st: Comparison of coefficients from different cox regression models

[Amal Khanolkar <Amal.Khanolkar@ki.se>]

Hi all,

I'm running two sets of cox regression models - same exposure and outcome and adjusted for the same covariates - except that the models are run stratified by gender.

I would now like to compare the coefficients obtained from the two models, i.e. the coefficients for men and women and test if they are statistically significantly different from each other. Ideally I guess this would be done by introducing an interaction term with the variable that specifies gender. Or is there another way more specific for cox regression?

For men:
 xi: stcox i.Genotype*i.BWcat age x290 if sex==1

For women:

 xi: stcox i.Genotype*i.BWcat age x290 if sex==2



Thanks,
Amal Khanolkar
*
*   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/