Saliva Has High Sensitivity and Specificity for Detecting SARS-CoV-2 Compared to Nasal Swabs but Exhibits Different Viral Dynamics from Days of Symptom Onset

Abstract

Background/Objectives: Saliva as a diagnostic medium for COVID-19 requires fewer resources to collect and is more readily adopted across a range of testers. Our study compared an Emergency Use Authorized direct saliva-to-RT-qPCR test against an FDA-authorized nasal swab RT-qPCR assay for participants who reported symptoms of respiratory infection. Methods: We analyzed 737 symptomatic participants who self-selected to test at either a community testing facility or a walk-in clinic due to respiratory symptoms and provided matched saliva and nasal swab samples. Samples were collected between March and September of 2023, both before and after the declared end of the public health emergency. Results: A total of 120 participants tested positive in at least one of the tests. For participants testing in the first 5 days of reported symptoms, the saliva test had a 94.0 positive percent agreement (PPA; 95% C.I. 88.9-99.1%) with the nasal test and a 99.0 negative percent agreement (NPA; 95% C.I. 98.1-99.9%). The viral load decreased beyond day 1 of reported symptoms for saliva testing. Viral load increased up to day 4 for nasal swabs and then decreased. The same number of discordant positive samples (five each) occurred for both tests within 5 days of symptoms onset. Conclusions: In the endemic phase of COVID-19 and for development of new tests, testing methods that are less invasive are more likely to be adopted. The results of saliva-based versus nasal swab PCR measurements relative to days of symptom onset are needed to optimize future testing strategies.

Keywords: COVID-19; SARS-CoV-2; diagnostic; qPCR; saliva.

Figure 1

Average Ct values for nasal swab and saliva samples and the difference between the two values by number of symptomatic days prior to testing.