Identifying Illness Through Scent
It’s been known for some time that rats and other animals can detect illness in others of their species based on scent. Rats will actively avoid sick packmates shortly after they fall ill, when there are few visible symptoms. Most people might believe that humans don’t notice sick friends quite so quickly and certainly not based on their scent. But is that belief really true?
It’s easy to identify someone with an illness if they show physical symptoms such as fever, sneezing, or exhaustion. It’s another matter to notice that person has just contracted a disease.
One way researchers have investigated an organism’s early response to disease is to inject it with lipopolysaccharide (LPS), an often-toxic substance found in many bacteria. Rats, for example, have an immediate immune response even when the dose isn’t toxic. Other rats can recognize “infected” rats and will stay away.
A team led by Mats Olsson wanted to know if humans have similar illness-detecting powers. The researchers found eight volunteers willing to be injected with small doses of LPS that would activate an immune response but would not be dangerous. The volunteers wore tight T-shirts, the better to capture the body odors released in their sweat.
Samples were taken over the next four hours, then put into spray bottles. In a separate session they were injected with a non-toxic saline solution — a placebo — and additional odor samples were taken as a control.
Next, 40 testers were recruited to smell both infected and uninfected samples from each of the earlier volunteers. Each sample was rated for intensity, pleasantness, and “health” on a scale ranging from 7 (total) to –7 (none). So a score of 7 on pleasantness would mean it was the most pleasant possible smell, a score of 0 would be neutral, and –7 would be the most unpleasant possible. Here are the results:
As you can see, the testers rated the odor from the LPS-infected volunteers as more intense, less pleasant, and less healthy than the odor from the same volunteers when they were not infected. All the results are significant, but the effect was much stronger for the intensity / pleasantness ratings than the health ratings, which are very close to neutral (remember, the scale goes to 7 in both directions).
One possible explanation is that the LPS simply caused the volunteers to sweat more, which might result in a stronger, more unpleasant odor. In fact, the reverse was true: The LPS-exposed volunteers sweated less than they did when given a placebo.
What’s particularly fascinating about this result is the fact that observers were able to detect these odor differences within just a few hours of exposure to the LPS. The human immune response had begun, but there was little additional evidence of illness — and in any case, the testers didn’t even see the people whose odors they were sampling.
What remains to be determined is how people respond to this information. Perhaps we use the odor of a sick person to learn if someone in our family or community needs help. Or perhaps, as in other animals, it is simply a warning to stay away.
What Olsson’s team has shown is that in many cases, there really is a scent of disease, readily detectable by humans, which takes effect within a few hours after exposure.
Olsson M.J., B. A. Kimball, A. R. Gordon, B. Karshikoff, N. Hosseini, K. Sorjonen, C. Olgart Hoglund, C. Solares, A. Soop, J. Axelsson & M. Lekander (2014). The Scent of Disease: Human Body Odor Contains an Early Chemosensory Cue of Sickness, Psychological Science, 25 (3) 817-823. DOI: http://dx.doi.org/10.1177/0956797613515681
Munger, D. (2014). Identifying Illness Through Scent. Psych Central. Retrieved on April 25, 2015, from http://psychcentral.com/blog/archives/2014/07/11/identifying-illness-through-scent/