A fully Bayesian application of the Copas selection model for publication bias extended to network meta-analysis

被引:51
作者
Mavridis, Dimitris [1 ,2 ]
Sutton, Alex [3 ]
Cipriani, Andrea [4 ,5 ]
Salanti, Georgia [1 ]
机构
[1] Univ Ioannina, Sch Med, Dept Hyg & Epidemiol, GR-45110 Ioannina, Greece
[2] Univ Ioannina, Dept Primary Educ, GR-45110 Ioannina, Greece
[3] Univ Leicester, Dept Hlth Sci, Leicester, Leics, England
[4] Univ Verona, Dept Publ Hlth & Community Med, I-37100 Verona, Italy
[5] Univ Oxford, Dept Psychiat, Oxford, England
基金
欧洲研究理事会;
关键词
small study effects; informative priors; antidepressants; prior elicitation; mixed treatment comparison; multiple treatments; ANTIDEPRESSANT TRIALS; FILL METHOD; TRIM; SIMULATION; IMPACT;
D O I
10.1002/sim.5494
中图分类号
Q [生物科学];
学科分类号
07 ; 0710 ; 09 ;
摘要
The Copas parametric model is aimed at exploring the potential impact of publication bias via sensitivity analysis, by making assumptions regarding the probability of publication of individual studies related to the standard error of their effect sizes. Reviewers often have prior assumptions about the extent of selection in the set of studies included in a meta-analysis. However, a Bayesian implementation of the Copas model has not been studied yet. We aim to present a Bayesian selection model for publication bias and to extend it to the case of network meta-analysis where each treatment is compared either with placebo or with a reference treatment creating a star-shaped network. We take advantage of the greater flexibility offered in the Bayesian context to incorporate in the model prior information on the extent and strength of selection. To derive prior distributions, we use both external data and an elicitation process of expert opinion. Copyright (c) 2012 John Wiley & Sons, Ltd.
引用
收藏
页码:51 / 66
页数:16
相关论文
共 35 条
[1]  
[Anonymous], 2011, R: A Language and Environment for Statistical Computing
[2]   Using journal impact factors to correct for the publication bias of medical studies [J].
Baker, Rose ;
Jackson, Dan .
BIOMETRICS, 2006, 62 (03) :785-792
[3]   Modelling multiple sources of dissemination bias in meta-analysis [J].
Bowden, Jack ;
Jackson, Dan ;
Thompson, Simon G. .
STATISTICS IN MEDICINE, 2010, 29 (7-8) :945-955
[4]   Empirical evaluation showed that the Copas selection model provided a useful summary in 80% of meta-analyses [J].
Carpenter, James R. ;
Schwarzer, Guido ;
Ruecker, Gerta ;
Kuenstler, Rita .
JOURNAL OF CLINICAL EPIDEMIOLOGY, 2009, 62 (06) :624-631
[5]   ONE-DIMENSIONAL MARGINAL DENSITY-FUNCTIONS OF A TRUNCATED MULTIVARIATE NORMAL DENSITY-FUNCTION [J].
CARTINHOUR, J .
COMMUNICATIONS IN STATISTICS-THEORY AND METHODS, 1990, 19 (01) :197-203
[6]  
Celeux G, 2006, BAYESIAN ANAL, V1, P651, DOI 10.1214/06-BA122
[7]   Meta-analysis and sensitivity analysis for multi-arm trials with selection bias [J].
Chootrakool, Hathaikan ;
Shi, Jian Qing ;
Yue, Rongxian .
STATISTICS IN MEDICINE, 2011, 30 (11) :1183-1198
[8]   What works?: selectivity models and meta-analysis [J].
Copas, J .
JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES A-STATISTICS IN SOCIETY, 1999, 162 :95-109
[9]  
Copas J, 2000, Biostatistics, V1, P247, DOI 10.1093/biostatistics/1.3.247
[10]   A sensitivity analysis for publication bias in systematic reviews [J].
Copas, JB ;
Shi, JQ .
STATISTICAL METHODS IN MEDICAL RESEARCH, 2001, 10 (04) :251-265