st: aflogit for clustered data?

["Sabine Gabrysch" <sabine.gabrysch@gmx.de>]

Hi!
I am trying to calculate adjusted attributable fractions using the aflogit 
command from the ado file.
My data is hierarchical / clustered / panel data and I use multilevel models 
with random intercepts, STATA commands xtlogit (2 levels) and xtmelogit (3 
levels).
Unfortunately, aflogit does not work after any of these, only after logit. 
The simple logit model ignoring the clustering gives coefficients that 
differ substantially and thus the PAFs will not be correct either...
Is there any way to get adjusted PAFs from multilevel models, be it with 
aflogit or alternative software?
Grateful for any advice.
Thank you,
Sabine
sabine.gabrysch@lshtm.ac.uk 

*
*   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/