Premium
P1‐554: DISCOVERING SUBTYPES OF ALZHEIMER'S DISEASE USING ELECTRONIC HEALTH RECORDS
Author(s) -
Alexander ie,
Direk Kenan,
Alexander Daniel C.,
Denaxas Spiros
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.2019.06.1159
Subject(s) - silhouette , cluster analysis , latent class model , depression (economics) , disease , medicine , anxiety , artificial intelligence , psychology , computer science , machine learning , psychiatry , economics , macroeconomics
24.0%; $501/participant); flyers (n1⁄413; 13.5%; $354/participant); community events (n1⁄410; 10.4%; $120/participant); targeted mailings (n1⁄48; 8.3%; $178/participant); press releases (n1⁄43; 3.1%; $90/participant), online (n1⁄43; 3.1%; $90/participant), advertisements (n1⁄42; 2.1%; $1,554/participant), community presentations (n1⁄42; 2.1%; $185/participant), support groups (n1⁄42; 2.1%; $185/ participant), and other (n1⁄47, 7.3%; $0/participant). Barriers to recruitment included transportation, type of exercise, existing schedules, vacations, MRI-incompatibility, and language. Conclusions:Our enrollment rate of 31.9% is higher than previously reported. For every enrolled participant, 1.8 needed to be screened in-person. Referrals were the most effective recruitment strategy. Indirect methods of recruitment (ads, website, and mailings) were less effective than direct methods (community events, presentations, and Alzheimer’s Association support). Networking and visibility were critical. Targeting caregivers through conferences with this emphasis was instrumental in our success.