Abstract
Background and aims: Accurate and timely diagnosis is crucial for coronavirus disease 2019 (COVID-19) outbreaks. Antigen-detecting rapid diagnostic tests (Ag-RDTs) are easily accessible and affordable, producing rapid results. They are an alternative to the limited gold-standard real-time reverse-transcription polymerase chain reaction (rRT-PCR) tests. This study assessed the performance of Ag-RDTs for COVID-19 outbreaks in institutional settings with high disease prevalence in Kelantan State, Malaysia.
Methods: This study analyzed a total of 303 individuals from five institutional outbreaks with paired nasopharyngeal specimens tested for COVID-19 by Ag-RDTs and rRT-PCR. The diagnostic performance of Ag-RDTs was evaluated through rRT-PCR as the gold standard based on cycle threshold (Ct) value, disease prevalence, and manufacturers.
Results: There was a moderate agreement between Ag-RDTs and RT-PCR (κ=0.603; 95% CI: 0.520- 0.686; P<0.001). The overall specificity was 97.9% (95% CI: 94.1%-99.6%), sensitivity was 63.3% (95% CI: 55.3%-70.8%), accuracy Ag-RDTs was 81.2% (95% CI: 76.4%-85.5%), while positive and negative predictive value was 96.6% (95% CI: 90.2%-98.9%) and 74.1% (95% CI: 70.0%-77.9%), respectively. Further, lower median Ct was reported in 100 (33.0%) true-positive cases compared to 58 (19.1%) false-negative cases (20.3 vs 31.4, P<0.001). The sensitivity was higher (P<0.001) in those with high viral load (Ct value≤25.0) with better performance and a prevalence>10%. In addition, no significant difference was observed between the studied manufacturers.
Conclusion: The Ag-RDTs performed well in diagnosing COVID-19 among outbreaks with higher viral load and disease prevalence. High-risk cases tested negative by Ag-RDTs may have low viral load and require confirmation by rRT-PCR.