Comparing the Diagnostic Accuracy of Point-of-Care Lateral Flow Antigen Testing for SARS-CoV-2 with RT-PCR in Primary Care (REAP-2)

Abstract

Background: Testing for COVID-19 with quantitative reverse transcriptase-polymerase chain reaction (RT-PCR) may result in delayed detection of disease. Antigen detection via lateral flow testing (LFT) is faster and amenable to mass testing strategies. Our study assesses the diagnostic accuracy of LFT compared to RT-PCR on the same primary-care patients in Austria. \n \nMethods: Prospective dataset of 2,562 patients presenting with mild to moderate flu-like symptoms to 20 practices in the district of Liezen, Austria, between October 22 and November 30, 2020. Symptomatic patients received clinical assessment, including both tests, and were split in two groups: Group 1 (true reactive): Suspected COVID-19 cases with a reactive LFT, who tested RT-PCR positive; and Group 2 (false non-reactive): Suspected COVID-19 cases with a non-reactive LFT, who tested RT-PCR positive. We report the number of cases detected with each test, evaluate the correlation of RT-PCR positivity with reactive LFT and report clinical sensitivity and specificity of LFT, positive predictive value (PPV), negative predictive value (NPV), and pre-test duration of symptoms and RT-PCR cycle threshold (Ct) value across groups. Regression analysis quantifies the association between reactive LFT and symptom duration and Ct value respectively. \n \nFindings: Of the 2,562 symptomatic patients, 1,037 were suspected of COVID-19: 826 (79.7%) tested RT-PCR positive 201 (19.8%) RT-PCR negative and 10 (0.5%) with inconclusive RT-PCR. Among patients with positive RT-PCR, 788/826 tested LFT reactive (Group 1) and 38 (4.6%) non-reactive (Group 2); Of those with negative RT-PCR, 179/201 tested LFT non-reactive and 22/201 reactive. Clinical sensitivity (95.4%) and specificity (89.1%), and PPV (97.3%) and NPV (82.5%) were high. Test outcomes of both LFT and RT-PCR were positively correlated (r=0.968,95CI=[0.952,0.985]). Reactive LFT was negatively correlated with Ct value (r=0.2999,p<0.001) and symptom duration (r=-0.1299,p=0.0043) while Ct value was positively correlated with symptom duration (r=0.3733),p<0.001). \n \nInterpretation: We show that LFT at scale during early COVID-19 is an accurate alternative to RT-PCR testing and may assist in curbing resurgence of disease. We note the importance of administering LFT properly, here combined with clinical assessment and delivered at scale in primary care. This needs to be considered when applying LFT as part of mass testing strategies. \n \nFunding Statement: No funding was available for this study. \n \nDeclaration of Interests: None declared. \n \nEthics Approval Statement: The study used secondary anonymised data for which approval was granted by the Institute of Advanced Studies Research Ethics Committee, Austria (reference number: CASE002_2021_HEHP).