Premium
TD‐P‐17: HOME‐BASED DIGITAL ACTIVITY BIOMARKERS REMOTELY MONITOR RELEVANT ACTIVITIES OF MCI AND ALZHEIMER'S DISEASE PATIENTS AND THEIR CARE PARTNERS
Author(s) -
Mattek Nora,
Thomas Neil W.,
Sharma Nicole,
Beattie Zachary,
Marcoe Jennifer,
Riley Thomas,
Dodge Hiroko H.,
Kaye Jeffrey
Publication year - 2019
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1016/j.jalz.2019.06.4328
Subject(s) - medicine , dementia , wearable computer , activities of daily living , recall , cohort , medical diagnosis , cognition , activity monitor , actigraphy , physical therapy , gerontology , disease , physical activity , psychology , psychiatry , computer science , embedded system , circadian rhythm , pathology , cognitive psychology
Feature odds ratios showed that medications had the greatest number of negatively associated features, physiological features were most consistently positively associated with dementia and features of health system utilization had the greatest extremes of association. Model recall (sensitivity) ranged from 0.69 at 20-18 years to diagnosis, to 0.76 at 20-1 years to diagnosis. Overall precision of the models ranged from 0.13 to 0.30 due to the high proportion of false positives. Conclusions: The three main takeaway messages are: firstly, based on heterogeneous presentations, patients can potentially be identified from their EHR data up to 20 years before diagnosis, currently at a high false positive rate. Secondly, features oscillate in their prominence and predictive strength throughout the prodrome. Thirdly, information around patients’ interaction with the health system have significant value in identifying patients in the prodrome. These data can be incorporated into ongoing efforts to determine progression in the earliest phases of dementia.