The Sense Nobody Checks
Every routine physical measures the same things: blood pressure, heart rate, cholesterol, weight, and reflexes if the doctor is thorough, but nobody has ever checked your nose.
Jayant Pinto, an otolaryngologist at the University of Chicago, wondered what a simple smell test might reveal about who was going to die. He gave 3,005 community-dwelling Americans aged 57 to 85 a five-item Sniffin' Sticks test: scratch-and-sniff pens loaded with peppermint, fish, orange, rose, and leather. Each person was classified as normosmic (normal smell), hyposmic (diminished), or anosmic (unable to smell). Then he waited five years. The results were unambiguous (Pinto et al., 2014).
The Results Were Not Subtle
Within five years, 39% of those who failed the smell test were dead, compared to 19% of those with moderate loss and just 10% of those with normal smell — and because the study was part of the National Social Life, Health and Aging Project, a nationally representative in-home survey of community-dwelling adults rather than a hospital cohort, these numbers generalize to the broader U.S. population of older Americans.
After adjusting for age, sex, race, education, socioeconomic status, overall health, smoking, alcohol use, diabetes, and body mass index, anosmics still had more than three times the odds of dying compared to normosmics (OR 3.37, 95% CI 2.04–5.57, p<0.001). The association was dose-dependent: each additional error on the five-item test increased mortality risk (p<0.001 trend test), meaning that even mild olfactory impairment carried measurable risk.
The comparison that made headlines was brutal: olfactory dysfunction predicted five-year mortality better than diagnoses of heart failure, cancer, or lung disease, and only severe liver damage outperformed the scratch-and-sniff test as a predictor of death.
Three Independent Replications Confirmed It
Devanand and colleagues at Columbia University tested 1,169 Medicare beneficiaries with a 40-item University of Pennsylvania Smell Identification Test, followed them for an average of four years, and found that mortality reached 45% in the lowest-scoring group versus 18% in the highest, with the association holding after adjusting for medical burden and dementia (Devanand et al., 2015).
In Sweden, Ekström and colleagues followed 1,774 adults for over a decade and found that smell loss predicted mortality regardless of whether participants later developed dementia, a result that eliminates the tidy explanation that olfactory decline is simply an early marker of Alzheimer's disease rather than an independent signal of physiological breakdown (Ekström et al., 2017).
Liu and colleagues pooled data from three large U.S. cohorts totaling 2,289 older adults and confirmed that poor olfaction was associated with increased 10-year mortality (HR 1.46, 95% CI 1.27–1.67), independent of demographics, cognition, and comorbidities (Liu et al., 2019).
An Original Calculation
According to the 2020 U.S. Census, approximately 54 million Americans are 65 or older. The National Institute on Deafness and Other Communication Disorders estimates olfactory dysfunction affects roughly 25% of adults over 65, which yields approximately 13.5 million people. The Pinto study found that 39% of anosmics and 19% of hyposmics died within five years, versus 10% of normosmics. If a $2 scratch-and-sniff test were administered during routine physicals to this population, it could flag millions of elevated-risk individuals that no current screening protocol captures, at a per-test cost lower than a single disposable blood pressure cuff sleeve.
Why the Nose Predicts the Body
The olfactory epithelium is one of the few places in the adult nervous system where neurons regenerate continuously throughout life. When that regeneration slows, it may signal a broader failure of the body's cellular repair machinery. The olfactory nerve is also the only cranial nerve directly exposed to the environment, making it a frontline sensor for cumulative toxic damage from pollution, pathogens, and airborne particulates. The nose is where aging shows first.
Chronic inflammation offers another pathway connecting the nose to the rest of the body. A 2020 review in Frontiers in Neuroscience synthesized the evidence and concluded that elevated interleukin-6 levels are associated with both olfactory dysfunction and frailty, and that inflammation likely mediates the relationship between smell loss and mortality. Controlling for cardiovascular disease, metabolic disorders, and frailty scores did not attenuate the association in any of the major cohort studies (Poncet et al., 2020).
The Strongest Case Against
The most forceful objection is that smell loss is a bystander: an early symptom of neurodegenerative diseases like Parkinson's and Alzheimer's, which kill by other mechanisms entirely. Olfactory decline precedes clinical onset of Parkinson's by years, and olfactory testing is already used in Parkinson's screening protocols at movement-disorder clinics worldwide. If anosmics die sooner, perhaps the neurodegenerative disease explains the mortality, not the smell loss itself.
This is a serious challenge, and partially right. But the Swedish replication by Ekström specifically tested this hypothesis by separating participants who developed dementia from those who didn't, and smell loss predicted mortality in both groups equally. The Columbia study by Devanand found the association weakened but remained statistically significant after controlling for dementia, and the Pinto team controlled for cognitive function in their original analysis and found no attenuation of the effect at all. Neurodegeneration is part of the story, but not the whole story, and the residual signal points to something more fundamental about how the body ages.
What We Didn't Prove
All four studies are observational, which means they establish that smell loss predicts death without proving it causes it, and no intervention trial has tested whether treating olfactory dysfunction or the conditions it signals reduces mortality.
The Pinto study used a five-item test, which is quick but crude, and misclassification is inevitable at the margins, particularly for hyposmics, where a single incorrect answer determines categorization. The odds ratio of 3.37 comes from a logistic regression with many adjustments, and no statistical model fully eliminates confounding in observational data.
Finally, the practical utility of the test remains unproven in a clinical setting. Identifying someone as anosmic raises a flag, but no clinical guideline specifies what to do next. Until intervention trials show that acting on olfactory screening changes outcomes, the test identifies risk without a validated pathway to reduce it.
The Bottom Line
Across four independent studies in three countries spanning more than 8,000 participants, the pattern holds: people who lose their sense of smell die sooner, independent of age, sex, cognition, chronic disease, and smoking. A test that costs less than a blood pressure cuff, takes under three minutes, and requires zero medical training catches a signal that the entire modern screening toolkit misses.
What You Can Do
Test yourself. The Brief Smell Identification Test (B-SIT) costs roughly $27 and is available without a prescription, offering twelve scratch-and-sniff panels with forced-choice answers that give you a validated score in under ten minutes. If your score falls below normal for your age and sex, bring it to your doctor.
Mention smell changes at your next physical. Most physicians never ask about olfactory function, so you have to raise it yourself. If you've noticed that food tastes blander, that you can't smell coffee brewing, or that your partner complains you've overloaded on perfume, say something. Self-reported smell loss is less reliable than objective testing, but it is a starting point.
Protect your nose. Smoking, chronic sinus infections, and environmental pollutant exposure all accelerate olfactory decline, and treating nasal inflammation, managing allergies, and reducing exposure to particulate matter won't guarantee you keep your sense of smell, but they remove known accelerants of a decline that turns out to matter far more than anyone assumed.