Townsend