Premium
Bayesian network meta‐analysis for unordered categorical outcomes with incomplete data
Author(s) -
Schmid Christopher H.,
Trikalinos Thomas A.,
Olkin Ingram
Publication year - 2014
Publication title -
research synthesis methods
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 3.376
H-Index - 35
eISSN - 1759-2887
pISSN - 1759-2879
DOI - 10.1002/jrsm.1103
Subject(s) - categorical variable , computer science , bayesian network , bayesian probability , meta analysis , missing data , data mining , bayesian statistics , social network analysis , artificial intelligence , statistics , econometrics , machine learning , bayesian inference , mathematics , world wide web , medicine , social media
We develop a Bayesian multinomial network meta‐analysis model for unordered (nominal) categorical outcomes that allows for partially observed data in which exact event counts may not be known for each category. This model properly accounts for correlations of counts in mutually exclusive categories and enables proper comparison and ranking of treatment effects across multiple treatments and multiple outcome categories. We apply the model to analyze 17 trials, each of which compares two of three treatments (high and low dose statins and standard care/control) for three outcomes for which data are complete: cardiovascular death, non‐cardiovascular death and no death. We also analyze the cardiovascular death category divided into the three subcategories (coronary heart disease, stroke and other cardiovascular diseases) that are not completely observed. The multinomial and network representations show that high dose statins are effective in reducing the risk of cardiovascular disease. Copyright © 2013 John Wiley & Sons, Ltd.