st: RE: metan and other meta-analysis commands in Stata

["Steichen, Thomas J." <SteichT@rjrt.com>]

In addition to the meta-analysis programs summarized by 
Jonathan Sterne, there are quite a few other meta-related 
programs available. I list those in the style Jonathan used, 
continuing his numbering:


8. The metap command
This program combines p-values using either Fisher's method,
Edgington's additive method or Edgington's normal curve method.
It was released in 1999 as a Version 6 program (no graphics)
and last updated in 2000. It requires the user to input p-value
for each study.

9. The metannt command
This program computes the number needed to treat (NNT) and the 
number of events avoided (or added) per 1000. It is designed to
aid interpretation of meta-analyses of binary data by presenting 
the effect sizes in absolute terms. It was released in 2003 as 
a Version 7 program (no graphics) and has not been updated.
It requires the user to input design parameters and uses metan
to calculate needed statistics.

10. The metainf command
This program investigate the influence of a single study on the 
overall meta-analysis estimate and shows graphically the results
when the meta-analysis estimates are computed omitting one study
in each turn. This program makes repeated calls to the meta
program for its analyses. It was released in 1998 as a Version 6
program using Version 6 graphics and was last updated in 2000.
It requires the user to provide input in the form needed by meta.

11. The metaninf command
This program is a port of the metainf command to use metan as
its analysis engine rather than meta. It was released in 2001 
as a Version 6 program using Version 6 graphics and was last 
updated in 2004. It requires the user to provide input in the
form needed by metan.

12. The galbr / galbr8 / rgalbr programs
These program provides a graphical display giving a visual 
impression of the amount of heterogeneity in a meta-analysis.
The galbr program was released in 1997 as a Version 6 program 
using Version 6 graphics and was last updated in 2000. The
galbr8 program was a port to Version 8 with version 8 graphics 
and was released in 2005. The rgalbr program uses a radial
graphical display. It is a Version 8 program and was released 
in 2005 only in test form via Statalist. Each requires the user 
to provide input in the form needed by meta.

13. The labbe program
This program draws a L'Abbe plot for event data (proportion 
of successes in the two groups). It is available via the
metaaggr package as a Version 7 program that uses Version 6
graphics. It requires the user to provide input in the form 
needed by metan.

14. The metagraph program
This program draws Forrest plot. It can be used directly after 
a meta command or the user can input the combined estimate and 
confidence interval. It requires the user to provide input in 
the form needed by meta. It is a Version 8 program released in
2005 and last updated in 2006.

15. The heterogi program
This is an immediate command that provides the statistics H 
and I^2 to quantify heterogeneity in a meta-analysis. It is a 
Version 8 program released in 2005. It requires the user to
input the Q stat and its df, as reported by meta or metan.
(The I^2 stat is now directly available in metan.)

16. The funnel / funnel2 programs
These programs were released in conjunction with metan to 
draw funnel plots.

17. The meta_lr program
This program graphs positive and negative likelihood ratios in 
diagnostic tests. It can do stratified meta-analysis of 
individual estimates. The user must provide the effect estimates 
(log positive likelihood ratio and log negative likelihood ratio) 
and their standard errors. Commands meta and metareg are
used for internal calculations. This is a Version 8 program
released in 2004.


In addition to the above programs, which are directly tied to 
the core meta or metan programs, there are other programs: 

The metaparm program performs meta-analyses and calculates 
confidence intervals and P-values for differences or ratios 
between parameters for different subpopulations for data 
stored in the parmest format.

The glst program calculates a log-linear dose-response 
regression model using generalized least squares for trend 
estimation of single or multiple summarized dose-response 
epidemiological studies.


Lastly, I would advise AGAINST releasing, as Jonathan 
commented, "...a new version of meta that acts as a wrapper 
for metan."  Too many older but useful programs depend on 
the presence of meta to function properly.

Tom

-----------------------------------------
Thomas J. Steichen
steicht@rjrt.com
-----------------------------------------




-----------------------------------------
CONFIDENTIALITY NOTE: This e-mail message, including any
attachment(s), contains information that may be confidential,
protected by the attorney-client or other legal privileges, and/or
proprietary non-public information. If you are not an intended
recipient of this message or an authorized assistant to an intended
recipient, please notify the sender by replying to this message and
then delete it from your system. Use, dissemination, distribution,
or reproduction of this message and/or any of its attachments (if
any) by unintended recipients is not authorized and may be
unlawful.


*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/