Modeling the Comparative Impact of Individual Quarantine vs. Active Monitoring of Contacts for the Mitigation of COVID-19

Corey M Peak; Rebecca Kahn; Yonatan H Grad; Lauren M Childs; Ruoran Li; Marc Lipsitch; Caroline O Buckee

The relationship between symptoms of a disease and infectiousness to others is critical to the success of containment strategies. Previous work has found that a disease's natural history, particularly the amount of transmission that occurs before symptom onset, greatly influences the ability to control outbreaks 5 and the relative effectiveness of individual quarantine vs. active monitoring. 6 Short-course diseases, such as influenza, and diseases with long periods of presymptomatic infectiousness, like hepatitis A, are impacted more strongly by quarantine than by active monitoring; however, quarantine is of limited benefit over active monitoring for the coronaviruses MERS and SARS, where persons usually show distinctive symptoms at or near the same time that they become infectious. Recent work on isolation for COVID-19 found a potentially large impact of perfect isolation, if one assumed there was limited presymptomatic transmission and a high probability of tracing contacts to be put under isolation immediately following symptom onset. 7 Our framework enables comparison of active monitoring and individual quarantine and considers parameters such as delays, and imperfect isolation to account for known nosocomial transmission of this respiratory virus. 8 One of the key uncertainties surrounding COVID-19 is the extent of asymptomatic and presymptomatic transmission. A recent study reporting asymptomatic transmission in Germany 9 was later found to be incorrect or misleading, 10 adding to the confusion. There has also been uncertainty about the serial interval -the time between symptom onset of infector-infectee pairs -which in turn reflects uncertainty about the extent of presymptomatic transmission. Early estimates by Nishiura et al 11 and Li et al 12 were derived from limited data (24 and 6 infector-infectee pairs, respectively) and in the latter case largely reflected the prior distribution derived from SARS cases in 2003. Given the severe impact of quarantine on both resources and individual liberty, it is vital to assess under what conditions quarantine can effectively control COVID-19, and among these under what conditions it is substantially more effective than less restrictive approaches such as active monitoring, particularly given uncertainty in essential disease parameters. Here, using methods described in previous work 6 and the latest epidemiological parameters reported for COVID-19, 11, 12 we compare the ability of individual quarantine and active monitoring to reduce the effective reproductive number ( ) of COVID-19 R e to below the critical threshold of one. While mass restrictions on movements within cities have been implemented during this outbreak and are sometimes referred to as quarantines, here we focus on the effectiveness of quarantine and active monitoring on an individual basis based on contact tracing.