We often see them in airports, sniffing around for drugs and other questionable substances. Now, it seems, dogs are being trained to use their powerful sniffers (aka snouts) to detect the coronavirus.
How do they do it? First, let’s look at the dog’s olfactory advantage for this work. Humans have a mere six million smell receptors; dogs have as many as 300 million.
Dogs trained in scent detection can discern low levels of what are called volatile organic compounds (VOCs) that are found in human blood, breath, urine, and saliva and have been associated with a number of diseases. I hope to write more about these fabulous pups and their detection of other diseases in the near future.
As to our urgent international need, here’s a sampling of some ongoing studies:
In the UK
They’re actually engaged right now in what might be considered the first step toward an airport version of a clinical trial.
There are six dogs, aptly called “The Super Six,” a combo of labrador retrievers and cocker spaniels. The premise is that the dogs will be able to detect the scent of the disease on asymptomatic travelers.
So our future may include appearing at the airport, suitcases ready, passports in hand, and then a cold little nose says, in effect, “No trip for you, buddy! You’re outta here!”
But as a dog worshipper, I can’t think of a happier way to help bring a worldwide pandemic under control.
This effort is backed by the UK government, which has donated funds to a research team from the London School of Hygiene and Tropical Medicine, in collaboration with Durham University and a charity named Medical Detection Dogs.
And the head of the research team, Professor James Logan, who also heads the department of disease control at the London School of Hygiene and Tropical Medicine, expresses optimism.
He observed in an interview with cnn.com that this work evolved from earlier research findings that people infected with malaria have a specific body odor—and “dogs can be trained to detect that with very high accuracy.”
The training involves dogs’ sniffing face masks and/or nylon socks worn by both individuals who have tested positive for COVID-19 and those who haven’t.
Why nylon socks? Professor Steve Lindsay, a public health entomologist at the university, acknowledges that it’s a “bit strange,” but their experiences have shown them it’s “a really good way of collecting odors from people and it’s such an easy way to do it.”
In the US (Two Studies) [NOTE: SEE CLARIFICATION BELOW]
Similar research is being done at the University of Pennsylvania’s School of Veterinary Medicine (Penn Vet). Same premise: uncover asymptomatic patients, who are most likely to spread the disease. Also: focus on screening in specifically challenging environments for testing, such as hospitals or businesses.
In the Penn Vet study, which began with eight dogs, the dogs were first given saliva and urine samples from COVID-19 positive patients in a lab-oratory (spelled out and hyphenated to avoid confusion with this post’s stars!). Then they were given negative samples.
The researchers plan to begin testing the trained dogs with live humans in July. They’ll test both sensitivity—the ability to correctly identify those who have the disease (true positive rate)—and specificity—the ability to correctly identify those who don’t have the disease (true negative rate).
Cynthia Otto, DVM, PhD, professor of Working Dog Sciences and Sports Medicine and director of the Penn Vet Working Dog Center, is leading a multidisciplinary group across the University.
According to Otto:
“The potential impact of these dogs and their capacity to detect COVID-19 could be substantial. This study will harness the dog’s extraordinary ability to support the nation’s COVID-19 surveillance systems, with the goal of reducing community spread.”
Another study, at Cold Spring Harbor Laboratory in New York State, bears the names of numerous authors. Involving 198 samples of axillary (armpit) sweat collected from various hospitals, the study was conducted with 18 dogs on three sites. It took the dogs between one and four hours to learn to recognize the odor and then four to ten hours to detect positive samples.
In a subset to demonstrate proof-of-concept, the researchers focused on eight dogs that had previously been trained to detect explosives or colon cancer and had now expanded their doggie resumes with this new specialty.
Their task was to pick out the positive sample from among negative or mock (made up) samples. After 368 trials, here are their percentages: four dogs scored 100%; one achieved 83%; another 84%; another 90%, and the eighth dog 94%.
All those percentages were deemed significantly different from what would occur by chance.
“We conclude that there is a very high evidence that the armpits sweat odor of COVID-19+ persons is different, and that dogs can detect a person infected by the SARS-CoV-2 virus.”
Perhaps, theorizes Annie the English major, that also means that armpit sweat can be added to the bodily substances cited earlier that contain volatile organic compounds.
And in Finland
This article’s title appeals to my weakness for bad puns: “The Finnish COVID dogs’ nose knows!”
At the University of Helsinki, researchers from the veterinary and human medicine faculties are working together. The first dogs have successfully differentiated between the urine samples of patients that have COVID-19 and those that don’t.
The researcher and DogRisk group leader, Anna Hielm-Bjorkman, observed:
“We have solid experience in training disease related scent detection dogs. It was fantastic to see how fast the dogs took to the new smell.”
The preliminary tests have demonstrated that the dogs learned fast and worked fast, outperforming the COVID-19 tests based on molecular approaches.
But now comes the big step, prior to moving the scent detection into practice. They’ll begin a randomized double-blinded setting, “introducing them to a larger number of patient samples that are either positive or negative.”
And they’ll throw in a curve ball: some of the negative samples will have other respiratory diseases.
They foresee the many beneficial possibilities, which include identifying infected individuals in nursing or retirement homes, and screening health care workers to discern those who are actually ill, rather than just having been exposed, thereby avoiding unnecessary quarantines.
And, to bring us back to where we began, they’re also looking toward screening at airport checkpoints and other border points.
These are, as far as I can tell, all preliminary studies, with the Finnish study moving closest to peer review status, and the Penn Vet study ready to take a big leap forward in just a few weeks.
But with the dogged determination of researchers, trainers, and humans’ best friends, it certainly looks as though we may soon see a warm and fuzzy side of the successful efforts to contain this terrible pandemic.
Though most of these tests will be processed in lab-oratories, if you were in an airport and given the choice, which would you prefer: a large swab inserted into your nostril, or a tail-wagging canine circling around you once or twice (no petting allowed!)?
Warmest thanks and profound love to my daughter the professional dog trainer, par excellence, who suggested the idea and provided me with the articles that formed the basis of this post.
CLARIFICATION: The study I attributed above to the Cold Spring Harbor Laboratory in the US was based on an abstract bearing its logo. I subsequently learned that the study is, in fact, being conducted in Paris, at the Alfort School of Veterinary Medicine; Ajaccio, France, at the South Corsica Fire and Emergency Dept, and Beirout (sic), Lebanon, at the French-Lebanese University Saint Joseph.