st: Metaregression for Poisson distributed outcome

["Alison Drake" <adrake2@u.washington.edu>]

I am having difficulty comparing 2 pooled incidence rates to see if they are
significantly different.  I do not think I can use the metareg command since
incidence rates would follow a Poisson rather than normal distribution. 

I used the metan command to pool incidence rates from 2 populations (call
them population 0 and population 1) from different studies.  Populations 0
and 1 are often not both presented in the same study (different studies are
included in population 0 and population 1), so incidence rate ratios (IRR)
are not possible to calculate.  Here is my metan syntax:
metan incidence lowci upci, random by(population)

Is there a way I can directly compare the pooled incidence rate generated
from the metan comman in population 0 to population 1 to see if they are
significantly different?  Any help would be greatly appreciated, either
STATA syntax to do this.or mathematical formulas to do this by hand if it is
not possible in STATA.


Thanks!
Alison Drake, MPH, PhD
Department of Global Health
University of Washington
email: adrake2@uw.edu




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