An intra-host SARS-CoV-2 dynamics model to assess testing and quarantine strategies for incoming travelers, contact management, and de-isolation

Abstract

Non-pharmaceutical interventions (NPIs) remain decisive tools to contain SARS-CoV-2. Strategies that combine NPIs with testing may improve efficacy and shorten quarantine durations. We develop a stochastic within-host model of SARS-CoV-2 that captures temporal changes in test sensitivities, incubation- and infectious periods. We use the model to simulate relative transmission risk for (i) isolation of symptomatic individuals, (ii) contact person management and (iii) quarantine of incoming travelers. We estimated that testing travelers at entry reduces transmission risks to 21.3% ([20.7, 23.9], PCR) and 27.9% ([27.1, 31.1], rapid diagnostic tests; RDT), compared to unrestricted entry. We calculated that 4 (PCR) vs. 5 (RDT) days pre-test quarantine are non-inferior to a 10 days quarantine for incoming travelers and that 8 (PCR) vs. 10 (RDT) days of pre-test quarantine are non-inferior to 14 days post-exposure quarantine. De-isolation of infected individuals 13 days after symptom onset may reduce the transmission risk to <0.2% [<0.01, 6.0].

COVIDStrategyCalculator: A software to assess testing and quarantine strategies for incoming travelers, contact management, and de-isolation

“…The underlying transit compartment model consists of different phases that resemble relevant attributes of the infection: i.e., whether the virus is (i) detectable, the individual (ii) has symptoms, and (iii) may be infectious. In an associated article 6 , we describe the mathematical details of the model, and exemplify the estimation of default parameters that capture clinically observed temporal changes and variability in test sensitivities, incubation and infectious periods, as well as times to symptom onset.…”

“…To enable the design and evaluate the efficacy of NPI strategies in preventing the risk of onward transmission, we present the COVIDStrategyCalculator software. The software implements a stochastic intra-host SARS-CoV-2 dynamics model, presented in an associated article 6 , to assess arbitrary, userdefined, NPI strategies ''on the fly.'' The software focuses on three common scenarios in policy making: (i) contact person management, (ii) quarantine of incoming travelers, and (iii) isolation strategies.…”

“…The software allows full flexibility with regard to parameter choices of the underlying model, which, for example, determine the time course of infection (e.g., the average ''incubation time'' or the ''time of infectiousness''), the proportion of asymptomatic cases, the test sensitivity, and much more. However, a set of default parameters has been derived in an associated article 6 that synthesizes the current knowledge on within-host infection dynamics of SARS-CoV-2 and temporal test sensitivity.…”

Rapid incidence estimation from SARS-CoV-2 genomes reveals decreased case detection in Europe during summer 2020

“…We used GInPipe to detect changes in SARS-CoV-2 case detection. Let us denote by P t (tested|infected) the proportion of infected individuals that are actually diagnosed with the virus in week t. According to Bayes' theorem we have P t (tested|infected) = P t (infected|tested)•P t (tested) P t (infected) (2) where P t (infected|tested) denotes the proportion of tested individuals that are infected, P t (tested) the proportion of individuals that are tested and P t (infected) the proportion currently infected in week t. We calculate P t (infected|tested) = r pos −(1−spec) sens− (1−spec) from the positivity rate r pos of the conducted tests, corrected for the clinical sensitivity sens= 0.7 and specificity spec= 0.999 of the diagnostic tests [57]. For calculating the probability of being tested P(tested), we considered linear-, Poisson-and Binomial models, all of which yielded identical results.…”

Rapid incidence estimation from SARS-CoV-2 genomes reveals decreased case detection in Europe during summer 2020

“…We used GInPipe to detect changes in SARS-CoV-2 case detection. Let us denote by P t (tested|infected) the proportion of infected individuals that are actually diagnosed with the virus in week t. According to Bayes' theorem we have P t (tested|infected) = P t (infected|tested)•P t (tested) P t (infected) (2) where P t (infected|tested) denotes the proportion of tested individuals that are infected, P t (tested) the proportion of individuals that are tested and P t (infected) the proportion currently infected in week t. We calculate P t (infected|tested) = r pos −(1−spec) sens−(1−spec) from the positivity rate r pos of the conducted tests, corrected for the clinical sensitivity sens= 0.7 and specificity spec= 0.999 of the diagnostic tests [54]. For calculating the probability of being tested P(tested), we considered linear-, Poisson-and Binomial models, all of which yielded identical results.…”