14 May 2009

Understanding risk is the first step toward protecting from any potential disaster—even a medical one. A simple trip to the doctor’s office emphasizes the value of routine tests for preventing a variety of disorders. Unfortunately, there is no quick test for predicting Alzheimer disease (AD)—at least not yet. But there are indicators that help identify those at increased risk for dementia. In yesterday’s Neurology, researchers led by Deborah Barnes at the University of California, San Francisco, debuted a late-life dementia risk index. The index combines some of the known risk factors for dementia into a 15-point scale. Like other commonly used indices, such as the Framingham Heart Index, this tool could help clinicians monitor, prevent, and treat dementia in those at highest risk. It might also be useful for identifying clinical trial participants, suggest the authors.

The index was devised by studying participants in the Cardiovascular Health Cognition Study, which is part of the larger Cardiovascular Health Study (CHS). The CHS is a longitudinal look at risk factors for heart disease and stroke in those over 65 years old (see Fried et al., 1991). The cognitive element was introduced from 1998 to identify people who developed dementia or mild cognitive impairment at follow-up. Barnes and colleagues looked at data from 3,375 participants (average age 76) who were not demented at baseline. Over six years, 14 percent developed dementia, and the authors used risk factors shared by those 480 individuals to develop the index.

The factors that were most predictive of dementia were age, cognitive function (using the 3MS modified Mini-Mental State Exam or the Digit Symbol Substitution Test), body-mass index less than 18.5, ApoE4 status, MRI evidence of white matter disease or enlarged brain ventricles, ultrasound evidence of carotid artery thickening, slow physical performance, history of by-pass surgery, and lack of alcohol consumption. Of those scoring three points or less on the index, 4.2 percent developed dementia, as did 22.8 percent of those scoring between four and seven. A score of eight or higher was indicative of highest risk, with 56 percent progressing to dementia. “I like to stress that the factors we identified are predictive, not causal,” said Barnes in an interview with ARF—lest people are tempted to suddenly ramp up their alcohol consumption.

Barnes said the work needs to be replicated. “This is the first step of a longer process,” she said. In fact, she wants to develop a shorter index since many of the items in the current version, such as the ApoE testing, carotid artery ultrasound, and MRI, are expensive or not readily available in a rural setting. “We want something that could be done in a doctor’s office, something that’s more of a parallel to the Framingham Heart Index,” said Barnes. The hope is that a shorter index would be as good as the longer one. If such an index becomes accepted, it would help identify high-risk patients who can then be monitored or treated early, suggested Barnes. It could also help implement prevention strategies, she said—once those strategies have been developed.—Tom Fagan