Is a Coronavirus Breathalyzer Test Possible?

The only thing as frustrating as waiting to find out if you’ve tested positive for the coronavirus, a process that can take days, might be enduring the test in the first place. The most common COVID-19 test involves a long swab stuck up your nostril to collect secretions from the back of your nose and throat—not exactly a pleasant sensation. But researchers are breaking ground on easier methods. One might be like testing the breath for booze consumption.

“It’s the idea that you have COVID breath,” Makeda Robinson, MD, PhD, an infectious disease fellow at Stanford University, tells Verywell. “It’s been shown before that viruses, and cells that are infected with viruses, can emit these volatile organic compounds or VOCs into your breath.”

Researchers have developed and conducted a small early study on a breathalyzer-like device to rapidly detect the presence of SARS-CoV-2, the virus that causes COVID-19. The findings were recently published in ACS Nano, a publication of the American Chemical Society.

How It Works

The device has a sensor made of nanomaterials linked to VOC-sensitive molecules. When VOCs from the breath interact with the sensor, the electrical resistance changes, creating a pattern. The researchers used machine learning to train the sensor to recognize key differences in exhaled breath and identify a possible COVID-19 signature based on the electrical resistance signals. 

To train the sensor to recognize the possible COVID-19 signature, researchers enlisted three groups: 49 patients with confirmed COVID-19, 58 healthy individuals, and 33 patients with a non-COVID lung infection. Each participant blew into the prototype for a few seconds. Next, the researchers needed to determine if the prototype’s potential COVID-signature was accurate, so they tested the device on a smaller number of the 140 participants.

Is It Accurate?

In differentiating COVID-19 cases from the control group of healthy individuals, the device was 76% accurate. The sensor distinguished between individuals with COVID-19 and those diagnosed with a different type of lung infection with 95% accuracy, and between recovered and sick COVID-19 patients with 88% accuracy.

“It’s all squarely on the left side of the development curve,” Farley Cleghorn, MD, MPH, an epidemiologist and global head of health at Palladium, tells Verywell. “It's the first report. And while it shows some promising data, even the authors themselves say we need a larger study.”

The test was least accurate in determining between COVID-19 infection and healthy subjects. The study recorded a number of false positives, which can lead to an increase in healthcare use and unnecessarily contribute to overburdening, Robinson says. But false positives might be less of a concern than missing people who are infected.

“If somebody unknowingly has [COVID-19] and they get on a plane or they go back to work, now you’re exposing a lot of people to it,” Robinson says.

The Test May Help Gauge Improvement

Robinson says she is particularly intrigued by the fact the sensor was able to differentiate, to a degree, between patients who were sick with COVID-19 and those who had recovered.

“It could possibly tell us more about how people are improving or not improving over time,” she says. “It’s not definite that it could do this, but I think the possibility is there.”

A limitation of the initial research is that it doesn’t account for many factors like age, gender, ethnicity, comorbidities, and even where the test is administered. “This is such a small study that was performed in Wuhan, China—so in a more sort of homogenous patient population,” Robinson says. “We don't know if the COVID-19 breath of that area will be similar to others.”

But the sensor may be able to be trained to account for those factors. "I think that's really the promise of machine learning,” Robinson says. “We can keep evolving. This test can actually keep getting better over time."

A Welcome Alternative to Swabs

There's still a need for better testing methods for SARS-CoV-2, especially options that are less invasive than the dreaded nasopharyngeal swab.

“It’s very unpleasant,” Rebecca Wurtz, MD, MPH, associate professor in the School of Public Health at the University of Minnesota, tells Verywell. “If it’s really obtained properly, you’re supposed to push the swab all the way into the back and then swirl it five times.” 

Wurtz says she hasn’t seen any studies on this, but she suspects user error in collection, especially in certain drive-up testing sites where you actually administer the swab yourself. “I think a lot of lab techs and nurses and other people who obtain specimens even in a controlled setting are afraid to push that swab all the way back and to cause such discomfort,” she says.

The U.S. Food and Drug Administration (FDA) has granted emergency use authorization (EAU) to 230 COVID-19 tests, including 186 molecular, 40 antibody, and 4 antigen tests. The breath test in development doesn't fall into any of these categories.

“There’s been sort of this second wave,” Robinson says of testing innovation. “We’ve gotten more information that different specimen types can actually have similar accuracies. And now we’ve kind of moved into asking ourselves, ‘Is this the best and only way?’” 

Sample collection is only half the battle. Although new and innovative lab methods are in the works, the gold standard for this pandemic has been reverse-transcriptase polymerase chain reaction (RT-qPCR) testing. PCR is a complex and time-consuming process involving chemicals and machinery in a lab, according to the FDA. Before testing begins, samples have to be shipped to an appropriate lab where they often wait in a queue. 

Cleghorn says labs are simply backed up. He went to a free COVID-19 testing site about a month ago, he tells Verywell. The turnaround time for results at that site was about 10 days. In that time, he says people could be “missing the window for doing all the right epidemiological control measures that are required for when someone is either positive for coronavirus or has been exposed.”

The long turnaround times for results are concerning to experts, especially while test-positive rates are still high in many places in the U.S. and with the cold and flu season about to start. “We recommend that positivity rate by jurisdiction be less than 5%,” Cleghorn explains, citing the COVID-19 benchmark set forth by the World Health Organization in May. “We have about 30 states in the U.S. that have not achieved that.” 

These statistics, Cleghorn adds, “have led to a much greater appetite for other tests. And that’s why there’s been huge, very fertile efforts to develop antibody tests, antigen tests, non-PCR tests—anything that will ease some of the burden we are experiencing.”

Use Cases for Breath Tests 

So will we all be tested for “COVID breath” as we go about our days in the near future? “The test itself is very similar to a breathalyzer for an alcohol test,” Robinson says of the prototype mentioned in the study. “And alcohol or ethanol is a VOC.”

With its ease of use, a breath test could be a useful tool for point-of-entry screening, Robinson adds. Right now, many places like schools, clinics, and offices perform temperature checks and ask individuals about symptoms. “I think there’s a real niche for a diagnostic test like this because it requires so little additional personnel and the turnaround time is so fast,” she says. “We’ve seen with the reopening of schools that there’s a lot of hit or miss there.” 

Hypothetically, a breathalyzer could help school administrators screen students more accurately and efficiently. And such a test could be used at movie theaters or sporting events, as well. “The sensitivity and specificity would have to be improved,” Wurtz says, “but the sensitivity and the specificity of the tests we have right now is not too good, and the time and the delay is significant.” 

The breathalyzer-like prototype uses gold nanoparticles on its sensor, and for Robinson, that raises a few questions. “How expensive are these tests?” she says. “How often do these sensors have to be replaced? Can we scale a test like this up to very, very high level? I have a question about whether or not this is a sustainable solution.”

Wurtz agrees that it’s too early to say if we’ll be taking COVID-19 breathalyzers before we go to school or work or a concert. “I don’t get too excited about these things,” she says. “But as an area of exploration, I think it’s a promising one.”

The information in this article is current as of the date listed, which means newer information may be available when you read this. For the most recent updates on COVID-19, visit our coronavirus news page.