st: post-hoc test for 2x3 table

["Mitchell F. Berman" <mfb1@columbia.edu>]

I have a 2x3 table, looking at use of a specific drug (yes or know) and 
three different surgical procedures.

Fishers exact test gives me an overall significance value and 
demonstrates that the use of the drug is not the same for the three 
different surgical procedures, but I want to say which of the surgical 
procedures have a different rate of use. I could subset out 2 x2 tables 
from my 2x3 table and compare two operations at a time.

Is a Bonferroni correction the only way to adjust for the fact that I'm 
doing multiple post-hoc comparisons?
What if it were a 2x5 table with 5 different surgical procedures?

Is there some better way to approach this type of categorical data?

Thanks in advance.

Mitchell Berman
Columbia University
*
*   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/