EARLY DETECTION OF ALZHEIMER-DISEASE - METHODS, MARKERS, AND MISGIVINGS

There is at present no reliable predictive test for most forms of Alzh eimer disease (AD). Although some information about future risk for di sease is available in theory through ApoE genotyping, it is of limited accuracy and utility. Once neuroprotective treatments are available f or AD, reliable early detection will become a key component of the tre atment strategy. We recently conducted a pilot survey eliciting attitu des and beliefs toward an unspecified and hypothetical predictive test for AD. The survey was completed by a convenience sample of 176 indiv iduals, aged 22-77, which was 75% female, 30% African-American, and of which 33% had a family member with AD. The survey revealed that 69% o f this sample would elect to obtain predictive testing for AD if the t est were 100% accurate. Individuals were more likely to desire predict ive testing if they had an a priori belief that they would develop AD (p = 0.0001), had a lower educational level (p = 0.003), were worried that they would develop AD (p = 0.02), had a self-defined history of d epression (p = 0.04), and had a family member with AD (p = 0.04). Howe ver, the desire for predictive testing was not significantly associate d with age, gender, ethnicity, or income. The desire to obtain predict ive testing for AD decreased as the assumed accuracy of the hypothetic al test decreased. A better short-term strategy for early detection of AD may be computer-based neuropsychological screening of at-risk (old er aged) individuals to identify very early cognitive impairment. Indi viduals identified in this manner could be referred for diagnostic eva luation and early cases of AD could be identified and treated. A new s elf-administered, touch-screen, computer-based, neuropsychological scr eening instrument called Neurobehavioral Evaluation System-3 is descri bed, which may facilitate this type of screening.