```Date: Sun, 25 Dec 2011 21:27:37 -0800 Reply-To: zmlr Sender: "SPSSX(r) Discussion" From: zmlr Subject: Problem of COX model with time dependent covariate Content-Type: text/plain; charset=us-ascii Hi all, I am trying to detect association between a covariate and a disease outcome. This covariate shows time-varying effect, so I use "Cox w/Time-Dep cov" in SPSS 17.0 to build cox model, the T_COV_ is equal to T_. The result is as follows: B SE Wald df Sig. Exp(B) eGFR_BASE -13.603 .548 616.007 1 .000 .000 T_COV_*eGFR_BASE .860 .115 55.859 1 .000 2.362 Some materials explained that "T_COV_*eGFR_BASE" is a time-covariate interaction item in fact, so I recode a new variable,named "TIME", which exactly equal to survival time, then input the conventional Cox model. The result is as follows: B SE Wald df Sig. Exp(B) eGFR_BASE 6.317 .370 291.547 1 .000 554.166 TIME*eGFR_BASE -3.372 .076 1944.385 1 .000 .034 My Questions are: 1, Why it's so different between two results? How T_COV_ was calculated in SPSS, and is it the same with the survival time? 2, I also try to calculate it using Cox in R, I use time:eGFR_BASE as time dependent item, the result is same as the second one above.But whether I can do the first procedure in R? Many thanks, Kevin -- View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Problem-of-COX-model-with-time-dependent-covariate-tp5101002p5101002.html Sent from the SPSSX Discussion mailing list archive at Nabble.com. ===================== To manage your subscription to SPSSX-L, send a message to LISTSERV@LISTSERV.UGA.EDU (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD ```

Back to: Top of message | Previous page | Main SPSSX-L page