1
$\begingroup$

I calculated the waist-to-hip ratio from, well, waist and hip measurements in cm. The ratio's values would come out in the range of 0.8-1.1, for my population.

When I performed a univariate cox proportional hazards regression for the time to incidence of diabetes, I got a stupidly high hazard ratio: 146.4. The risk ratios for my other variables are all well within 0.5-to-5 at most.

The last time something like this occurred, I had a very, very small hazard ratio (which is pretty much the same as a big one, I guess). The result was significant. HbA1c was recorded as decimal values instead of a percentage. So I multiplied it by 100, and the hazard ratio was interpretable, and the resultant p-value still fell within significance. (I got this idea from another StackExchange answer). I didn't understand why that worked then, and I don't understand why it isn't working now.

146.4, with p-value 0.0004, is now 1.03481, with p-value 0.78. Can anybody help me develop some intuition for what's going on here?

I'm using R with packages survminer and survMisc, if that makes any difference.

Improve this question
$\endgroup$
3
  • 1
    $\begingroup$ Sounds like a programming error on your side / something you did inconsistently between the two analyses or some kind of numerical problem with the function you use (e.g. numbers get too big to handle in the first analysis, particularly if exponentiating is required). I would first thoroughly check that it's not the first option. $\endgroup$
    – Björn
    Commented Jul 17, 2017 at 7:06
  • $\begingroup$ Oh no. I don't do anything fancy... here is a sample of the code, I am not doing anything special. Do I need to contact the package editors of the coxph or Surv function or something? I messaged my statistician to try and replicate my results in SAS. --- summary(coxph(Surv(survTimePTDM, SigPTDMs)~HbA1c100, data=read.delim("clipboard"))) $\endgroup$
    – user36847
    Commented Jul 17, 2017 at 13:35
  • $\begingroup$ Yes, my clipboard is in order - it's just a copy of some excel columns. $\endgroup$
    – user36847
    Commented Jul 17, 2017 at 13:35

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.