Premium
Allowing for uncertainty due to missing data in meta‐analysis—Part 2: Hierarchical models
Author(s) -
White Ian R.,
Welton Nicky J.,
Wood Angela M.,
Ades A. E.,
Higgins Julian P. T.
Publication year - 2007
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/sim.3007
Subject(s) - missing data , markov chain monte carlo , computer science , markov chain , odds , random effects model , statistics , econometrics , monte carlo method , meta analysis , data mining , mathematics , machine learning , logistic regression , medicine
We propose a hierarchical model for the analysis of data from several randomized trials where some outcomes are missing. The degree of departure from a missing‐at‐random assumption in each arm of each trial is expressed by an informative missing odds ratio (IMOR). We require a realistic prior for the IMORs, including an assessment of the prior correlation between IMORs in different arms and in different trials. The model is fitted by Monte Carlo Markov Chain techniques. By applying the method in three different data sets, we show that it is possible to appropriately capture the extra uncertainty due to missing data, and we discuss in what circumstances it is possible to learn about the IMOR. Copyright © 2007 John Wiley & Sons, Ltd.