stcox with strata option

[tim bradshaw]

Hi there - I'm trying to understand the use of the strata option with a stcox command.

I'd be interested to hear opinion on the difference between the following, and whether one gives a better estimate of the effect of the exposure than the other.

stcox i.exposure var1 var2 var3

and

stcox i.exposure, strata(var1 var2 var3)

[Marcos Almeida]

A reason to use the option - strata - is the (sometimes, potential; oftentimes, real) violation of the PH assumption.

Please take in mind that, when we use this option, we cannot get the hazard ratio for the stratified variable under the - stcox - command.

Below, further explanation from the Stata Manual:

Stratified estimation.
hen you type . stcox xvars, strata(svars) you are allowing the baseline hazard functions to differ for the groups identified by svars.
This is equivalent to fitting separate Cox proportional hazards models under the constraint that the coefficients are equal but the baseline hazard functions are not.

You may wish to take a look at example 8 as well.