Anyone who has lost their sense of smell during a bad cold knows how flat food tastes without it. For most people it comes back. For some older adults, a lasting loss of smell may be an early sign that something is changing in the brain.
A 2017 study from the University of Chicago found that older adults who struggled to identify five common scents were more likely to be diagnosed with dementia five years later. It got a lot of headlines, and some of them promised more than the research can deliver.
This guide explains how the smell test worked, what the researchers found, what the results can't tell us, and what to do if you or someone you love has noticed a change in their sense of smell.
How the Smell Test Worked
The study followed nearly 3,000 adults aged 57 to 85. All of them lived at home and had normal thinking and memory at the start, according to the study's published summary in the Journal of the American Geriatrics Society.
Each person was given five "Sniffin' Sticks," which look like felt-tip pens but are infused with a scent instead of ink. They smelled each one and picked the right answer from four choices. The five scents were peppermint, fish, orange, rose and leather.
Anyone who identified four or five correctly counted as having a normal sense of smell. Three or fewer meant some degree of smell loss, and about 22 percent of participants fell into that group. Five years later, the researchers checked who had been diagnosed with dementia.
What the Researchers Found
After accounting for other risk factors, people with smell loss had more than twice the odds of being diagnosed with dementia five years later. The published figure is an odds ratio of 2.13, with a 95 percent confidence interval of 1.32 to 3.43.
The risk also rose with the number of mistakes. The more scents a person got wrong, the greater their chance of a dementia diagnosis, rather than smell loss simply being present or absent.
The University of Chicago's summary gives a breakdown of the scores. About 78 percent of participants identified at least four scents. Around 14 percent got three right, 5 percent got two, 2 percent got one, and 1 percent couldn't identify any. According to that summary, almost everyone who couldn't name a single scent had been diagnosed with dementia five years later, and nearly 80 percent of those who got only one or two right had dementia too.
Those last groups were small, about 1 percent and 7 percent of participants, so those figures deserve more caution than the headline result across the whole sample.
What the Results Can and Can't Tell Us
It shows a link, not a cause. The study found that poor smell identification and later dementia diagnoses tend to go together. It doesn't show that losing your sense of smell causes dementia.
It isn't a diagnosis. A poor score means higher risk in a group of people. It can't tell you what will happen to one person. Smell loss also has other causes, including sinus problems and viral infections such as COVID-19.
The authors see it as one input. In their conclusion, the researchers describe the five-item test as an efficient, low-cost part of a physical exam that could give useful information when assessing dementia risk. They don't describe it as a stand-alone test.
Screening is debated. A commentary published in the same journal issue was titled "Screening Is Not Benign," a reminder that testing people who have no symptoms has costs as well as benefits, including worry and follow-up tests.
Newer work points the same way. A more recent 12-year population-based study reported that people with both cognitive and smell impairment had a higher short-term risk of progressing to dementia, and that smell loss may be a useful early marker on its own.
Why Would Smell and the Brain Be Linked?
Nobody has a complete answer yet. Lead researcher Jayant Pinto said the results show how closely smell is tied to brain function and health, and that researchers still need to understand the mechanisms behind the link. His co-author Martha McClintock suggested that smell, an evolutionarily ancient sense, may point to a mechanism that also underlies human cognition.
The same research group had reported in 2014 that loss of smell in older adults predicted death within five years, and that it did so better than a diagnosis of heart failure, cancer or lung disease. Smell may be a general marker of how well the body and brain are holding up, in addition to any specific link to dementia.
Why a Lost Sense of Smell Matters in Daily Life
Whatever the link to dementia, losing your sense of smell affects everyday life. Pinto has pointed out that smell influences nutrition and mental health.
People who can't smell may not know whether food has spoiled, may not detect smoke in a fire, and may not realise they need a shower. Loss of smell is also closely associated with depression, since food and other everyday pleasures become less enjoyable.
If someone you care for has lost their sense of smell, a few practical steps help:
Check that smoke alarms are working and test them regularly.
Label leftovers with dates and clear out the fridge each week.
Keep an eye on appetite and weight, since food that tastes of nothing is easy to skip.
Ask, gently, whether they've stopped cooking or eating the things they used to enjoy.
What to Do If You Notice a Change in Smell
A change in smell is not a dementia diagnosis, and it doesn't need to cause panic. It is worth acting on quietly.
Write down what you've noticed. When did it start? Did it follow a cold or infection? Is it getting worse?
Note any other changes. Memory lapses, confusion, mood shifts, withdrawing from things they enjoy, or changes in appetite all help a doctor see the bigger picture. Our guides on early versus mid-stage dementia and mild cognitive impairment explain what different stages look like.
Mention it to a doctor. Bring the notes and say the smell change out loud, alongside anything else that's worried you. A doctor can look for other causes, such as a sinus problem, and decide whether further checks make sense.
Keep tracking. A slow change is much easier to judge from written notes than from memory.
How Elli Cares Can Help You Track Changes
Small changes are easy to miss when you see someone every day, and easy to forget by the time you're sitting in the appointment. Elli Cares gives families a simple place to record what they notice. Daily check-ins and family observation notes can capture things like "said the soup tasted of nothing" or "has stopped cooking," and a downloadable report makes it easy to share the pattern with a doctor.
Elli doesn't test smell, diagnose dementia or predict risk. It helps families bring clearer information to the conversation. You can read more on our Seniors and Families page.
Frequently Asked Questions
Can a smell test predict dementia?
In a University of Chicago study of nearly 3,000 older adults, people who identified three or fewer of five scents had more than twice the odds of a dementia diagnosis five years later. It shows higher risk in a group and can't predict what will happen to one person.
Which scents were used in the test?
The five scents were peppermint, fish, orange, rose and leather, presented in Sniffin' Sticks pens. Participants chose the right answer from four options for each one.
If I lose my sense of smell, will I get dementia?
Not necessarily. Smell loss has several causes, including sinus problems and viral infections, and a poor test result isn't a diagnosis. It is worth mentioning to a doctor alongside any memory or behaviour changes.
Can I test my sense of smell at home?
The study used a validated set of scent pens rather than household items, so home experiments won't give a reliable result. If you're worried about your sense of smell, ask a doctor.
What else can loss of smell affect?
Nutrition, mood and safety. People who can't smell may miss spoiled food or smoke, may eat less because food tastes bland, and are more likely to feel low.
