SARS-CoV-2 Testing With Saliva vs. Nasopharyngeal Swabs

Quick Takes

  • Sample collection through saliva or nasopharyngeal swabbing does not differ significantly in sensitivity.
  • Confidence intervals were wide in stratified analyses. Whether certain subgroups would benefit from a specific approach to sample collection is unclear.
  • Collection of specimens by saliva compared with nasopharyngeal swab was estimated to save $636,105 per 100,000 persons sampled.

Study Questions:

What is the difference in sensitivity and cost for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) detection between nasopharyngeal swabs and saliva testing?


The authors performed a systematic review and meta-analysis examining studies including ≥5 paired nasopharyngeal swab and saliva samples and reporting diagnostic accuracy for SARS-CoV-2 detection. The primary outcome of interest was the difference between saliva samples and nasopharyngeal swabs in sensitivity for SARS-CoV-2 detection. A positive result with either sample was considered the reference standard. Specificity could not be determined because any positive result was assumed to be true. Subgroup analyses assessed difference in sensitivity according to age (adult or pediatric), presence of symptoms, study setting, method of saliva collection, analytic form, and use of transport media. Costs associated with materials (swabs, transport media, containers, and personal protective equipment) and personnel to collect the samples were accounted for to estimate the per-person cost for each test separately.


Overall, 37 studies were included, comprising 7,169 participants with 7,332 paired saliva samples and nasopharyngeal swabs. A total of 2,327 (32%) were positive on either nasopharyngeal swab or saliva. The sensitivity for salivary testing was overall 3.4% lower (95% confidence interval [CI], -9.9% to +3.1%) compared to the nasopharyngeal swab. The pooled sensitivity for saliva testing was 86.9% (95% CI, 82.3%-90.4%). There were no major differences in stratified analyses, with differences in sensitivity between salivary and nasopharyngeal testing ranging between -9.3% to +1.5%. Heterogeneity was high in stratified analyses. CIs for differences between salivary and nasopharyngeal testing were very wide for stratified analyses. Nasopharyngeal testing would identify 79 more persons with SARS-CoV-2 per 100,000 persons sampled. Collection of specimens by saliva compared with nasopharyngeal swab was estimated to save $636,105 (95% uncertainty interval [UI], $467,427 to $831,770) per 100,000 persons sampled.


Sample collection through saliva or nasopharyngeal swabbing does not differ significantly in sensitivity. Salivary collection, however, is much less costly.


Nasopharyngeal sample collection for SARS-CoV-2 presents many challenges beyond the discomfort of a frontal lobe tickle, including requiring a trained health care professional and personal protective equipment. A salivary test is not traumatic and can be self-administered in any location, making testing for SARS-CoV-2 convenient and facilitating serial testing. As testing for SARS-CoV-2 is now increasingly becoming a requirement for travel and other purposes, salivary sample collection appears to be a viable and more cost-effective alternative to nasopharyngeal swabbing and can help expand testing. This relatively large meta-analysis confirms findings from previous studies suggesting that a salivary sample collection is likely to be equivalent in sensitivity to nasopharyngeal swabbing. The highest yield approach and target patient population remains ill-defined, as the CIs of the subgroup analyses were very wide. Nevertheless, the trade-offs appear significantly in favor of salivary testing, since nasopharyngeal swabbing is imperfect, and patients with a high pretest probability for COVID-19 should be treated as such a regardless of testing results.

Clinical Topics: COVID-19 Hub, Prevention

Keywords: Coronavirus, Cost-Benefit Analysis, COVID-19, Diagnostic Tests, Routine, Nasopharynx, Primary Prevention, Saliva, severe acute respiratory syndrome coronavirus 2

< Back to Listings