Online Program Home
My Program

Abstract Details

Activity Number: 652 - Recent Innovation in Generalized Evidence Synthesis
Type: Topic Contributed
Date/Time: Thursday, August 2, 2018 : 10:30 AM to 12:20 PM
Sponsor: Section on Statistics in Epidemiology
Abstract #329503
Title: Bayesian Network Meta-Regression Models Using Heavy-Tailed Multivariate Random Effects with Covariate-Dependent Variances
Author(s): Ming-Hui Chen* and Hao Li and Joseph G Ibrahim and Sung Duk Kim and Arvind K. Shah and Jianxin Lin and Andrew M. Tershakovec
Companies: University of Connecticut and University of Connecticut and University of North Carolina Chapel Hill and National Cancer Institute and MRL, Merck & Co., Inc. and MRL, Merck & Co., Inc. and MRL, Merck & Co., Inc.
Keywords: Bayesian cumulative ranking curve; Conditional CPO; DIC; Ezetimibe drug; MCMC; Statin drug
Abstract:

Many clinical trials have been carried out on safety and efficacy evaluation of cholesterol lowering drugs. To synthesize the results from different clinical trials, we examine treatment level (aggregate) network meta-data from 29 double-blind, randomized, active or placebo-controlled statins +/- Ezetimibe clinical trials on adult treatment-naive patients with primary hypercholesterolemia. In this paper, we assume a multivariate t distribution for the random effects in the proposed network meta-regression model. We further propose a log-linear model for the variances of the random effects so that the variances depend on the arm-by-trial level aggregate covariates. A variation of deviance information criterion and the logarithm of the pseudo marginal likelihood based on conditional CPO's are developed for model comparisons. An efficient Metropolis-within-Gibbs sampling algorithm is developed to carry out the posterior computations. We apply the proposed methodology to conduct an in-depth analysis of the network meta-data from 29 trials with 11 treatment arms.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program