How accurate is a 15-minute COVID antigen strip?
COVID-19 Ag Respi-Strip was highly specific (99.5%) but only moderately sensitive (57.6%) vs PCR, best when viral load was high (CT <22).
Source
Development and Potential Usefulness of the COVID-19 Ag Respi-Strip Diagnostic Assay in a Pandemic Context
What they did
Developed a colloidal-gold ICT targeting conserved SARS-CoV nucleoprotein, then measured analytical reproducibility/cross-reactivity and clinical performance on 328 nasopharyngeal aliquots against qRT-PCR.
What they found
Between-observer disagreement 1.7%, robustness 98%, no cross-reactivity with other virus-infected samples. Overall sensitivity 57.6%, specificity 99.5%, accuracy 82.6%; ICT cut-off around CT <22. Authors propose first-line use complementary to molecular tests.
The limits
What it doesn't show
Moderate sensitivity means many PCR-positive cases are missed, especially at higher CT; evaluation was during an ascending epidemic curve and does not replace PCR confirmation.
Key terms
- ICT
- Immunochromatographic test using membrane + colloidal gold for rapid antigen detection.
- Nucleoprotein antigen
- Conserved SARS-CoV/SARS-CoV-2 target used by the monoclonal antibodies.
- Sensitivity
- Share of PCR-positive samples the antigen strip correctly flags positive.
- Specificity
- Share of PCR-negative samples the strip correctly flags negative.
- CT value
- PCR cycle threshold; lower CT usually means higher viral load.
- qRT-PCR
- Reference molecular standard used to judge the antigen assay.
Flashcards
Research intelligence for this paper
See its role on concept claims, tensions it is part of, placement history, and related discoveries.
Quiz yourself
Overall sensitivity of the antigen strip was about:
Common questions
Clinical sensitivity/specificity?
57.6% sensitivity and 99.5% specificity overall.
How fast?
Authors frame it as a ~15-minute first-line antigen assay.
Clinical sample size?
328 nasopharyngeal aliquots across three labs.
When does it work best?
Around CT <22 (higher viral load).
More on Diagnostic accuracy