• Original Reporting
  • Sources Cited
Original Reporting This article contains new, firsthand information uncovered by its reporter(s). This includes directly interviewing sources and research / analysis of primary source documents.
Sources Cited As a news piece, this article cites verifiable, third-party sources which have all been thoroughly fact-checked and deemed credible by the Newsroom in accordance with the Civil Constitution.
Yale School of Medicine professor Derek Toomre holds up a phone during a video call showing his u-Smell-it app, which links up with a scratch-and-sniff card to create a simple olfactory test. Toomre hopes the test can be used to screen for cases of COVID-19. (Screenshot)

It may be cliché, in this time of heightened stress and frenzy, to remind you to stop and smell the roses. But seriously: Stop. And. Smell. The. Roses. You might find out you have COVID-19.

Anosmia, the loss of your sense of smell, is by now a well-known symptom of a coronavirus infection. Studies have estimated its prevalence among people with COVID-19 at as much as 80%, which would make it as common of a symptom as cough or fever among symptomatic patients.


The latest from the coronavirus outbreak in Colorado:

  • MAP: Cases and deaths in Colorado.
  • TESTINGHere’s where to find a community testing site. The state is now encouraging anyone with symptoms to get tested.
  • VACCINE HOTLINE: Get up-to-date information.


But unlike those other two symptoms, anosmia also occurs at high rates in people with coronavirus infections that are otherwise asymptomatic. And it’s also an unusual symptom — there aren’t a lot of other illnesses that cause it. So that makes it more predictive than cough or fever, both of which can be caused by many other kinds of other illnesses.

For all these reasons, two researchers at the University of Colorado and a Yale School of Medicine professor who grew up in Boulder think they’ve sniffed out a better way to conduct mass testing for coronavirus infections.

Forget the nose swabs, the temperature guns, the overwhelmed testing labs. What if the solution to conducting a comprehensive surveillance program at places like offices or college campuses was as simple as a stack of 50-cent scratch-and-sniff cards and a phone app?

University of Colorado professor Daniel Larremore (Provided by the University of Colorado)

“If you’ve lost your sense of smell unexpectedly, that should be a signal to make sure the mask is on tight,” said Daniel Larremore, a professor at CU who is one of the three researchers.

In a paper posted earlier this month on the preprint server medRxiv, the researchers use a mathematical model to argue that conducting sniff tests three times a week would be as effective in curtailing virus spread as conducting weekly PCR tests. The server has become a common place during the pandemic for scientists to post hot-off-the-workbench studies before they have gone through peer review.

It’s not so much that the sniff tests are super-accurate. PCR tests are still considered the gold standard for accuracy in detecting coronavirus infections.

But, when it comes to mass testing, frequency is just as important as accuracy. And PCR tests are cumbersome to administer and analyze. They’re also costly — something that can add up quickly when organizations are trying to conduct testing on a large scale.

Thus, a cheaper and simpler, although less accurate, test that can be conducted more frequently can be just as valuable. 

“Your lack in accuracy is more than made up in frequency,” said Derek Toomre, a professor of cell biology at the Yale School of Medicine who was an author on the study, along with Larremore and CU professor Roy Parker.

And that’s how the u-Smell-it test (slogan: “Do you smell it?”) was born.

A promotional shot for the u-Smell-it scratch-and-sniff olfactory test. Its developers hope the test can be used to screen for cases of COVID-19. (Handout)

The test is part science and part elementary school arts and crafts. It involves a card with five unmarked scratch-and-sniff spots on it as well as a QR code. Test-takers use a smartphone app to scan the QR code. Then they scratch the spots and answer multiple-choice questions in the app about what they smelled — mint, vanilla, strawberry or other common scents. There’s also an option to say you smelled nothing.

The whole process takes about 45 seconds, at the end of which the app will tell you if you’ve passed or failed.

Toomre, who grew up in Boulder and went to CU for undergrad, said a dedicated, blind sniff test for coronavirus has significant advantages over just telling people to mind their nose. The first is that people who are infected often don’t realize they’ve lost their sense of smell — other studies have found around 45% to 50% of people infected with coronavirus report a loss of smell when surveyed, but that number rises to around 80% when people are actually administered a smell test, Larremore said.

And Toomre said it’s also important for the test to be blind to avoid tricks of the mind. If, for instance, you are used to sticking your nose down into your cup of coffee every morning, your brain might continue smelling that coffee even if your nose can’t. So Toomre said, in order for the test to work best, people can’t know in advance what they are supposed to be smelling.

Toomre has asked the federal Food and Drug Administration to approve the test, allowing it to be used for coronavirus surveillance testing. It’s not meant to be a diagnostic test, Toomre said. People who fail the sniff test should follow up with a PCR test.

But it could be valuable to workplaces or college campuses looking to conduct mass testing programs, and it could also be useful for people to have around the house, especially if they think they might be coming down with other symptoms and want to know if they should go get a PCR test.

This idea, though quirky, is not exclusive to Toomre, Larremore and Parker. Researchers at Penn State University have also explored the idea.

Toomre said he’s talking with manufacturers, with the hope of producing hundreds of millions of cards that can be sold for 50 cents or less, assuming it receives federal approval. He said he also plans to donate large numbers of tests to nonprofits.

It’s a long way from his pre-pandemic work, which focused on cellular imaging.

“I am comfortable being uncomfortable,” Toomre said, “which I definitely am in this area.”

John Ingold

John Ingold is a co-founder of The Colorado Sun and a reporter currently specializing in health care coverage. Born and raised in Colorado Springs, John spent 18 years working at The Denver Post. Prior to that, he held internships at the Rocky Ford Daily Gazette, the Colorado Springs...