Current research does not support use of smell tests for Alzheimer's dementia

May 4 2012

Current research does not support the use of smell tests for predicting Alzheimer's dementia, according to a comprehensive review by Robert Wood Johnson Foundation Clinical Scholars at the University of Michigan and the VA Center for Clinical Management Research.

The paper was published online this week in the journal Laryngoscope.

"Smell tests have been touted as a possible way of predicting Alzheimer's dementia because of a reported association with decreased sense of smell," says Gordon Sun, M.D., a general otolaryngologist at U-M and RWJF/US Department of Veterans Affairs Clinical Scholar at the University of Michigan Medical School. "Our team set out to determine whether these beliefs are based on existing high-quality evidence."

This is the first systematic review of the quality and quantity of research on olfactory identification tests (smell tests) as prognostic tools for Alzheimer's dementia. The authors reviewed nearly 1,200 articles dating back to 1984, finding only two prospective longitudinal cohort studies and 30 cross-sectional studies that met inclusion criteria.

The authors considered the longitudinal studies more significant because they track patients over time and represent the gold standard study design for prognostic evaluations. Cross-sectional studies, on the other hand, evaluate patients during one specific point in time and are more susceptible to findings that are influenced by hidden variables.

The authors found that there is evidence to show an association between the loss of smell and Alzheimer's dementia, but not to conclude that loss of smell is a predictor of developing the disease later in life.

Sun and colleagues caution that there are many other possible explanations for this association, noting that other medical problems are also linked with loss of smell.

"A nonspecific association between poor smell function and Alzheimer's dementia is not the same as actually being able to use a smell test to predict Alzheimer's," says Sun. "Unfortunately, this misinterpretation of the research has led to the promotion of these tests by the media and public figures like Dr. Oz. This study helps set the record straight about where the evidence currently stands."

Alzheimer's is the most common cause of dementia and is expected to double every 20 years through the year 2040.

"Understandably, researchers, clinicians, and the public are eager for a simple, accurate, and inexpensive way to predict or diagnose Alzheimer's early, but we're not there yet," warns Sun. "My concern is that by promoting smell tests at this point, we create false hope or even false alarm among seniors and their families. Additional research is needed before we can rely on smell tests to predict the later onset of Alzheimer's."

In the meantime, the study authors recommend patients consult with their primary care physicians first, if there are any concerns that they or your loved one may be at risk for Alzheimer's dementia. Patients may also obtain information from another trusted source, such as the Alzheimer's Association.

Source: Laryngoscope