BACKGROUND: The identification of individuals who are at high risk of dementia may lead to earlier detection and diagnosis. In this study, we describe the development and preliminary validation of a very brief, Internet-based risk assessment tool. METHODS: The Dementia Risk Assessment collected information on demographic, neurologic, and medical risk factors; perceived cognitive functioning; and emotional symptoms. Participants answered questions pertaining to themselves ("patient page") or to a relative or friend ("proxy page"). Patients also completed a novel short-delayed recognition memory test. Subjects received personalized feedback about their risk factors and were advised to seek clinical evaluation for concerning symptoms. By November 2009, 500 respondents completed the patient page. RESULTS: The 352 respondents aged 50 years or older (M = 62 years) included a high percentage of women (72%) and college graduates (67%). Among the respondents, only 10% complained of having severe memory problems, whereas 38% reported having a first-degree relative who experienced severe memory loss. The prevalence of major neurologic risk factors was low. Although the delay interval for the recognition memory test averaged around 2 minutes, there was sufficient variability in performance. Individuals with scores ≤10th percentile were significantly older and more likely to be men, have hypertension, and report having severe memory problems. CONCLUSIONS: History of major dementia risk factors can be ascertained through the Internet, and several of these factors are shown to be moderately predictive of performance on a brief test of episodic memory. These data provide preliminary validation of the Dementia Risk Assessment for identifying current cognitive impairment, and suggest its potential for assessing future dementia risk. Ongoing studies will further establish the validity of these findings against proxy reports of cognitive and functional decline, the results of more detailed cognitive assessments, and formal clinical diagnoses.
BACKGROUND: The identification of individuals who are at high risk of dementia may lead to earlier detection and diagnosis. In this study, we describe the development and preliminary validation of a very brief, Internet-based risk assessment tool. METHODS: The Dementia Risk Assessment collected information on demographic, neurologic, and medical risk factors; perceived cognitive functioning; and emotional symptoms. Participants answered questions pertaining to themselves ("patient page") or to a relative or friend ("proxy page"). Patients also completed a novel short-delayed recognition memory test. Subjects received personalized feedback about their risk factors and were advised to seek clinical evaluation for concerning symptoms. By November 2009, 500 respondents completed the patient page. RESULTS: The 352 respondents aged 50 years or older (M = 62 years) included a high percentage of women (72%) and college graduates (67%). Among the respondents, only 10% complained of having severe memory problems, whereas 38% reported having a first-degree relative who experienced severe memory loss. The prevalence of major neurologic risk factors was low. Although the delay interval for the recognition memory test averaged around 2 minutes, there was sufficient variability in performance. Individuals with scores ≤10th percentile were significantly older and more likely to be men, have hypertension, and report having severe memory problems. CONCLUSIONS: History of major dementia risk factors can be ascertained through the Internet, and several of these factors are shown to be moderately predictive of performance on a brief test of episodic memory. These data provide preliminary validation of the Dementia Risk Assessment for identifying current cognitive impairment, and suggest its potential for assessing future dementia risk. Ongoing studies will further establish the validity of these findings against proxy reports of cognitive and functional decline, the results of more detailed cognitive assessments, and formal clinical diagnoses.