Diagnostic performance of patient self-collected oral swab (tongue and cheek) in comparison with healthcare worker-collected nasopharyngeal swab for severe acute respiratory syndrome coronavirus-2 detection

Abstract

The present study was conducted to compare the performance of patient self-collected oral swab (OS) with healthcare worker (HCW)-collected nasopharyngeal swab (NPS) for SARS-CoV-2 detection by reverse transcription polymerase chain reaction (RT-PCR) in real-world setting. Paired OS and NPS were collected from 485 consecutive individuals presenting with symptoms of coronavirus disease-19 (COVID-19) or asymptomatic contacts of COVID-19 cases. Both specimens were processed for RT-PCR and cycle threshold (Ct) value for each test was obtained. Positive percent agreement (PPA), negative percent agreement (NPA), overall percent agreement (OPA) and kappa were calculated for OS RT-PCR compared with NPS RT-PCR as reference. A total of 116/485 (23.9%) participants were positive by NPS RT-PCR. OS had PPA of 71.6%, NPA of 98.8%, OPA of 92.4% and kappa of 0.771. Almost all participants (483/485, 99.6%) reported OS as a convenient and comfortable sample for SARS-CoV-2 testing over NPS. All participants with Ct values <25 and majority (90.8%) with Ct values <30 were detected by OS. To conclude, OS self-sampling was preferred in comparison with NPS due the ease and comfort during collection. The performance of OS RT-PCR for SARS-CoV-2 detection, however, was sub-optimal in comparison with NPS RT-PCR.