Pairwise comparison of Cox regression adjusted survival??

#1
Hey guys,

another beginner question by me. I hope someone can help me out:

I have 3 groups of subjects (patients A, B and C) and I have their survival data which allows me to conduct a Kaplan-Meier analysis. I used the log-rank test to compare the three groups. I also conducted a pairwise comparison of the 3 groups (A vs. B, A vs. C, B vs. C). In addition I computed hazard ratios using the group variable as an ordinal variable and using group A as the reference group, which gives me the HR of B vs. A and C vs. A. So far so good.

Then I adjusted for confounding factors using multivariable Cox regression. Using SPSS I can draw separate survival curves for the group variable (e. g. 3 curves). It is possible to conduct a pairwise comparison of these 3 groups in a multivariable Cox regression model??

I hope I made myself clear enough. Appreciate your help!