Competing risks regressin models

[Richard Lin2]

Hello,

I used competing risk regression model using the command below, but would like to model it using cause specific hazard in COX PH. Any suggestion of the commands?

. stcrreg txyear10, compete(nrm==2)

failure _d: nrm == 1
analysis time _t: nrmm100

Thanks,
Richard

[Andrea Discacciati]

Use Cox regression censoring the competing event.
Something like:

stset nrmm100, fail(nrm = 1)
stcox txyear10

[Richard Lin2]

Thanks Andrea, I see this is actually the same command for cox ph regression model then graphed with cumulative hazard