Using a Smell Test to Diagnose Alzheimer’s Disease

Image Credit: Sally Plank

Using a Smell Test to Diagnose Alzheimer’s Disease

Alzheimer’s disease (AD) pathology appears to start in the part of the brain that handles smell before subsequently spreading to additional brain regions and then, ultimately, taking over much of the rest of the brain. This led some to speculate that Alzheimer’s disease may begin in the nose. Perhaps there’s some environmental agent that might enter the brain through some portal in the nostrils?

This is the so-called olfactory vector hypothesis. The anatomy of the nose is well suited for the transfer of things directly into the brain, since the olfactory nerves that stick out into the nose project directly into the brain, bypassing the blood-brain barrier. The nose was actually a major infection route for the polio virus. Public health officials started cauterizing the nasal passages of schoolchildren by spraying caustic chemicals up their noses in an effort to prevent the disease.

The concern is if people breathe in some ionized metals like aluminum dust, for example, it could be transported into the brain through these olfactory nerves at a rate of about 2 millimeters an hour, which is practically 2 inches a day. Doubt has been cast on this theory, however, by a case report of a woman born with a birth defect in which she had no smell nerves yet still developed Alzheimer’s-like pathology. And so, to date, all the supporting evidence is really just circumstantial. It is clear, though, that changes in the sense of smell are among the first clinical signs of Alzheimer’s, occurring during the preclinical phase—that is, before there’s any noticeable cognitive decline. Could we use these changes to predict or diagnose the disease?

For years, researchers have been trying to find markers of brain illness hidden in people’s ability to smell using all sorts of fancy gadgets. For example, functional MRI scans can detect differences in brain activation in response to an odor. In my video, Peanut Butter Smell Test for Alzheimer’s, you can see the responses to lavender. You’ll see a representation of a normal brain’s responses to the odor versus an Alzheimer’s brain. This unequivocally demonstrates that we can pick up changes in smell function due to Alzheimer’s. But, do we really need a million-dollar machine?

An ingenious group of researchers at the University of Florida discovered all we may need is some peanut butter and a ruler.

Considering that the left side of the brain primarily processes what we smell through our left nostril and the right side of our brain covers the right nostril, and understanding that Alzheimer’s strikes the left side more than the right, what if you performed the following experiment: Close your eyes and mouth, breathe normally through the nose, then close one nostril, and hold a foot-long ruler out from the open nostril. Once your eyes, mouth, and one nostril are closed, open a container of peanut butter at the bottom of the ruler (one foot away from your open nostril). Move the peanut butter closer by 1 centimeter upon each exhale until you can detect the odor. Then, repeat the whole procedure again using your other nostril.

This is exactly what the University of Florida researchers did with their subjects. What did they find? The normal elderly control subjects in the study smelled the peanut butter as soon as it came within an average of 18 centimeters (about 7 inches) from either nostril. It was about the same, roughly 7 inches, in the right nostrils of Alzheimer’s patients. But in their left nostrils, it was a mere 2 inches! The peanut butter had to be only 2 inches away before the Alzheimer’s patients could detect it through their left nostrils. This happened every single time. Indeed, the researchers found that a “left nostril impairment of odor detection was present in all the patients with probable AD.” There was no left-right difference in the control group; they could smell the peanut butter when it was the same distance away from both their left and right nostrils. In the Alzheimer’s group, however, there was a 12-centimeter difference.

The disparity was so great that we may be able to set a cutoff value for the diagnosis of Alzheimer’s. The researchers reported that “[c]ompared to patients with other causes of dementia this nostril asymmetry of odor detection…was 100% sensitive and 100% specific for probable AD,” meaning no false positives and no false negatives. Compared to healthy people, it was 100% sensitive in picking up cases of probable Alzheimer’s and 92% specific. What exactly does that mean? In this study, if you had Alzheimer’s, there was a 100% chance of having that wide left-right discrepancy. But, if you did have that discrepancy, the chance of having Alzheimer’s was only 92%. This means there were some false positives.

The reason it’s only “probable” Alzheimer’s is because the only way we can really confirm someone has the disease is on autopsy. The current criteria for diagnosing Alzheimer’s require an extensive evaluation, combined with fancy positron emission tomography (PET) scans and spinal taps. All of these tests are expensive and hard to get, can be invasive, and can have potential complications. On top of that, they are neither highly sensitive nor specific. The left-right nostril / peanut butter odor detection test, however, was fast, simple, non-invasive, and inexpensive. They concluded that may make peanut butter an ideal instrument for the early detection of Alzheimer’s disease.

Does all this sound a bit too good to be true? It may be. A University of Pennsylvania research team was unable to replicate the results. Click here to read their paper. So, at this point, the data are mixed. I’ll do another post once more studies are published and we have a better handle on whether it’s useful or not.

Of course, it’s better to prevent Alzheimer’s in the first place. Check out these videos for more information.

In health,

Michael Greger, M.D.

