Social distancing boosts quarantine effect on COVID-19 curve: COCHRANE rapid analysis

  • Nussbaumer B, & et al
  • Cochrane Collaboration
  • 9 Apr 2020

  • curated by Liz Scherer
  • Clinical Essentials
Access to the full content of this site is available only to registered healthcare professionals. Access to the full content of this site is available only to registered healthcare professionals.

Takeaway

  • Early quarantine combined with other preventative public health measures (e.g., physical distancing, community closures) forms the optimal strategy for reducing COVID-19 morbidity and mortality.
  • Evidence grade is low-very low in this Cochrane rapid analysis.

Why this matters

  • Quarantine alone might be insufficient to contain SARS-CoV-2 transmission and acquisition.
  • Especially those with suspected exposure should practice physical distancing, avoid community gatherings, and quarantine as much as possible.

Key results

  • 29 studies (10 on COVID-19, 15 SARS, 2 SARS/other infectious diseases, 2 MERS).
  • With quarantine only:
    • Case reductions (4 studies) from 44% (COVID-19) to 81% (SARS).
    • Mortality reductions (2 studies) from 31%-34% (COVID-19) to 63% (SARS).
  • Quarantine + other measures (4 studies): 
    • +case isolation, social distancing: 67% case reduction (ages ≥70, 1 study).
    • +closures, mass gathering cancellations, travel restrictions: R 0 reduced from 2.9 to 2.3 in 2 weeks (1 study).
    • +closures: peak of onward transmission reduced by 29.9%.
    • +restricting travelers from ongoing outbreak countries: small effect.

Study design

  • Cochrane rapid review evaluating effects (transmission, mortality, resource utilization) of quarantine alone or in combination with other measures in:
    • Contacts of confirmed/suspected COVID-19 or persons living in high transmission areas;
    • Returning travelers from countries with declared outbreaks.
  • Funding: Cochrane Collaboration.

Limitations

  • Time constraints.
  • Titles limited to English, Chinese languages.
  • Confounding likely.