Diagnostic accuracy and feasibility of a rapid SARS-CoV-2 antigen test in general practice – a prospective multicenter validation and implementation study

Abstract

Background: PCR testing is considered the gold standard for SARS-CoV-2 diagnosis but its results are earliest available hours to days after testing. Rapid antigen tests represent a diagnostic tool enabling testing at the point of care. Rapid antigen tests have mostly been validated by the manufacturer or in controlled laboratory settings only. External validation at the point of care, particularly in general practice where the test is frequently used, is needed. Furthermore, it is unclear how well point of care tests are accepted by the practice staff. Methods: In this prospective multicenter validation study in primary care, general practitioners included adult individuals presenting with symptoms suggesting COVID-19. Each patient was tested by the general practitioner, first with a nasopharyngeal swab for the point of care test (Roche SARS-CoV-2 Rapid Antigen Test) and then with a second swab for PCR testing. Using the RT-PCR result as a reference, we calculated specificity, sensitivity, positive predictive value and negative predictive value, with their 95% confidence intervals. General practitioners and medical assistants completed a survey to assess feasibility and usefulness of the point of care tests. Results: In 40 practices in Würzburg, Germany, 1518 patients were recruited between 12/2020 and 06/2021. The point of care test achieved a sensitivity of 78.3% and a specificity of 99.5% compared to RT-PCR. With a prevalence of 9.5%, the positive predictive value was 93.9% and the negative predictive value was 97.8%. General practitioners rated the point of care test as a helpful tool to support diagnostics in patients with signs and symptoms suggestive for infection, particularly in situations where decision on further care is needed at short notice. Conclusion: The point of care test used in this study showed a sensitivity below the manufacturer’s specification (Sensitivity 96.25%) in the practice but high values for specificity and high positive predictive value and negative predictive value. Although widely accepted in the practice, measures for further patient management require a sensitive interpretation of the point of care test results.