Generalized linear mixed models are believed to overcome the problems of the standard random-effects model because they use a binomial-normal likelihood. These models can be fitted in SAS and in R using the metafor package by Viechtbauer. Additionally, the standard REM suffers from transformation bias () and bias in the estimation of the random-effect variance τ 2.Īn attractive alternative approach for the meta-analysis of binary outcomes uses a class of generalized linear mixed models (GLMMs). When the outcome of interest is a transformation of a binomial outcome such as the logit transformation, the standard random-effects model assumes that within-study variability can be described by an approximate normal likelihood, i.e. Since the individual studies might differ in populations and structure, their effects are often assumed to be heterogeneous, and the use of methods based on random-effects models is recommended. Meta-analysis is a statistical technique for synthesizing outcomes from several studies. The problem of finding uniformly good methods of the meta-analysis for binary outcomes is still open. It is difficult to recommend the use of GLMMs in the practice of meta-analysis. In our simulations, the hypergeometric-normal model provided less biased estimation of the heterogeneity variance than the standard random-effects meta-analysis using the restricted maximum likelihood (REML) estimation when the data were sparse, but the REML method performed similarly for the point estimation of the odds ratio, and better for the interval estimation. The present study is the first to provide extensive simulations on the performance of four GLMM methods (models with fixed and random study effects and two conditional methods) for meta-analysis of odds ratios in comparison to the standard random effects model. This gap may be due to the computational complexity of these models and the resulting considerable time requirements. However, this belief is based on theoretical considerations, and no sufficient simulations have assessed the performance of GLMMs in meta-analysis. GLMMs are believed to overcome the problems of the standard random-effects model because they use a correct binomial-normal likelihood. An attractive alternative approach for the meta-analysis of binary outcomes uses a class of generalized linear mixed models (GLMMs). However, the standard meta-analysis of odds ratios using a random-effects model has a number of potential problems. The odds ratio, in particular, is by far the most popular effect measure. Systematic reviews and meta-analyses of binary outcomes are widespread in all areas of application.
0 kommentar(er)
