COVID-19 Detection Dogs

By Scott Weese on July 28, 2020

A colleague asked me about scent detection dogs the other day. My response was that I hadn’t heard much after all the initial buzz, which might suggest things weren’t going well.

However, as opposed to the horrible pre-print I wrote about earlier, a paper in BMC Infectious Diseases (Jendry et al) provides some more robust and interesting information. It’s a pilot study, meaning it’s small, preliminary and underpowered, but it shows potential. Whether that’s ‘potential for dogs to be able to detect SARS-CoV-2 under certain circumstances‘ or ‘potential for dogs to be an effective detection tool’ isn’t clear. That’s the big question.

Here’s a breakdown and some commentary.

They collected saliva samples and respiratory secretions from hospitalized COVID-19 patients, and healthy people that were PCR negative.

This may not be ideal, depending on the goal. 


My vision is using these dogs in the community to detect infectious people in high risk situations (e.g entrance to transit, buildings, schools). 

If that’s the case, people that are hospitalized with severe COVID-19 might not be the ideal population. A dog isn’t going to replace a PCR machine in the hospital. There still might be some benefit in a hospital but the practical aspects of collecting a sample from a patient, taking it to the dog as a quick screen, then taking it for proper testing, doesn’t seem overly practical.

They didn’t test samples for other human coronaviruses (those that cause the common cold). It’s a potential limitation but I don’t think it’s a big deal.

They also don’t explain where they got their negative samples (were dogs detecting SARS-CoV-2 or something unique to the positive sample population, like being from a hospital). Clear description of the study populations is critical and it’s somewhat lacking here.

Because of the potential susceptibility of dogs to this virus, samples were inactivated prior to exposing them to the dog.

That’s a reasonable step but raises more issues of practicality and how dogs would be used (exposed directly to a high risk patient specimen vs detection of someone walking by).

Eight dogs were trained using standard methods. They had a 2 week habituation process for the training system, then had 5 days of training until their rate of detection was greater than what would be expected by chance alone. They then started the study

Small sample size but reasonable for a proof-of-principle study.

The ability of dogs to detect positive samples increased over time. 

There was some variation between dogs but all were pretty good. The overall sensitivity (percentage of positive samples that were detected) was 83%, ranging from 70-95%. The specificity (percentage of negative detections that were from truly negative samples) was 96%, ranging from 92-99%.

For a screening test, we’d actually want the reverse….high sensitivity at the expense of specificity. That would mean you’d detect most positives.

 A lower specificity is ok if there’s an ability to follow up a screening test with a more specific test, and the implications of an initial false positive aren’t high, that’s acceptable. 

If this was a situation where someone that was detected as positive by a dog was sent home for 14 days, a high false positive rate would be a problem. If it just means they’re pulled aside and tested using a lab test, that’s not as big of a deal. It’s a hassle but maybe not a deal breaker.

A low sensitivity and high specificity means you run into fewer hassles with false positives but miss more positive people. The fact that 17% of samples from people hospitalized with active COVID-19 were called negative is a concern for picking up positives in less severely affected people and detection from less voluminous and close samples (e.g. detection directly from someone walking by vs detection from an aliquot of secretions from the lower airways).

I’d file this in the ‘interesting but preliminary’ folder. Anything that can help identify infectious people is useful. If dogs could be part of that, great. They’d have to be able to do it from a distance (since a handler and dog getting very close to large numbers of people might cause more problems than they fix).

In my perfect world….

A dog would be parked at the entrance of a school, office building, subway….

It would be able to detect infected people from a short distance.

It would signal when it detected a positive person.

That person would (discretely) be pulled aside for testing, which would (again, in my perfect world) be done quickly, right there.

If positive, the person would be told right away and sent home. If negative, they’re good to go (maybe wondering why they smell like a coronavirus).

Overall, it’s an interesting and useful preliminary study. 


The key is to follow up preliminary studies with more detailed work, something that often doesn’t get done. This shows there’s potential, and more rigorous work is needed (although I suspect headline writers will jump on this and over-interpret things).

 It’s interesting and another remarkable example of what a dog’s nose can do. How/if this will be applied functionally is still very much up in the air.  I’ll be a bit surprised if this ever becomes a common/useful tool but I’d love to be wrong about that.

......