Document Server@UHasselt >
Research >
Research publications >

Please use this identifier to cite or link to this item: http://hdl.handle.net/1942/2063

Title: Direct likelihood analysis versus simple forms of imputation for missing data in randomized clinical trials
Authors: Beunckens, Caroline
Molenberghs, Geert
Kenward, Michael G.
Issue Date: 2005
Citation: CLINICAL TRIALS, 2(5). p. 379-386
Abstract: Background In many clinical trials, data are collected longitudinally overtime. In such studies, missingness, in particular dropout, is an often encountered phenomenon. Methods We discuss commonly used but often problematic methods such as complete case analysis and last observation carried forward and contrast them with broadly valid and easy to implement direct-likelihood methods. We comment on alternatives such as multiple imputation and the expectation-maximization algorithm. Results We apply these methods in particular to data from a study with continuous outcomes. The outcomes are modelled using a general linear mixed-effects model. The bias with CC and LOCF is established in the case study and the advantages of the direct-likelihood approach shown. Conclusions We have established formal but easy to understand arguments for a shift towards a direct-likelihood paradigm when analysing incomplete data from longitudinal clinical trials, necessitating neither imputation nor deletion.
Notes: Limburgs Univ Ctr, Ctr Stat, B-3590 Diepenbeek, Belgium. London Sch Hyg & Trop Med, London WC1, England.Molenberghs, G, Limburgs Univ Ctr, Ctr Stat, Bldg D, B-3590 Diepenbeek, Belgium.geert.molenberghs@luc.ac.be
URI: http://hdl.handle.net/1942/2063
DOI: 10.1191/1740774505cn119oa
ISI #: 000233178800001
ISSN: 1740-7745
Category: A1
Type: Journal Contribution
Validation: ecoom, 2006
Appears in Collections: Research publications

Files in This Item:

Description SizeFormat
Published version1.66 MBAdobe PDF

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.