Premium
Analysis of longitudinal data in an Alzheimer's disease clinical trial
Author(s) -
Thomas Ronald G.,
Berg Julie D.,
Sano Mary,
Thal Leon
Publication year - 2000
Publication title -
statistics in medicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.996
H-Index - 183
eISSN - 1097-0258
pISSN - 0277-6715
DOI - 10.1002/(sici)1097-0258(20000615/30)19:11/12<1433::aid-sim435>3.0.co;2-f
Subject(s) - clinical trial , disease , context (archaeology) , medicine , alzheimer's disease , sample size determination , dementia , missing data , computer science , statistics , machine learning , paleontology , mathematics , biology
Evidence of delayed progression is the primary mechanism for demonstrating therapeutic efficacy in clinical trials in Alzheimer's disease. In the major trials of therapeutic treatment of AD, to date, measures based on clinical judgement and cognitive performance, instead of mortality, have been used as the primary response measures. There is good reason for this since the course of the disease is quite long, and AD trials designed around mortality would require either very large sample sizes or very long follow‐up in order to have adequate power. However, the evaluation of progression in AD using clinical markers is subject to a number of challenges often found in longitudinal databases, for example, missing data, floor and ceiling effects and non‐linearity. Unfortunately, few of these issues are being addressed in the typical analysis of progression data. This paper explores these analytic issues in the context of the recently completed Alzheimer's Disease Cooperative Study trial of vitamin E and Selegeline in moderate AD patients. Copyright © 2000 John Wiley & Sons, Ltd.