COVID-19: which 4 antibody tests have the highest sensitivity, specificity — and when?

  • Green DA & al.
  • J Clin Microbiol
  • 8 Jun 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.


  • An IgG serology study shows overall specificity of 98.0%-99.6% and sensitivity of 88.1%-98.8% in samples collected >14 days after disease detection for 4 high-throughput serologic tests for antibodies to SARS-CoV-2: Abbott, Epitope, Euroimmun, and Ortho-Clinical.
  • Repeat molecular diagnostic testing yielding eventual positive results can be considered a true positive for SARS-CoV-2, according to a separate study published in the same journal.

Why this matters

  • Serologic testing yields the most accurate results at >14 days after symptom onset/PCR diagnosis.  
  • Molecular testing should be repeated up to 2 days after a negative result in patients whose symptoms warrant high clinical suspicion for COVID-19.

Key results

    • IgG study:
      • Overall specificity/positive predictive values by assay, assuming 5% prevalence, >14 days since symptom onset:
        • Abbott: 99.6%/92.8%.
        • Epitope: 99.6%/90.6%.
        • Euroimmun: 98.0%/71.2%.
        • Ortho-Clinical: 99.6%/92.5%. 
      • All but 1 (Abbott) hit 100% sensitivity overall among samples >14 days following symptom onset.
    • Diagnostics study:
      • Of 3432 patients, 2630 had initial negative, invalid, or indeterminate results, and 802 had initial positive results.
      • Overall test positivity: 39.9% with repeat testing vs 49.0% with single test (P<.001>
      • Of 2413 patients first testing negative, 18.6% became positive on repeat testing on subsequent days.
        • Negative predictive value: 81.3% (95% CI, 79.7%-82.8%).
      • Same-day repeat testing of patients with invalid, negative, indeterminate results yielded total 17.0% positives initially missed.
      • Positive-negative conversion probability was minimal by 15-20 days after the initial test, and 50% at 28 (95% CI, 27-29) days.

    Study design

    • IgG study: IgG tests evaluated in 224 serially collected serum samples (from 56 patients with confirmed COVID-19).
    • Diagnostics study: evaluation of SARS-CoV-2 molecular assays in dataset covering patients in New York City.
    • Funding: None disclosed.


    • IgG study: small numbers of patients.
    • Diagnostics study: repeat testing in only 15%.