Testing Frequency Matters | An Evaluation of the Diagnostic Performance of a SARS-CoV-2 Rapid Antigen Test in United States Correctional Facilities

Abstract Background: The CDC recommends serial rapid antigen assay collection within congregate facilities for screening and outbreak testing. Though modeling and observational studies from community and long-term care facilities have shown serial collection provides adequate sensitivity and specificity, the diagnostic accuracy of this testing strategy within correctional facilities remains unknown. Methods: Using Connecticut Department of Corrections (DOC) data from November 21st 2020 to June 15th 2021, we estimated the accuracy of a rapid assay, BinaxNOW, under three collection strategies, a single test in isolation and two and three serial tests separated by 1-4 day intervals. Diagnostic accuracy metrics were estimated in relation to RT-PCRs collected within one day before the first or after the last included rapid antigen tests in a series. Results: Of the 17,669 residents who contributed at least one RT-PCR or rapid antigen during the study period, 3,979 contributed [≥]1 paired rapid antigen test series. In relation to RT-PCR, the three-rapid antigen test strategy had a sensitivity of 89.6% (95% confidence intervals: 86.1-92.6%) and specificity of 97.2% (CI: 95.1-98.3%). The sensitivities for two and one-rapid antigen test strategy were 75.2% and 52.8%, respectively, and the specificities were 98.5% and 99.4%, respectively. The sensitivity was higher among symptomatic residents and when the RT-PCR was collected before the rapid antigen tests. Conclusions: We found the serial collection of an antigen test resulted in high diagnostic accuracy. These findings support serial testing within correctional facilities for outbreak investigation, screening, and when rapid detection is required (such as intakes or transfers).