Likelihood-based random-effects meta-analysis with few studies: empirical and simulation studies.

Affiliation

Department of Medical Statistics, University Medical Center Göttingen, Humboldtallee 32, Göttingen, 37073, Germany. [Email]

Abstract

Standard random-effects meta-analysis methods perform poorly when applied to few studies only. Such settings however are commonly encountered in practice. It is unclear, whether or to what extent small-sample-size behaviour can be improved by more sophisticated modeling.

Keywords

Count data,Generalized linear mixed model (GLMM),Hartung-Knapp-Sidik-Jonkman (HKSJ) adjustment,Normal-normal hierarchical model (NNHM),Random-effects meta-analysis,