Comparing meta-analyses for chronic obstructive pulmonary disease

Abstract

In the paper by Oppe et al., a cost-effectiveness analysis of alternative treatments for chronic obstructive pulmonary disease (COPD), based on data from four different trials, is considered. The goal is to compare the usual (frequentist and Bayesian) fixed-effects (FE) and random-effects (RE) normal model for carrying out meta-analysis. Under RE and FE models, the meta-estimation of some quantities of interest for the disease are also carried out using three out of the four trials, and afterwards data from the fourth are incorporated into the meta-estimation. From these sequential estimators, some conclusions on the FE and RE procedures are drawn. Furthermore, as far as the cost-effectiveness is concerned, the main conclusion of the paper is that the Bayesian RE procedure overrides the Bayesian FE and frequentist methods for cost-effectiveness meta-analysis.