ABSTRACT
This chapter presents meta-analysis methods for clinical trials using the R system. It illustrates three datasets. The first dataset is the 22 multi-center betablocker trial to reduce mortality after myocardial infarction. The second dataset is the classical and famous data from Cochrane Collaboration logo that resulted from systematic reviews of the entire, pre-1980 clinical trial literature of corticosteroid therapy in premature labor and its effect on neonatal death. The third dataset contains estimates of treatment effect from eight randomized controlled trials of the effectiveness of amlodipine as compared to placebo in improving work capacity in patients with angina. The chapter also presents useful meta-analysis models including the weighted mean method known as fixed-effects model and the DerSimonian-Laird random-effects model implemented in R libraries rmeta and meta. It demonstrates how to use R and the R functionalities from both libraries to analyze these three datasets.
