Quarantining potentially infected individuals is a widely used approach for controlling the spread of viruses and other infectious pathogens. Even in the case of the current coronavirus disease 2019 (COVID-19) pandemic, caused by the novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), a 14-day quarantine has been recommended by the US Centers for Disease Control and Prevention (CDC) and the World Health Organization (WHO). This is based on our current understanding of the virus’s incubation period – the period of time it takes for pre-symptomatic individuals to show signs of infection.
*Important notice: medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.
Experts have raised concerns that this quarantine period is very burdensome and does not offer much added benefit compared to shorter quarantine durations that could increase compliance because of its reduced disruptions to day-to-day activities.
A frequently proposed strategy for shortening the quarantine duration without an increase in risk is using an RT-PCR test or a rapid antigen test to “test out” of quarantine. Recently, the CDC stated that it may be acceptable sometimes to quarantine for 7 days with a negative RT-PCR test using a sample collected within 48 hours of exit, or even quarantining for 10 days with no test. However, RT-PCR tests are not 100% reliable and can miss infections, especially if someone is pre-symptomatic.
Estimating the proportion of infections undetected by different testing methods in various quarantine durations
In a recent medRxiv* preprint paper, researchers from Wake Forest and Johns Hopkins Universities in the USA discussed their attempt to quantify the proportion of infections that are not detected by different testing methods and quarantine durations. The researchers calculated the probability of infected individuals testing negative for SARS-CoV-2 on a given day post-infection and remaining symptom-free for a time period.
To estimate these probabilities, they designed a biologically plausible model under the assumption that symptom onset and test sensitivity are independently linked to an unobserved point in time post-infection when the viral load crosses a critical threshold.
“An RT-PCR test after the first few days of quarantine appreciably decreases the likelihood of an infected individual going undetected.”
Infected individuals have 20.1% chance of testing RT-PCR negative on day 5 post-infection
The results of the study showed that infected individuals have a 20.1% chance (95% CI 9.8-32.6) of testing negative on RT-PCR on day 5 post-infection and remaining asymptomatic until day 7. They also suggested that the additional information offered by a test decreases as they move further from the test date.
We estimate that an infected individual has a 20.1% chance (95% CI 9.8-32.6) of testing RT-PCR negative on day five post-infection and remaining asymptomatic until day seven, similar to an 8-day quarantine without a test.”
Testing negative on RT-PCR or other tests can significantly increase the confidence that a quarantined individual is uninfected. However, a negative test’s importance goes down each day an individual goes after the test without developing any symptoms. Thus, the authors concluded that a less sensitive test that gives rapid results might be preferable to a more sensitive test with delayed results.
According to the authors, the significance of timely test results has also been proven to be critical in other contexts, and it is the reason why a negative test on day 5 post-infection offers little benefit compared to 7 days quarantine without a test.
A negative test can substantially increase the confidence that an individual in quarantine is uninfected, potentially achieving an impact similar to that of quarantines 3-4 days longer.”
*Important notice: medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.
The long-term cardiac impact of COVID-19