New models for describing outliers in meta-analysis

Baker, RD; Jackson, D

Abstract

An unobserved random effect is often used to describe the between-study variation that is apparent in
meta-analysis datasets. A normally distributed random effect is conventionally used for this purpose. When
outliers or other unusual estimates are included in the analysis, the use of alternative random effect
distributions has previously been proposed. Instead of adopting the usual hierarchical approach to
modelling between-study variation, and so directly modelling the study specific true underling effects,
we propose two new marginal distributions for modelling heterogeneous datasets. These two distributions
are suggested because numerical integration is not needed to evaluate the likelihood. This makes the
computation required when fitting our models much more robust. The properties of the new distributions
are described, and the methodology is exemplified by fitting models to four datasets.

Citation

Baker, R., & Jackson, D. (2015). New models for describing outliers in meta-analysis. Research Synthesis Methods, 7(3), 314-328. https://doi.org/10.1002/jrsm.1191

• Journal Article Type: Article
• Acceptance Date: Oct 14, 2015
• Publication Date: Nov 27, 2015
• Deposit Date: Jan 12, 2016
• Publicly Available Date: Apr 5, 2016
• Journal: Research Synthesis Methods
• Print ISSN: 1759-2879
• Electronic ISSN: 1759-2887
• Publisher: Wiley
• Volume: 7
• Issue: 3
• Pages: 314-328
• DOI: https://doi.org/10.1002/jrsm.1191
• Publisher URL: http://dx.doi.org/10.1002/jrsm.1191
• Related Public URLs: http://onlinelibrary.wiley.com/journal/10.1002/(ISSN)1759-2887