As the COVID-19 pandemic created an urgent need for rapid, accurate, and non-invasive testing, researchers explored innovative diagnostic tools to complement conventional methods. Dogs’ extraordinary sense of smell, capable of detecting volatile organic compounds (VOCs) associated with disease, emerged as a promising avenue for mass screening.
In this pivotal study published in PLoS ONE, Dominique Grandjean and colleagues evaluated the ability of trained search and rescue dogs to identify SARS-CoV-2-positive individuals using sweat samples. The study involved seven dogs trained to detect COVID-19 odor signatures in 218 axillary sweat samples—62 from infected individuals and 156 from uninfected participants—under a randomized, double-blind testing protocol.
Each dog’s diagnostic performance was assessed by calculating sensitivity (the ability to correctly identify positive cases) and specificity (the ability to correctly identify negatives). Sensitivity ranged from 87% to 94%, while specificity ranged from 78% to 92%. Four of the dogs exceeded 90% in both measures. These results approached the accuracy of standard laboratory-based diagnostic tests such as PCR assays.
The study also analyzed predictive values across different infection probabilities, showing that dogs maintained negative predictive values of ≥98% when infection probability was low (10%), and ≥88% even when infection probability rose to 50%. These figures demonstrate the dogs’ exceptional reliability in identifying true negatives, an essential feature for mass public health screening.
Grandjean et al. emphasized that while dogs cannot fully replace laboratory diagnostics, they can play a vital role in large-scale, rapid pre-screening—particularly in high-traffic environments like airports, hospitals, and event venues. The authors also called for standardized training protocols and deployment infrastructure to ensure reproducibility and consistency across detection teams.
Ultimately, the research reinforces the immense potential of canine olfactory detection in medical diagnostics, offering a humane, cost-effective, and rapid screening tool for infectious diseases and beyond.
Source: Grandjean, D., Gallet, C., Julien, C., Sarkis, R., Muzzin, Q., Roger, V., Roisse, D., Levert, C., Breton, E., Galtat, A., Forget, A., Charreaudeau, S., Gasmi, F., Jean-Baptiste, C., Petitjean, S., Hamon, K., Duquesne, J., Coudert, C., & Desquilbet, L. (2022). Identifying SARS-COV-2 infected patients through canine olfactive detection on axillary sweat samples; study of observed sensitivities and specificities within a group of trained dogs. PLoS ONE, 17. Published February 14, 2022.
