TY - JOUR
T1 - Bayesian Inference on Risk Differences
T2 - An Application to Multivariate Meta-Analysis of Adverse Events in Clinical Trials
AU - Chen, Yong
AU - Luo, Sheng
AU - Chu, Haitao
AU - Wei, Peng
N1 - Funding Information:
The authors are grateful to Dr. Steven Snapinn, the Associate Editor, and two anonymous reviewers for the helpful comments that have greatly improved this article. Yong Chen’s research was partially supported by a start-up fund and the PRIME award from the University of Texas School of Public Health. Sheng Luo’s research is partially supported by two NIH/NINDS grants U01NS043127 and U01NS43128. Haitao Chu was supported in part by the U.S. Department of Health and Human Services Agency for Healthcare Research and Quality grant R03HS020666, and P01CA142538 P30CA077598 from the U.S. National Cancer Institute. Peng Wei was supported in part by the National Institutes of Health grant HL095511.
PY - 2013/4
Y1 - 2013/4
N2 - Multivariate meta-analysis is useful in combining evidence from independent studies that involve several comparisons among groups based on a single outcome. For binary outcomes, the commonly used statistical models for multivariate meta-analysis are multivariate generalized linear mixed effects models which assume risks, after some transformation, follow a multivariate normal distribution with possible correlations. In this article, we consider an alternative model for multivariate meta-analysis where the risks are modeled by the multivariate beta distribution proposed by Sarmanov. This model has several attractive features compared to the conventional multivariate generalized linear mixed effects models, including simplicity of likelihood function, no need to specify a link function, and a closed-form expression of distribution functions for study-specific risk differences. We investigate the finite sample performance of this model through simulation studies and illustrate its use with an application to multivariate meta-analysis of adverse events of tricyclic antidepressant treatment in clinical trials.
AB - Multivariate meta-analysis is useful in combining evidence from independent studies that involve several comparisons among groups based on a single outcome. For binary outcomes, the commonly used statistical models for multivariate meta-analysis are multivariate generalized linear mixed effects models which assume risks, after some transformation, follow a multivariate normal distribution with possible correlations. In this article, we consider an alternative model for multivariate meta-analysis where the risks are modeled by the multivariate beta distribution proposed by Sarmanov. This model has several attractive features compared to the conventional multivariate generalized linear mixed effects models, including simplicity of likelihood function, no need to specify a link function, and a closed-form expression of distribution functions for study-specific risk differences. We investigate the finite sample performance of this model through simulation studies and illustrate its use with an application to multivariate meta-analysis of adverse events of tricyclic antidepressant treatment in clinical trials.
KW - Bivariate beta-binomial model
KW - Exact method
KW - Hypergeometric function
KW - Relative risk
KW - Sarmanov family
UR - http://www.scopus.com/inward/record.url?scp=84880986848&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84880986848&partnerID=8YFLogxK
U2 - 10.1080/19466315.2013.791483
DO - 10.1080/19466315.2013.791483
M3 - Article
C2 - 23853700
AN - SCOPUS:84880986848
SN - 1946-6315
VL - 5
SP - 142
EP - 155
JO - Statistics in Biopharmaceutical Research
JF - Statistics in Biopharmaceutical Research
IS - 2
ER -