Skip to main content
. 2021 Jul 19;6(1):135–156. doi: 10.20411/pai.v6i1.422

Table.

The Benefits and Limitations of Antigen and RT-PCR Testing1 for SARS-CoV-22

Parameter Antigen Testing RT-PCR Comments
Sensitivity Moderate High Nucleic acid amplification tests are generally more sensitive than antigen detection tests. Early infection (ie, lower viral loads) would be expected to be missed more frequently with tests with lower sensitivity.
Specificity High High Specificity depends on assay design (eg, the antigen-antibody interaction for antigen tests, and primer/probe selection for RT-PCR, among a variety of other factors).
Cost Low Moderate to High Cost will vary based on platform and the inclusion of nucleic acid extraction step prior to RT-PCR.
Time-to-Result Fast Fast-to-Slow Antigen test results are generally available in 30 minutes or less. There are some isothermal and one RT-PCR platform that offer quick time-to-results (ie, an hour or less), but generally testing with in-laboratory RT-PCR can be completed within a day. Time to result varies significantly by platform, transport time, and specimen backlog.
PPV3 in a High Pretest Likelihood Setting4 High High When there is a high pretest likelihood of infection/disease, antigen tests perform very well, despite a sensitivity that is not as great as RT-PCR.
PPV3 in a Low Pretest Likelihood Setting5 Low-to-Moderate Moderate All tests produce false positive reactions when used in low prevalence settings, but generally antigen detection tests suffer from this to a greater degree than RT-PCR assays.6
1.

The RT-PCR assays considered here include a nucleic acid extraction prior to amplification and are performed in a laboratory, unless otherwise noted.

2.

This comparison assumes well-designed assays that are performed according to the manufacturer's instructions or current Emergency Use Authorization (EUA) guidance.

3.

PPV = Positive Predictive Value of the test result (ie, a positive test truly represents the presence of SARS-CoV-2).

4.

A high pretest likelihood setting would consist of a patient demonstrating signs and symptoms consistent with COVID-19 and/or testing occurring in the setting of a high prevalence of disease in the community.

5.

A low pretest likelihood setting would consist of a patient without the signs and symptoms consistent with COVID-19 (eg, asymptomatic screening) and a low prevalence of disease in the community.

6.

The higher rate of false-positive reactions for antigen detection tests is likely because only a single aberrant reaction (ie, an antigen-antibody reaction) is necessary for a false positive test result, whereas for traditional RT-PCR several aberrant reactions (eg, 2 primers and a probe mis-hybridization) must occur to produce an erroneously positive test result. Specimen mis-handling (eg, splashing, mix ups) can occur with any test and are not considered here.