Electronic Health Records-Based Tool Uses Data to Detect Undiagnosed Dementia

A study in the Journal of the American Geriatrics Society suggests a newly developed tool based on data in electronic health records (EHRs) could identify people with undiagnosed dementia and tag their records for future follow-up, reports the U.S. National Institute on Aging . Earlier research indicated that under-recognition of dementia is a major issue in the aging population, but little has been done to translate findings from models of future risk of dementia into systems for use in primary care settings to detect undiagnosed cases. The researchers devised the EHR Risk of Alzheimer's and Dementia Assessment Rule (eRADAR) by analyzing 4,330 individuals across 16,665 Adult Changes in Thought visits, and determined that 1,015 visits led to a dementia diagnosis. Of these, 49 percent had not already been coded as such in EHR records. The team then studied various markers by diagnosis to identify which ones were key predictors of undiagnosed dementia and to produce the eRADAR model, which returns a score that increases with the likelihood that an individual has dementia. Subjects who had eRADAR scores in the top 5 percent were more than five times as likely than the overall population to have unidentified dementia, suggesting that it would be valuable to screen more individuals with high scores.