Premium
P3‐018: ELECTRONIC ADAS‐COG (EADAS‐COG) DATA QUALITY: HOW DO COUNTRIES COMPARE?
Author(s) -
Feaster H. Todd,
Barbone Jordan M.,
Garcia-Valdecasas Macarena,
Solomon Todd M.
Publication year - 2018
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.2018.06.1373
Subject(s) - cog , medicine , clinical trial , data quality , computer science , artificial intelligence , metric (unit) , operations management , economics
baseline RAQ scores predicted trial completion and study medication compliance. Youden index was used to estimate optimal cut points. Results: Mean baseline patient RAQ was 31.4 (95%CI 27.0 to 34.7) and mean study partner RAQ was 29.6 (95%CI 24.7 to 33.4). Both patient and study partner RAQ scores at baseline appeared similar to the respective scores at weeks 26 and 52 (all p>0.05). Higher baseline patient RAQ score predicted higher probability of compliance with study medication at weeks 26 and 52 (both p<0.05). Higher baseline study partner RAQ predicted higher probability of trial completion (OR 1⁄4 1.13; p<0.05). Larger disparity between patient and partner RAQ scores predicted lower probability of trial completion (OR1⁄4 0.95; p<0.05). A cut point of 28 optimally discerned between patients who did and did not complete the trial or would be unlikely to be compliant with study medications at a given study visit. In contrast, slightly lower cut points discerned between study partners, 27 and 24 for completion and compliance, respectively. Conclusions:The RAQ may help identify patients and study partners at risk of dropout. Further study is needed in diverse groups and to develop interventions to address the needs of participants identified as at risk for trial dropout.