Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Bootstrapping METAN by study identifier: epidemiological study, bootstrap meta-analysis


From   "Cangelosi, Michael" <MCangelosi@tuftsmedicalcenter.org>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: Bootstrapping METAN by study identifier: epidemiological study, bootstrap meta-analysis
Date   Wed, 9 Jun 2010 11:40:51 -0400

Dear Statalist:

I have an epidemiological meta-analysis that is examining a few
categories of outcomes with multiple endpoints assigned to these
categories.  I don't care too much about the endpoints themselves.
METAN has been very helpful to get pooled relative-risks of some
outcomes that have only one measurement per study.  

However, I need to modify the usual routine to account for multiple
measurements within a same category (e.g. study X has multiple endpoints
assigned to category (1); so would receive more weighting than a study Y
which has only one endpoint assigned to category (1)).  I am
contemplating bootstrapping METAN, grabbing only one outcome measurement
per study.  While a command something like:

bootstrap ES=r(ES) seLogRR=r(selogrr), metan A B C D by(OutcomeType)

seems close, METAN is still not grabbing only one outcome per study.
Has anyone already performed a similar analysis?  Does METAN have any
capability to address multiple measurements per study? (the .hlp file
seems to suggest no).

Suggestions?

Michael J. Cangelosi MPH, MA

P.S. I Recently posted this message in HTML and suspect message is
trapped, if anyone received both versions, I offer my sincere apoligies.

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index