st: Exposure/offset and independent variables

[Evan Cole <cole.s.evan@gmail.com>]

Stata users,

I am trying to estimate the number of emergency room visits for the
patient populations of different clinics.
I am using a negative binomial model with the number of patients at
the clinic as the exposure option.
The outcome is the count of ER visits.
I also control for certain characteristics of the patient population,
including the number of patients with 2+ chronic diseases, average
age, and the number of patients that are adult female.
My question is how to treat the independent variables.  If I use the
number of patients with 2+ chronic diseases, will my exposure variable
of total number of patients turn that variable into a rate?  Or should
I put that variable in the model as a rate (% of patient population
with 2+ chronic diseases)?
These two approaches give me very different results.
Any help would be greatly appreciated.

Evan
Doctoral Candidate
Tulane University
*
*   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/