The dogtor is in: Study finds dogs can sniff out COVID-19 with 96% accuracy

Poncho, a two-and-a-half year old yellow Labrador retriever, along with seven other Labs and one Belgian Malinois who had never done medical detection work before, were able to sniff out COVID-positive samples with remarkable accuracy. (Pat Nolan)

Who knew one of the keys to ending the global health crisis of the last year might come from the group who started a simultaneous crisis of their own by barking through every Zoom call?

A new study led by researchers from the University of Pennsylvania’s School of Veterinary Medicine found that mass COVID-19 screening could be solved by a quick sniff—a dognostic test, if you will.

In keeping with the illustrious tradition of canine heroism (to wit: Lassie, Scooby-Doo, Air Bud), a group of nine specially trained dogs proved themselves to be not just man’s best friend but also a physician’s best friend by detecting positive samples of the coronavirus with a tail-wagging 96% accuracy rate.

RELATED: DoorDash to stock COVID-19 test kits for home delivery

Eight Labrador retrievers and one Belgian Malinois who had never done medical detection work were recruited for the study. They ranged in age from one-and-a-half to a little over six years and included four very good boys and five very good girls.

The sniff squad was trained first in general scent detection. Once mastered, they moved on to learning to distinguish between urine samples from COVID-positive and -negative patients and to respond only to the positive samples, which had been inactivated so they were safe to sniff.

After three weeks at the training facility—hereafter referred to only as the Lab Lab—all nine dogs demonstrated a remarkable ability to detect positive samples, with an average accuracy rate of 96%, proving once and for all that dogs of any age can indeed be taught new tricks.

Unfortunately, it will likely be awhile before we see COVID-sniffing dogs patrolling public areas and venues. For one thing, the dogs in the study were repeatedly tripped up by a sample from a patient who had tested negative for COVID but had previously had the infection.

For another, the cost of goods required for scaling up sniff testing could be rather hefty, when kibble, treats, the occasional squeaky toy and other consumables are factored in.

There's also the fact that the COVID testing market is both highly competitive and rapidly narrowing. Top dogs like Abbott, Thermo Fisher and Hologic continue to rapidly increase their testing output to meet demand, through acquisitions, new lab tests and the introduction of coronavirus-detecting machines.

And while the Centers for Disease Control and Prevention reports that nearly 400 million tests have been administered in the U.S. to date, analysts at Jefferies predict quarterly drops in the COVID diagnostic market of up to 30% in that demand going forward, before closing out this year with a peak of about $8.5 billion. In the future, testing demand levels may resemble the seasonal flu, they said, with averages of about $150 million to $250 million annually. 

Plus, unlike other coronavirus diagnostics currently on the market, the sniff method requires the occasional potty break or jaunt in the park.

RELATED: With green lights for routine, at-home screening, FDA sets up over-the-counter tests for COVID-19's endgame

Still, like a dog with a bone, the Penn scientists are forging ahead with a related and much larger study in which four-legged investigators are being taught to distinguish between the odors left on T-shirts by people who test positive or negative or have been vaccinated.

“We are collecting many more samples in that study—hundreds or more—than we did in this first one, and are hopeful that will get the dogs closer to what they might encounter in a community setting,” senior author Cynthia Otto, director of the vet school’s Working Dog Center, said.

This isn’t the research world’s first foray into dognostics. Past studies have demonstrated that, thanks to their highly sensitive noses, dogs can also sniff out various forms of cancer from an array of blood samples. In fact, their olfactory abilities are now being used to train machine learning algorithms to detect cancer from urine samples with the same levels of accuracy.