Accuracy of COVID-19 self-tests with unsupervised nasal or nasal plus oropharyngeal self-sampling in symptomatic individuals in the Omicron period

• DOI: 10.1101/2022.03.24.22272891
• Publication Date: 2022-03-25T04:40:10Z
• Authors: Ewoud Schuit; Roderick P Venekamp; Lotty Hooft; Irene K Veldhuijzen; Wouter van den Bi...; Suzan D Pas; Vivian F Zwart; Esther B Lodder; Marloes Hellwich; Marco Koppelman; Richard Molenkamp; Constantijn Wijers; Irene H Vroom; Leonard C Smeets; Carla R S Nagel-I...; Wanda G H Han; Susan van den Hof; Jan AJW Kluytmans; Janneke H M van d...; Karel G M Moons
• Keywords: 03 medical and health sciences; 0302 clinical medicine; 16. Peace & justice; 3. Good health

Abstract Background Performances of rapid antigen diagnostic tests (Ag-RDTs) with nasal self-sampling, and oropharyngeal plus (OP-N) in the Omicron period are unknown. Methods Prospective accuracy study among 6,497 symptomatic individuals aged >16 years presenting for SARS-CoV-2 testing at three test-sites. Participants were sampled RT-PCR (reference test) received one Ag-RDT to perform unsupervised either self-sampling (during emergence Omicron, after share was >90%, phase-1) or OP-N (in a subsequent phase-2; >99%). The evaluated Acon Flowflex (Flowflex; phase-1 only), MP Biomedicals (MPBio), Siemens-Healthineers Clinitest (Clinitest). Findings During phase-1, 45% Flowflex, 29% MPBio, 35% participants confirmatory testers (previously tested positive by self-test own initiative). Overall sensitivities 79.0% (95% CI: 74.7-82.8%) 69.9% (65.1-74.4%) 70.2% (65.6-74.5%) Clinitest. Sensitivities substantially higher (93.6%, 83.6%, 85.7%, respectively) than those who other reasons (52.4%, 51.5%, 49.5%, respectively). decreased 6.1 (p=0.16 Chi-square test), 7.0 (p=0.60), 12.8 (p=0.025) percentage points, respectively, when transitioning from >95% Omicron. phase-2, 53% 44% testers. 83.0% (78.8%-86.7%) MPBio 77.3% (72.9%-81.2%) Comparing sampling, slightly (87.4% 86.1%, respectively), (69.3% 59.9%, Interpretatio Ag-RDTs during but only statistically significant influenced proportion Addition improved Funding Dutch Ministry Health, Welfare, Sport. Research into context Evidence before this require no minimal equipment, provide result within 15-30 minutes, can be used range settings including self-testing home. Self-testing may potentially lower threshold allows obtain test quickly their convenience, which could support early detection infectious cases reduce community transmission. Real world evidence on performance variant is needed accurately inform end-users policymakers. Therefore, we conducted large prospective commercially available (the test, Clinitest) using as reference standard. Our aims evaluate whether accuracies changed over time Omicron; determine addition same swab yielded accuracies. What adds comprehensive almost 6,500 symptoms routine public health service COVID-19 test-sites Netherlands. study, between 21 December 2021 10 February 2022, samples national pathogen surveillance increased first week 99% last study. dominant divided sampling phase (phase-1; present >90% samples) an (phase-2; In visited test-site because (confirmatory testers). Ag-RDT. 94%, 84%, 86%, testers, 52%, 50%, had getting tested. 87.0% (79.7-92.4%), 83.1% (72.9-90.7%), 80.0% (51.9-95.7%), week, points final Phase-2, When comparing 59.9%). Implications all performed trend sensitivity Clinitest, most notably confirm self-test. Based these findings, manufacturers consider extending instructions use include combined evaluating well.