Statalist


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

st: How to make inference on results from two logistic regression models


From   "roland andersson" <rolandersson@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: How to make inference on results from two logistic regression models
Date   Tue, 9 Dec 2008 14:37:54 +0100

I am analysing the difference in risk of surgical complications beween
laparoscopic and open appendectomy in data from a large administrative
register. What i really want is to describe and analyse the risk in
three geographical areas after adjustment for a number ov covariates
(age, sex, diagnnosis at operation). I have done that by running
separate models for each geographical area and keep only the result
for the laparoscopy/open variable in the multivariate model like this:

                         Numbers
Area                 Open                   Laparoscop
Univariate                        Multivariate
Low volume	109/23352 (0.5%) 16/1992 (0.8%) 1.73 (1.02-2.92)
0.042	2.23 (1.28-3.89)	0.005
Medium	       127/35599 (0.4%) 37/7309 (0.5%)	  1.42 (0.98-2.05)
0.061	1.55 (1.04-2.30	0.030
High	               224/41296 (0.5%) 122/13378 (0.9%) 1.69 (1.35-2.11)
<0.001	1.81 (1.43-2.28)	<0.001

How can I make inference on the results of the multivariate models? I
assume there are other ways to analyse this in one single model. Any
comment are appreciated.

Roland Anderson
*
*   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   |   What's new   |   Site index