Objectives: The workshop will provide an overview of Comprehensive Meta-Analysis, a computer program for meta-analysis that was developed with funding from the National Institutes of Health (NIH) in the United States. The program incorporates state-of-the-art functionality for computing treatment effects, performing a meta-analysis, assessing the impact of study-level characteristics, and creating high-resolution forest plots. Attendees will learn how they can use Comprehensive Meta-Analysis (Version 2) to perform these tasks, and will be asked for suggestions that will help shape future development of the program.

Description: We will show how Comprehensive Meta-Analysis can be used to perform the following tasks.

1. Compute a treatment effect and its standard error. We will show how to enter data in almost any format and have the program compute the treatment effect and standard error automatically. We will also discuss formulas used for computing various indices of treatment effect including all those available in RevMan.
2. Perform the meta-analysis. We will show how to work with various computational options (fixed, random, and mixed effects), how to examine diagnostics, and how to display and understand the weights assigned to each study. We will show how to assess the potential impact of study-level characteristics (using subgroup analysis, analysis of variance and meta-regression), how to assess and adjust for publication bias, and how to work with multiple outcomes per study.
3. Create a high-resolution forest plot. Attendees will also learn how to customize the forest plot and to create a high-resolution version of the plot for export to PowerPoint, Word, and other programs.

Target audience: Review authors, and anyone who performs meta-analyses using RevMan, Stata, or other software.

Style: Training workshop

Disclosure statement: Comprehensive Meta Analysis is being sold commercially. The presenters of this workshop stand to benefit financially from sales of the program.