Research methods for meta-analyses

Abstract

Meta-analysis uses numerical tools to pool data and to estimate a summary effect size for the comparison of two interventions from a set of randomised controlled trials identified in a systematic review. An effect size is a single number that expresses the difference in outcome from the interventions. The most commonly used effect sizes for dichotomous outcomes, for example, mortality, are the odds ratio and the relative risk. The results of a meta-analysis are usually presented in a complex figure, known as a forest plot, which shows both the individual studies and the summary statistics. Sensitivity analyses are performed to clarify the effect of the experimental design bias on the effect size. Clinical and statistical heterogeneity of the included studies are explored by the additional tools of fixed effect versus random effects models and subgroup analyses.