Containing Emerging Epidemics: a Quantitative Comparison of Quarantine and Symptom Monitoring

Corey M Peak; Lauren M Childs; Yonatan H Grad; Caroline O Buckee

It is widely known that generation intervals are difficult to observe in nature, but challenges also arise in measurements of generation intervals in simulations. For example, generation time distributions may change over the course of an epidemic due to depletion . CC-BY-ND 4.0 International license is made available under a The copyright holder for this preprint (which was not peer-reviewed) is the author/funder. It . https://doi.org/10.1101/072652 doi: bioRxiv preprint of the susceptible individuals (1) . Analogous to the approach of (2), our method of direct observation of simulated serial intervals is restricted to the exponential growth period of an epidemic, as produced by a branching model. However, the potential for the length of generation intervals to be under-estimated near the peak of an epidemic can cause a downward bias in the published serial intervals upon which our parameterization methods are based (2, 3) . Therefore, this downward bias can result in a bias towards "faster" disease parameters (namely, a leftward bias in !""#$% , !"# , and ! ). Such a bias would reduce the simulated effectiveness of all interventions considered in this paper. for influenza A and hepatitis A and by <10% for pertussis. However, due to a much shorter incubation period of influenza A versus hepatitis A ( Table 2) , the relative costeffectiveness measured by the reduction per day of quarantine (outlined bars) is substantially higher for influenza A than hepatitis A. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder. It . https://doi.org/10.1101/072652 doi: bioRxiv preprint Fig S3. Partial rank correlation coefficients for all outcomes. Partial rank correlation coefficients (x-axis) measuring the influence of disease characteristics and intervention performance metrics (rows) on the impact, comparative effectiveness, and comparative cost-effectiveness of the interventions under study. Disease-specific estimates are shown with colored bars and pooled estimates with larger grey bars. For example, increasing the delay in tracing a named contact DCT has a generally small effect negative effect on RS-RQ when pooled across diseases (large grey bar), but for influenza A specifically (purple bar), DCT has a rather large negative effect on RS-RQ. Note that pooled estimates for comparative cost-effectiveness are not available due to non-monotonic relationships across diseases.