Agreement Between the Rapid Antigen Test for SARS-CoV-2 and the RT-qPCR Diagnostic Test

Abstract

Objective: to assess the agreement between the rapid antigen testing (RATs) for SARS-CoV-2 and quantitative reverse transcription polymerase chain reaction (RT-qPCR). Methods: analytical CROSS-sectional study, conducted in three medical units from January 1, 2021, to June 30, 2022, in Mexico City, Mexico. Non-probabilistic sampling was performed using data from the Online Notification System for Epidemiological Surveillance. Data analysis was performed using frequency measures, Cohen’s kappa index, and maximum likelihood estimation. Results: Of 2173 participants with both diagnostic tests: 565 respiratory samples were concordant with a positive result (26.09% CI95% 24.25 - 27.99%), and 1229 with a negative result (56.55% CI95% 54.44 - 58.65%). The sensitivity of RAT VERSUS RT-qPCR was estimated to be 65.17% (95%CI 61.99 - 68.33%), while the specificity was 94.10% (95%CI 92.82 - 95.38%); a positive predictive value of 88.01% (95%CI 85.23 - 90.41%), and a negative predictive value of 80.27% (95%CI 78.19 - 82.24%). The Cohen’s kappa index was 0.62 (substantial agreement), and a calculated likelihood ratio of 40% at pre-test prevalence, a post-test probability of 88.3% was observed for a positive result with RAT in case of having COVID-19. Conclusion: The present study demonstrated substantial concordance between RAT and RT-qPCR, supporting the feasibility of using both tests. This provides clinicians with a valuable tool for informed decision making in the diagnostic context of COVID-19.