GOSH - a graphical display of study heterogeneity.

Res Synth Methods 2012 Sep 11;3(3):214-23. Epub 2012 Sep 11.

Center for Evidence-based Medicine, Program in Public Health, Brown University, Providence, RI, USA.

Estimates from individual studies included in a meta-analysis often are not in agreement, giving rise to statistical heterogeneity. In such cases, exploration of the causes of heterogeneity can advance knowledge by formulating novel hypotheses. We present a new method for visualizing between-study heterogeneity using combinatorial meta-analysis. The method is based on performing separate meta-analyses on all possible subsets of studies in a meta-analysis. We use the summary effect sizes and other statistics produced by the all-subsets meta-analyses to generate graphs that can be used to investigate heterogeneity, identify influential studies, and explore subgroup effects. This graphical approach complements alternative graphical explorations of data. We apply the method to numerous biomedical examples, to allow readers to develop intuition on the interpretation of the all-subsets graphical display. The proposed graphical approach may be useful for exploratory data analysis in systematic reviews. Copyright © 2012 John Wiley & Sons, Ltd.

Download full-text PDF

Source
http://dx.doi.org/10.1002/jrsm.1053DOI Listing
September 2012
4 Reads

Publication Analysis

Top Keywords

graphical display
8
graphical approach
8
heterogeneity
5
studies explore
4
explore subgroup
4
influential studies
4
heterogeneity identify
4
identify influential
4
subgroup effects
4
effects graphical
4
alternative graphical
4
complements alternative
4
approach complements
4
investigate heterogeneity
4
graphical
4
generate graphs
4
studies meta-analysis
4
meta-analysis summary
4
subsets studies
4
meta-analyses subsets
4

References

(Supplied by CrossRef)
A new approach to outliers in meta-analysis
Baker et al.
Health Care Management Science 2008

Cooper et al.
2009
Quantifying heterogeneity in a meta-analysis
Higgins et al.
Statistics in Medicine 2002
A re-evaluation of random-effects meta-analysis
Higgins et al.
Journal of the Royal Statistical Society: Series A 2009
Cumulative meta-analysis of therapeutic trials for myocardial infarction
Lau et al.
The New England Journal of Medicine 1992

Similar Publications