Cognitive Score Changes Over Time May Hint at Dementia

Study authors stress new method is a complement to current testing

TUESDAY, Aug. 19, 2008 (HealthDay News) -- Yearly changes in a person's performance on cognitive testing may be associated with dementia, new research suggests.

Using a newly developed model to assess the effect of variations in a person's score from year to year, researchers found that just a one point change in variability on cognitive test scores could indicate as much as a fourfold increase in the risk of developing dementia.

"Above and beyond the prediction made by performance on neuropsychological tests, if you take into account the degree to which the individual varies in their cognitive abilities, there's an independent risk of dementia," said study author Roee Holtzer, an assistant professor of psychology and neurology at the Albert Einstein College of Medicine at Yeshiva University in New York City.

Results of the study are published in the Aug. 20 issue of the Journal of the American Medical Association.

Holtzer was quick to point out that the newly developed variability model was a complement to currently used tests, and is not meant to replace them. Additionally, he said, the findings need to be replicated.

Currently, doctors use cognitive function and memory tests to diagnose dementia or mild cognitive impairment. But, these tests compare one person to the "norm" for their age and education level.

Holtzer and his colleagues theorized that changes within an individual's score over time might be an additional tool for diagnosing dementia.

To that end, they developed a model to assess "within-person" variability on commonly used neuropsychological tests that compensated for sex, education, medical illness, and performance on each individual test. The model was then used to evaluate 897 people older than 70 between October 1993 and December 2007. Baseline testing was done, and then the participants had follow-up exams every 12 to 18 months.

During the follow-up, 61 cases of dementia were identified. Twenty-six of these people had scores in the highest quartile of the within-person variability index.

The researchers found that when they controlled for sex, education and medical illness, each one-point difference in the variability scores increased the risk of dementia by nearly four times. If they also factored in adjustments for scores on individual tests, the risk was still more than doubled for each one point difference in the variability model.

"The investigators found that relatively brief measures of memory and executive function reassessed annually would allow physicians to predict of dementia before the patient scores in the clearly impaired range," said Dr. Gary Kennedy, director of the division of geriatric psychiatry at Montefiore Medical Center in New York City.

"This study brings us one step closer to a simple, safe and reliable test of dementia risk," added Kennedy. And, that's important, he said, because once medications are developed to treat dementia, you have to be able to correctly identify who needs the treatment.

More information

To read more about diagnosing dementia, visit the Family Caregiver Alliance.

SOURCES: Roee Holtzer, Ph.D., assistant professor, psychology and neurology, Albert Einstein College of Medicine, Yeshiva University, New York City; Gary Kennedy, M.D., director, geriatric psychiatry, Montefiore Medical Center, New York City; Aug. 20, 2008, Journal of the American Medical Association
Consumer News