COVID-19 Testing: Saliva NAAT Comparable to Nasopharyngeal Swabs

Coronavirus 2019-nCoV Sample. New Epidemic Corona Virus. Corona virus outbreaking. Corona Virus in Lab. Scientist hold tube with Test with the Virus Name Coronavirus. sputum examination
Study authors assessed the diagnostic accuracy of saliva NAAT for COVID-19 due to the increased resources, such as trained personnel, required for nasopharyngeal swab NAAT.

Saliva nucleic acid amplification testing (NAAT) for the diagnosis of COVID-19 was found to be comparable to nasopharyngeal swab NAAT during a systematic review and meta-analysis. These results were published in JAMA Internal Medicine.

Researchers searched publication databases through August 2020 for articles on methods for detecting COVID-19 infections. This meta-analysis included 16 studies conducted throughout the United States, Asia, Europe, and Australia.

All studies tested for COVID-19 in pairs of samples to maintain diagnostic accuracy. The average number of sample pairs among the studies was 370 (standard deviation [SD], 616). Patients were tested in a variety of situations, including at hospital admission, when returning from traveling, and at drive-through testing sites. A minority of studies (n=5) tested only individuals who presented with COVID-19 symptoms.

Of the 5922 samples included in this meta-analysis, 941 were positive for COVID-19.

The pooled sensitivity for saliva NAAT was 83.2% (95% credible interval [CrI], 77.4-91.4) and specificity was 99.2% (95% CrI, 98.2-99.8). The sensitivities of each individual study ranged from 60.6% to 89.4% and the specificities from 98.8% to 99.3%.

Among only those studies that underwent peer review, the pooled saliva NAAT sensitivity was 85.6% (95% CrI, 77.0-92.7) and specificity was 99.1% (95% CrI, 98.0-99.8).

The pooled nasopharyngeal swab NAAT sensitivity was 85.7% (95% CrI, 76.5-93.4) and specificity was 98.9% (95% CrI, 97.4-99.7).

In studies that were conducted in ambulatory settings (n=9), the pooled saliva NAAT performance (sensitivity, 84.5%; specificity, 99.0%) remained comparable to that of the pooled nasopharyngeal swab NAAT performance (sensitivity, 88.0%; specificity, 98.7%).

This analysis was limited by the underlying studies, which were dominated by those which recruited patients without COVID-19 symptoms, indicating that sicker patients were underrepresented.

These data suggested that saliva NAAT may be as effective as nasopharyngeal swab NAAT for detecting COVID-19 infections, especially in the context of large, community-based testing programs.


Butler-Laporte G, Lawandi A, Schiller I et al. Comparison of saliva and nasopharyngeal swab nucleic acid amplification testing for detection of SARS-CoV-2: a systematic review and meta-analysis. JAMA Intern Med. 2021;181(3):353-360. doi:10.1001/jamainternmed.2020.8876