Title: METAN: Stata module for fixed and random effects meta-analysis
Abstract: Meta-analysis is a statistical technique for combining results from multiple independent studies, with the aim of estimating a single overall effect. The routines in this package provide facilities to conduct meta-analyses of binary (event) or continuous data from two groups, or intervention effect estimates with corresponding standard errors or confidence intervals. This is an updated version of metan as published in Stata Journal Issue 8, and prior to that in STB-44, authored by Michael J Bradburn, Jonathan J Deeks and Douglas G Altman. Updates include a wide range of random-effects models; cumulative and influence analysis; meta-analysis of proportions; and better handling of heterogeneity, continuity correction and returned values. The routine for constructing forest plots has been separated off (‘forestplot’ command) and hugely extended; extremely flexible and generalised forest plots may now be produced. Also included is an “immediate” command ‘metani’, which accepts numlists or matrices as input rather than variables in memory; the ‘metannt’ program for binary data, which displays estimated intervention effects in terms of the absolute reduction in risk and number needed to treat; and the ‘labbe’ program which produces L'Abbe plots to examine whether the assumption of a common odds ratio, risk ratio or risk difference is reasonable. Further information on this package, including validation scripts, may be found at https://github.com/UCL/metan; and a description of available Stata meta-analysis commands may be found at http://www.stata.com/support/faqs/stat/meta.html.
Publication Year: 2007
Publication Date: 2007-10-29
Language: en
Type: article
Access and Citation
Cited By Count: 60
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot