C
Good Morning!
I'm working on a survival analysis involving transplant data.
STATUS - graft status
TIME - days until graft rejection
GROUP - a/b
COVAR1 - a/b/c
COVAR2 - yes/no
COVAR3 - yes/no
In SAS, I used the following code:
PROC PHREG DATA=PRELIMINARY;
MODEL TIME*STATUS = GROUP COVAR1 COVAR2 COVAR3 COVAR1_T COVAR2_T COVAR3_T / ties=efron;
COVAR1_T = COVAR1*TIME;
COVAR2_T = COVAR2*TIME;
COVAR3_T = COVAR3*TIME;
RUN;
COVAR1_T and COVAR2_T turned out to be significant which, I think, means the proportionality assumption has been violated.
I'm worried that I haven't gone about this correctly and am also kind of stuck at what to do next. Any ideas?
Thank you in advance!
Christyn
I'm working on a survival analysis involving transplant data.
STATUS - graft status
TIME - days until graft rejection
GROUP - a/b
COVAR1 - a/b/c
COVAR2 - yes/no
COVAR3 - yes/no
In SAS, I used the following code:
PROC PHREG DATA=PRELIMINARY;
MODEL TIME*STATUS = GROUP COVAR1 COVAR2 COVAR3 COVAR1_T COVAR2_T COVAR3_T / ties=efron;
COVAR1_T = COVAR1*TIME;
COVAR2_T = COVAR2*TIME;
COVAR3_T = COVAR3*TIME;
RUN;
COVAR1_T and COVAR2_T turned out to be significant which, I think, means the proportionality assumption has been violated.
I'm worried that I haven't gone about this correctly and am also kind of stuck at what to do next. Any ideas?
Thank you in advance!
Christyn