Many meta-analysis studies evaluate rates as parameter to assess the overall estimate of effects. However, none of these studies address systematic approaches for the meta-analysis of rates. This paper outlines the conditions, analysis and software operation procedures for the meta-analysis of rates. It also compares different operation procedures of three types of commonly-used R software (Comprehensive Meta-Analysis, Stata and MetaAnalyst) through real application examples. The biggest challenge for the meta-analysis of rates is to determine whether rates can be pooled, and how to evaluate heterogeneity between studies' outcomes needs further discussion.
