Evaluation of the Access Bio CareStartTM rapid SARS-CoV-2 antigen test in asymptomatic individuals tested at a community mass-testing program in Western Massachusetts

Abstract

Background: Point-of-care antigen-detecting rapid diagnostic tests (RDTs) for Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) represent a scalable tool for SARS-CoV-2 infections surveillance. Data on their performance in real-world community settings is paramount for their implementation. Method: We evaluated the accuracy of CareStartTM COVID-19 Antigen test (CareStart) in a testing site in Holyoke, Massachusetts. We compared CareStart to a SARS-CoV-2 reverse transcriptase quantitative polymerase chain reaction (RT-qPCR) reference, using anterior nasal swab samples. We calculated the sensitivity, specificity, and expected positive and negative predictive values at different SARS-CoV-2 prevalence estimates. Results: We performed 666 tests on 591 unique individuals. 573 (86%) were asymptomatic. There were 52 positive tests by RT-qPCR. The sensitivity of CareStart was 49.0% (95% Confidence Interval (CI): 34.8 - 63.4) and specificity was 99.5% (95% CI: 98.5 - 99.9). Among positive RT-qPCR tests, the median cycle threshold (Ct) was significantly lower in samples that tested positive on CareStart. Using a Ct [≤] 30 as a benchmark for positivity increased the sensitivity to 64.9% (95% CI: 47.5 - 79.8). Conclusions: CareStart has a high specificity and moderate sensitivity. The utility of RDTs, such as CareStart, in mass implementation should prioritize use cases in which a higher specificity is more important.