SARS-CoV-2 patient self-testing with an antigen-detecting rapid test: a head-to-head comparison with professional testing

Background Antigen-detecting rapid diagnostic tests (Ag-RDTs) have been widely recommended as a complement to RT-PCR. Considering the possibility of nasal self-sampling and the ease-of-use in performing the test, self-testing may be an option. Methods and Findings We performed a manufacturer-independent, prospective diagnostic accuracy study of nasal mid-turbinate self-sampling and self-testing when using a WHO-listed SARS-CoV-2 Ag-RDT. Symptomatic participants suspected to have COVID-19 received written and illustrated instructions. Procedures were observed without intervention. For comparison, Ag-RDTs with nasopharyngeal sampling were professionally performed. Estimates of agreement, sensitivity, and specificity relative to RT-PCR on a combined oro-/nasopharyngeal sample were calculated. Feasibility was evaluated by observer and participant questionnaires. Among 146 symptomatic adults, 40 (27.4%) were RT-PCR-positive for SARS-CoV-2. Sensitivity with self-testing was 82.5% (33/40 RT-PCR positives detected; 95% CI 68.1-91.3), and 85.0% (34/40; 95% CI 70.9-92.9) with professional testing. The positive percent agreement between self-testing and professional testing on Ag-RDT was 91.4% (95% CI 77.6-97.0), and negative percent agreement 99.1% (95% CI 95.0-100). At high viral load (>7.0 log10 SARS-CoV-2 RNA copies/ml), sensitivity was 96.6% (28/29; 95% CI 82.8-99.8) for both self- and professional testing. Deviations in sampling and testing (incomplete self-sampling or extraction procedure, or imprecise volume applied on the test device) were observed in 25 out of the 40 PCR-positives. Participants were rather young (mean age 35 years) and educated (59.6% with higher education degree). Most participants (80.9%) considered the Ag-RDT as rather easy to perform. Conclusion Ambulatory participants suspected for SARS-CoV-2 infection were able to reliably perform the Ag-RDT and test themselves. Procedural errors might be reduced by refinement of the Ag-RDTs for self-testing, such as modified instructions for use or product design/procedures. Self-testing may result in more wide-spread and more frequent testing. Paired with the appropriate information and education of the general public about the benefits and risks, self-testing may therefore have significant impact on the pandemic.