Meta-analysis is a statistical method to compare and combine effect sizes from a pool of relevant empirical studies. It is now a standard approach to synthesize research findings in many disciplines, including medical and healthcare research. This paper is the third paper of a mini-series introducing systematic review and meta-analysis. First, common effect sizes used in meta-analysis are presented. Fixed-, random- and mixed-effects models are then introduced. Next, a real data set from a published meta-analysis will be used to illustrate the procedures and interpretations. Last, software packages that may be used to conduct meta-analyses will be highlighted.
© 2012 The Authors International Journal of Rheumatic Diseases © 2012 Asia Pacific League of Associations for Rheumatology and Blackwell Publishing Asia Pty Ltd.