Hello everyone,
I am about to construct various Cox regression models that will each include a time dependent covariate. I intend to use SPSS for model generation. Can anyone advise me how to use SPSS (i.e. to generate T_Cov) to do the following (the user guide says that "T_Cov should take as one of its arguments the observed survival time for an uncensored case, T_" ....thus I am a little stumped) :
1) Model 1:
I have overall survival time recorded (i.e time from admittance to a programme to date last seen). I have a status variable recorded as 1 if patient has died and 0 if patient is still alive. x_i(t) is a time dependent covariate where xi(t)=1 when a patient has had a transplant and x_i(t)=0 when they have not. How do I form x_i(t) in the "Expression for T_Cov" box in SPSS?
2)Model 2:
Again I have overall survival time recorded (i.e time from admittance to a programme to date last seen). I have a status variable recorded as 1 if patient has died and 0 if patient is still alive. This time, we are interested in analysing outcome accounting for cumulative dosage of drugs received. I have recorded the amount of drugs received by each patient at date last seen. z_i(t) is a time dependent covariate as the longer the individuals survive, the higher the cumulative dosage tends to be. How do I form z_i(t) in the "Expression for T_Cov" box in SPSS?
Thanks so much for your help on these queries,
Kindest Regards,
Kim
You may leave the list at any time by sending the command
SIGNOFF allstat
to [log in to unmask], leaving the subject line blank.
|