anyone knows minimal sample size of cox regression analysis ? - (Oct/24/2006 )

Hello,
I am doing a survival analysis with a sample size of 16.
However, the log-rank (or wilcoxon) test showed significant result (p<0.001) but "NO RESULT" for cox regression model.

Someone told me that my sample size is too small. Is that it?
Thanks!!

Hi there,

Yes, when you have samples less than say 50, then you're bound to get results that do not have a strong confidence level. The Cox method does not assume any particular "survival model" but it is not truly non-parametric because it does assume that the effects of the predictor variables upon survival are constant over time and are additive in one scale. I've been told that you should not use Cox regression without the guidance of a Statistician as you're bound to get errors.