Specially trained dogs, sniffing sweat samples, were able to detect COVID in nearly all cases, while rapid nasal tests only picked up on roughly 84% of cases, according to the study, published in the medical journal Plos One. 

Rapid tests won out, however, when it came to accurately identifying negative cases: Dogs correctly identified about 90% of negative cases, but rapid tests correctly identified nearly all.

Two of the false-positive cases identified by dogs had other coronaviruses: NL63 and OC43, both of which cause respiratory infections, the authors wrote.

The first-of-its-kind study showed “excellent” performance of trained canines when compared to rapid testing. Dogs showed “no real difference in sensitivity,” or the ability of a test to correctly identify patients with a disease. Canines are less invasive than other forms of COVID testing, the authors pointed out, and can also complete testing more quickly than a lab.

Specially trained dogs may be particularly useful in detecting COVID among special-needs patients, including those with Alzheimer’s, who find it difficult to endure nasal-swab testing, and in mass settings like airports, public events, and schools, the authors wrote.