Davangere Devanand, MD: Odor Identification and Cognitive Testing Predict Dementia

Davangere Devanand, MD

Credit: Columbia University Department of Psychiatry

In an interview with The Educated Patient, Davangere P. Devanand, MD, professor of psychiatry and neurology and director of geriatric psychiatry at the Columbia University Irving Medical Center, discusses the results of a study that evaluated impaired odor identification and global cognition as simple alternatives to neuroimaging biomarkers to predict cognitive decline and dementia.

Patients were recruited from the Mayo Clinic Study of Aging (MCSA), a long-term study following more than 1,000 people in Rochester, Minnesota. The study included 647 participants, followed for an average of 8.1 years. At baseline, participants underwent several assessments: the 12-item Brief Smell Identification Test (BSIT) for odor identification, the modified Blessed Information Memory Concentration Test (BIMCT) for cognition, structural brain MRI, and PET imaging with 11C-Pittsburgh compound B (PiB) and fluorodeoxyglucose (FDG, in a subset).

Over the follow-up period, 102 participants showed cognitive decline and 34 developed dementia. Analysis showed that PiB PET imaging was a robust predictor of cognitive decline. However, impaired performance on both the BSIT and BIMCT, as well as MRI and FDG PET measures, were also significant predictors.

The findings indicate that combining smell and cognitive testing can provide a reliable prediction of cognitive decline and dementia as advanced imaging. This approach offers a practical, low-cost alternative to PET scans, making it accessible for broader use in early detection, especially in settings where advanced neuroimaging may be unavailable.

Results support the utility of smell and cognition tests as effective tools for identifying individuals at risk for cognitive impairment and dementia.

What is the connection between smell, memory, and Alzheimer's disease?
Davangere P. Devanand, MD: There are different regions of the brain responsible for identifying smells. We can sense a smell, but recognizing it requires memory, language and other brain functions—all areas affected early in Alzheimer's disease. This is why, when Alzheimer's patients are given a smell-identification test, they tend to perform poorly. This deficit is also observed in people with mild cognitive impairment, a condition that may eventually progress to Alzheimer's. Studies, including some from over 20 years ago, consistently show this connection.

We combined a smell-identification test with a brief cognitive test to assess this link further. Cognitive tests are often used in clinics to screen older adults with memory concerns. We used the smell test along with a cognitive test.

We found that combining the smell-identification and cognitive tests strongly predicted who would experience cognitive decline or develop dementia. Surprisingly, this combined test was as effective in predicting cognitive decline as amyloid PET imaging, which is currently a standard tool for Alzheimer's research but requires a specialized PET scanner and costs several thousand dollars. Our combined test, by contrast, is inexpensive and easy to administer.

How do you believe this research will impact future treatment and prevention?
DP: Combining a smell-identification test with a brief cognitive test could offer a simple and cost-effective alternative to expensive and complex brain imaging. While it won’t replace PET imaging, it could serve as an accessible option where imaging isn't available.

Who might benefit the most from these tests?
DP: Older adults experiencing memory or cognitive issues, unsure if these are signs of Alzheimer’s or normal aging, may benefit. This testing can help provide some clarity.

Is there anything else you'd like our audience to know?
DP: The field is evolving rapidly, with new tests emerging, including blood tests. Diagnostic accuracy is improving, and we can expect it to get even better in the future.

This interview was edited for clarity.