Premium
P3–120: Predictors of dropout in clinical trials of subjects with mild cognitive impairment (MCI)
Author(s) -
Emond Jennifer A.,
Jin Sheila,
Bochenek Jackie,
Thal Leon J.,
Petersen Ronald C.,
Sano Mary,
Edland Steven D.
Publication year - 2006
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.2006.05.1388
Subject(s) - dropout (neural networks) , univariate , dementia , clinical trial , psychology , marital status , medicine , clinical psychology , gerontology , censoring (clinical trials) , multivariate statistics , disease , demography , environmental health , statistics , population , mathematics , pathology , machine learning , sociology , computer science
Background: Knowledge of which factors influence retention rates in clinical trials may inform recruitment and retention strategies of future clinical trials. Studies with a low dropout rate have increased statistical power. A low dropout rate also reduces the chance of informative censoring. Hence, focused retention efforts may improve a trial’s power and validity. Objective: To better understand the influence of baseline characteristics on dropout rates in secondary prevention trials testing treatments to delay conversion from mild cognitive impairment (MCI) to Alzheimer’s disease. Methods: Subjects are 769 participants from 69 sites enrolled in a three year secondary prevention trial conducted by the Alzheimer’s Disease Cooperative Study (supported by a grant from the National Institute on Aging (UO1-AG104083), the Institute for the Study of Aging, Pfizer Inc, and Easai Inc). Analysis was by Cox proportional hazards regression. Baseline characteristics significant at the 0.10 level in univariate analyses were considered for inclusion in a final multivariate model. Results: A total of 230 subjects (29%) discontinued for reasons other than death or conversion to dementia. Age, marital status, informant relationship, gender, education, and race were significant predictors of dropout in univariate analyses. In multivariate analysis, risk of dropout remained significantly higher (p 0.05) among non-white subjects and among subjects with less than college education. Conclusions: Dropout rates were higher among minorities and those with less education. Retention efforts focused on these areas may improve the statistical power and validity of secondary prevention trials.