Modeling layered non-pharmaceutical interventions against SARS-CoV-2 in the United States with Corvid

  1. Dennis L. Chao
  2. Assaf P. Oron
  3. Devabhaktuni Srikrishna
  4. Michael Famulare

Reviewed by

Read the full article

Listed in

Log in to save this article

Abstract

Background

The novel coronavirus SARS-CoV-2 has rapidly spread across the globe and is poised to cause millions of deaths worldwide. There are currently no proven pharmaceutical treatments, and vaccines are likely over a year away. At present, non-pharmaceutical interventions (NPIs) are the only effective option to reduce transmission of the virus, but it is not clear how to deploy these potentially expensive and disruptive measures. Modeling can be used to understand the potential effectiveness of NPIs for both suppression and mitigation efforts.

Methods and Findings

We developed Corvid, an adaptation of the agent-based influenza model called FluTE to SARS-CoV-2 transmission. To demonstrate features of the model relevant for studying the effects of NPIs, we simulated transmission of SARS-CoV-2 in a synthetic population representing a metropolitan area in the United States. Transmission in the model occurs in several settings, including at home, at work, and in schools. We simulated several combinations of NPIs that targeted transmission in these settings, such as school closures and work-from-home policies. We also simulated three strategies for testing and isolating symptomatic cases. For our demonstration parameters, we show that testing followed by home isolation of ascertained cases reduced transmission by a modest amount. We also show how further reductions may follow by isolating cases in safe facilities away from susceptible family members or by quarantining all family members to prevent transmission from likely infections that have yet to manifest.

Conclusions

Models that explicitly include settings where individuals interact such as the home, work, and school are useful for studying the effectiveness of NPIs, as these are more dependent on community structure than pharmaceutical interventions such as vaccination. Corvid can be used to help evaluate complex combinations of interventions, although there is no substitute for real-world observations. Our results on NPI effectiveness summarize the behavior of the model for an assumed set of parameters for demonstration purposes. Model results can be sensitive to the assumptions made about disease transmission and the natural history of the disease, both of which are not yet sufficiently characterized for SARS-CoV-2 for quantitative modeling. Models of SARS-CoV-2 transmission will need to be updated as the pathogen becomes better-understood.

Article activity feed

  1. ScreenIT

    SciScore for 10.1101/2020.04.08.20058487: (What is this?)

    Please note, not all rigor criteria are appropriate for all manuscripts.

    Table 1: Rigor

    NIH rigor criteria are not applicable to paper type.

    Table 2: Resources

    No key resources detected.

    Results from OddPub: Thank you for sharing your code.

    Results from LimitationRecognizer: We detected the following sentences addressing limitations in the study:
    Our model has many obvious limitations. It was developed as the epidemiology of SARS-CoV-2 was becoming understood early in the pandemic. We assumed that SARS-CoV-2 could be modeled as a more transmissible influenza with a longer serial interval. The seasonality of SARS-CoV-2 transmission is not yet known and is not part of the model. Our model does not estimate hospitalizations or deaths due to COVID-19. Morbidity and mortality rate estimates are rapidly evolving and depend on local health facilities, and they can be applied to our model output of SARS-CoV-2 infections. Hospitals, healthcare workers, and other high-risk professionals are not included in our model. Finally, our work is based on an older influenza model [13], so it uses US Census data from the year 2000. The 2000 Census included a detailed commuter survey that Corvid requires to simulate movement between census tracts, so it would be difficult to update it with newer Census data. So far, few models of SARS-CoV-2 explicitly include key settings of transmission [10, 22]. We believe that these kinds of models are required to mechanistically simulate combinations of NPIs, which would be difficult to simulate without considering how these interventions target transmission in specific settings but may increase transmission elsewhere. However, human behavior in response to NPIs is difficult to model, and we believe that model results should be interpreted cautiously. Many of the population structure assumptions in Co...

    Results from TrialIdentifier: No clinical trial numbers were referenced.

    Results from Barzooka: We did not find any issues relating to the usage of bar graphs.

    Results from JetFighter: We did not find any issues relating to colormaps.

    Results from rtransparent:
    • Thank you for including a conflict of interest statement. Authors are encouraged to include this statement when submitting to a journal.
    • Thank you for including a funding statement. Authors are encouraged to include this statement when submitting to a journal.
    • No protocol registration statement was detected.

    About SciScore

    SciScore is an automated tool that is designed to assist expert reviewers by finding and presenting formulaic information scattered throughout a paper in a standard, easy to digest format. SciScore checks for the presence and correctness of RRIDs (research resource identifiers), and for rigor criteria such as sex and investigator blinding. For details on the theoretical underpinning of rigor criteria and the tools shown here, including references cited, please follow this link.

  2. Version published to 10.1101/2020.04.08.20058487v1 on medRxiv