COVID-19: antibodies last at least 5 months, substantially reduce reinfection risk

  • Hall V & al.
  • Public Health England
  • 15 Jan 2021

  • 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.

As an exception during this period of health crisis, some of the publications mentioned are at the time of writing still in pre-publication, undergoing peer review and subject to change. The results of this pre-print study should be interpreted with utmost caution.

  • Interim SIREN (SARS-CoV-2 Immunity and Reinfection EvaluatioN) findings suggest that a history of SARS-CoV-2 infection confers significant protection against reinfection for at least 5 months.

Why this matters

  • Carrying antibodies may not be 100% effective against reinfection.
  • The authors say, however, that the substantial protective effect they identified is the "minimum likely effect."

Key results

  • 20,787 enrolled (6614 in positive cohort, 14,173 in negative cohort).
  • 84% were women, and 88% were White.
  • Median age, 45.9 (interquartile range, 35.8-53.6) years.
  • 409 new infections.
  • 79% of PCR-positive cases were symptomatic, 12% were asymptomatic, and no data were available for 9%.
  • Reinfection (44 total events):
    • 3.3 reinfections/100,000 follow-up days in the positive cohort vs
    • 17.0 new infections/100,000 follow-up days in the negative cohort.
  • On multivariate analysis:
    • Probable reinfections (2 cases): aOR, 0.01 (P<.01>
    • Combined probable + symptomatic possible (15 cases): aOR, 0.06 (P<.01>
    • Probable + all possible (44 cases): aOR, 0.17 (P<.01>

Study design

  • Prospective UK multicenter cohort study, including health care workers in months following initial infection, June 18-November 9, 2020.
  • Funding: Department of Health and Social Care and Public Health England; Scottish, Welsh, and Northern Irish governments.


  • Limited data, samples.
  • Measurement error (primary infection onset date).
  • Recall bias.
  • Short follow-up.