Should I use Cox Regression or Kaplan-Meier Survival Analysis?

[mkit8]

Hello,

I am having trouble determining whether I should be using Cox Regression or Kaplan-Meier Survival Analysis to examine my data. I would like to examine the differences in survival between two age groups who have a certain type of cancer after adjusting both models for 7 covariates (1 of these variables is binary, 4 are categorical, and 2 are nominal).

This is my first time doing anything with survival analysis and I would appreciate guidance.

Thank you!

[Ksharp]

Does the life time censored ? If not , no need survival analysis .

[mkit8]

Hi, KSharp,

Yes, I am using survival months and then censoring for my 'cause specific death' variable. 

[Ksharp]

Yes. try survival analysis.
PROC LIFETIME ........

[Arik]

Hi

I currently read two papers about this. They compare the different approaches in example studies. I'm not sure if they totally fit to your topic, but maybe they'll be helpful.

https://clincancerres.aacrjournals.org/content/early/2012/01/25/1078-0432.CCR-11-2097DOI: 10.1158/10...

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2873048/

[riya275]

Since you want to adjust for covariates then you should use cox proportion.

[OsoGris · Accepted solution]

You need to use Cox regression (PROC PHREG) to account for the covariates in your model.   The KM Survival Analysis (PROC LIFETEST) does not do any modelling.