Diagnostic accuracy of SARS-CoV-2 rapid antigen self-tests in asymptomatic individuals in the Omicron period: cross sectional study (preprint)

ABSTRACT

Objectives To assess the performances of three commonly used rapid antigen diagnostic tests (Ag-RDTs) used as self-tests in asymptomatic individuals in the Omicron period. Design Cross-sectional diagnostic test accuracy study. Setting Three public health service COVID-19 test sites in the Netherlands. Participants 3,600 asymptomatic individuals aged ≥16 years presenting for SARS-CoV-2 testing for any reason except confirmatory testing after a positive self-test. Interventions Participants were sampled for RT-PCR (reference test) and received one self-test (either Acon Flowflex (Flowflex), MP Biomedicals (MPBio), or Siemens-Healthineers Clinitest (Clinitest)) to perform unsupervised at home within three hours and blinded to the RT-PCR result. Main Outcome(s) and Measures(s) Diagnostic accuracies (sensitivity, specificity, positive and negative predictive values) of each self-test compared to RT-PCR. Results Overall sensitivities of the three self-tests were 27.5% (95% CI 21.3-34.3%) for Flowflex, 20.9% (13.9-29.4%) for MPBio, and 25.6% (19.1-33.1%) for Clinitest. After applying a viral load cut-off (≥5.2 log10 SARS-CoV-2 E-gene copies/mL), sensitivities increased to 48.3% (95% CI 37.6-59.2%), 37.8% (22.5-55.2%), and 40.0% (29.5-51.2%), respectively. No consistent differences were found in sensitivities by COVID-19 vaccination status, having had a prior SARS-CoV-2 infection, gender or age across the three self-tests. Specificities were >99% for all tests in most analyses. Conclusions The sensitivities of three commonly used SARS-CoV-2 Ag-RDTs when used as self-tests in asymptomatic individuals in the Omicron period, were very low. Our findings indicate that Ag-RDT self-testing in asymptomatic individuals may only detect the minority of infections at that point in time and may not be sufficient to prevent the spreading of the virus to other (vulnerable) persons. Repeated self-testing in case of a negative self-test is advocated to improve the diagnostic yield of the self-tests, and individuals should certainly be advised to re-test when symptoms develop. Summary box What is already known on this topic If sufficiently reliable, SARS-CoV-2 self-testing by asymptomatic persons prior to admission in places where groups gather could have a huge public health impact by lowering the reproduction number or keep it below one for longer periods. Current evidence suggests that SARS-CoV-2 rapid antigen diagnostic tests (Ag-RDTs) when used as self-tests by asymptomatic individuals perform suboptimal, but sample sizes of the previous studies were too small to draw robust conclusions, and also empirical data on the accuracy of Ag-RDT self-tests in asymptomatic individuals during the Omicron period are scarce. What this study adds Compared to RT-PCR testing, overall sensitivities of three commercially available SARS-CoV-2 Ag-RDTs when used as self-tests by asymptomatic individuals (primary analysis population of non-confirmatory testers; n= 3600, 87% of full analysis population) in the Omicron period, were very low 27.5% (95% CI 21.3-34.3%) for the Acon Flowflex test, 20.9% (13.9-29.4%) for the MP Biomedicals test, and 25.6% (19.1-33.1%) for the Siemens Healthineers Clinitest Ag-RDT, which increased to 48.3% (95% CI 37.6-59.2%), 37.8% (22.5-55.2%), and 40.0% (29.5-51.2%), respectively, when applying a viral load cut-off (≥5.2 log10 SARS-CoV-2 E-gene copies/mL). Our findings indicate that Ag-RDT self-testing in asymptomatic individuals may only detect the minority of infections at that point in time and may not be sufficient to prevent the spreading of the virus to other (vulnerable) persons. Repeated self-testing in case of a negative self-test is advocated to improve the diagnostic yield of the self-tests, and individuals should certainly be advised to re-test when symptoms develop.