"... so that I can find out that group A with sex=male and tumor stage = x are at higher risk of dying/getting metastases compared to the other groups".
That sounds like an interaction. You can find examples of Cox regression models with interactions here (near the bottom of the page):
http://www.ats.ucla.edu/stat/spss/examples/asa2/chap5.htmHTH.
manu wrote
Hi,
I have a question regarding COX-analysis. I have read through various forums already and I am not
finding an appropriate answer resp. an answer that I can understand. (Please note, I'm not a native, so
it is most likely that there are some mistakes in my writing).
I have a study population of cancer patients. These patients differ in their genetic coding, so my
population is divided into three groups, and I want to find out prognostic factors (age, sex, tumor stage) that may influence the outcome of a certain group as I assume that they are at higher risk of death or getting metastases (so that I can find out that group A with sex=male and tumor stage = x are at higher risk of dying/getting metastases compared to the other groups).
I have ran some univariate models so I know that gender and other factors influence the outcome of the whole study population.
So should I run the model for each of the group seperately? Or is there a way in SPSS to find that out without running two or three models (most probably I will have to merge two groups as they are very similar in their genetic coding).
I would be really pleased of someone could help me as I am not sure how to do it.
:)
manuel
--
Bruce Weaver
bweaver@lakeheadu.ca
http://sites.google.com/a/lakeheadu.ca/bweaver/"When all else fails, RTFM."
PLEASE NOTE THE FOLLOWING:
1. My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above.
2. The SPSSX Discussion forum on Nabble is no longer linked to the SPSSX-L listserv administered by UGA (
https://listserv.uga.edu/).
Locked