PS: If you haven’t yet, you can subscribe to my free videos here and watch my live, year-in-review presentations:


Michael Greger M.D., FACLM

Michael Greger, M.D. FACLM, is a physician, New York Times bestselling author, and internationally recognized professional speaker on a number of important public health issues. Dr. Greger has lectured at the Conference on World Affairs, the National Institutes of Health, and the International Bird Flu Summit, testified before Congress, appeared on The Dr. Oz Show and The Colbert Report, and was invited as an expert witness in defense of Oprah Winfrey at the infamous "meat defamation" trial.

21 responses to “Using a Smell Test to Diagnose Alzheimer’s Disease

Comment Etiquette

On, you'll find a vibrant community of nutrition enthusiasts, health professionals, and many knowledgeable users seeking to discover the healthiest diet to eat for themselves and their families. As always, our goal is to foster conversations that are insightful, engaging, and most of all, helpful – from the nutrition beginners to the experts in our community.

To do this we need your help, so here are some basic guidelines to get you started.

The Short List

To help maintain and foster a welcoming atmosphere in our comments, please refrain from rude comments, name-calling, and responding to posts that break the rules (see our full Community Guidelines for more details). We will remove any posts in violation of our rules when we see it, which will, unfortunately, include any nicer comments that may have been made in response.

Be respectful and help out our staff and volunteer health supporters by actively not replying to comments that are breaking the rules. Instead, please flag or report them by submitting a ticket to our help desk. is made up of an incredible staff and many dedicated volunteers that work hard to ensure that the comments section runs smoothly and we spend a great deal of time reading comments from our community members.

Have a correction or suggestion for video or blog? Please contact us to let us know. Submitting a correction this way will result in a quicker fix than commenting on a thread with a suggestion or correction.

View the Full Community Guidelines

      1. I doubt it, because the inability to smell is due to the nerve damage from surgery, not from the nerves carrying toxic metals like aluminum from the air into the brain.

      2. E: I don’t think so. As I understood it, the lose of sense of smell in the left nostril was a symptom, not a cause.

  1. Thank you Dr. Gregor for this interesting report. My question: do you know what kind of peanut butter was used in the study? Brand? Homemade…that is just peanuts ground up? Was there salt in the mixture? I am wondering because the only kind of peanut butter we eat is a freshly ground mix of fresh unroasted, unsalted peanuts that we get at our local grocery. (I follow Fuhrman’s nutritarian food plan and do not roast nuts or salt them). Roasting and salting might affect the nose test.
    Finally, my mother had some form of dementia but undiagnosed back in the 80’s. Memory highly impacted and now I have some similar symptoms. Would like to try the peanut butter test but want it to be as close to the study as possible. Thank you.

    1. Kat: You need to read all the way to the bottom where Dr. G says, “Is it sound a bit too good to be true?” A 2nd study could not replicate the results.

    1. The sense of smell and taste are impacted when the person is deficient in zinc. I wonder if this is connected in any way.

  2. Snopes has an article which my friends all linked to when I posted this on Facebook. The Snopes article addresses David Avocado Wolfe pushing this nose test. Dr. Greger, please tell us you haven’t changed your middle name!

    I think this article hurts the mission of . I’ll now have lost all credibility within my alumni group to future posts, and people will associate the website with quack pseudo-science profiteers.

    1. Sorry, Josh, I don’t understand the problem. Dr. G wasn’t touting the peanut butter test as a way to diagnose Alzheimer’s disease. He merely shared one study that had positive results, followed by another that did the same test but couldn’t replicate results. So Dr. G’s conclusion was pretty much the same as Snopes – mixed. If your friends / colleagues don’t have the attention span to read Dr. G’s the whole article & just want to jump to their own conclusions, then that just makes them look silly.

      This is not a quack pseudo-science website, and there are no profiteers here.

  3. Dang Nutritionfacts is going the pseudoscience route, slowly but surely I knew it would happen and it might get worse.

    1. Dang, you should direct your ‘worries’ to Stamps et al at the University of Florida, who found and reported this result in 2013, not Snopes, Wolfe or Greger..
      Or to the late Ronald Fisher, who developed the F test used by these scientists to find a probability of this being a statistical fluke p<0.0001
      p<.05 is often taken as reason to publish.
      My concern, which didn't appear here though I seemed to have been first to comment, had to do with possible confounding effects of the experimental room's air currents; but apart from that, why not let the result stand without aspersions?
      Also, Wolfe's often being wrong doesn't entail his always being wrong; if that were so, he would be a wonderful resource.

    2. No it’s not, harambe, you just didn’t bother to read the whole article. You need to get to the part where Dr. G asks, “Is it too good to be true?”

      IMHO, it would be interesting to find out why the studies got such disparate results. Not that I’m particularly interested in hearing anyone’s opinion about it. But I would be interested in seeing some evidence based facts.

  4. Alzeimer disease is the most common cause of dementia .Anyhow the article is very well and every one need to read it.Visit our site for Drugs and Medicine knowledge

Leave a Reply

Your email address will not be published. Required fields are marked *

Pin It on Pinterest

Share This