Can you see the wood for the trees? Making sense of forest plots in systematic reviews 2. Analysis of the combined results from the included studies

摘要

Forest plots are graphs whose primary purpose is to present the results obtained in several individual studies in a systematic review. In the first article in this series, we explained how this was done in relation to the "skeleton" and "body" of each forest plot. Forest plots also present the findings of any meta-analysis (statistical combination) for the results of the included studies. In this article we explain how the "bells & whistles" and high-level detail available on a forest plot portray among other things the overall summary estimate of effect, its 95 percent confidence intervals (95% CIs) and the amount of variation between the included studies (heterogeneity). We continue to use the results of a recent Cochrane review on the effects of recombinant factor Vila (rFVIIa) on the number of thromboembolic events as our example#