Example 1: i want to test if Diabetes is a predictor of myocardial infarction. The result is this:
Covariate b SE Wald P Exp(b) 95% CI of Exp(b)
Diabetes 1,1624 0,3164 13,4996 0,0002 3,1976 1,7254 to 5,9257
How can i interpret this result? the p is less than 0,05 but i don't understand if it is in favor of patients with diabetes or without diabetes.
Example 2: And with continuos variables, for example:
Covariate b SE Wald P Exp(b) 95% CI of Exp(b)
RVD -1,0549 0,1800 34,3351
Maybe this discussion thread can help you think about your interpretation.
http://stats.stackexchange.com/questions/6026/how-do-i-interpret-expb-in-cox-regression
It will depend on how your diabetes variable is coded. For instance in your first example, if diabetes is coded 0= no diabetes and 1=diabetes then diabetes is related to a increased risk of MI.
Several of the suggestions on the thread I linked offer good resources for both interpreting Cox regressions as well as learning how to correctly describe and talk about the hazards ratio: Exp(b).
Good luck!
You need the Kaplan-Meier curves which may help.. My favorite book is given in the first link below. You might also use lasso with Cox regression. the regularization path may help. See the books in the second and third(Chapt 17 and 19) links. Lander's book has R code. Matthew's suggestions are quite good as well. Best wishes, David.
https://www.amazon.com/Statistical-Methods-Survival-Data-Analysis/dp/0471369977/ref=sr_1_2?ie=UTF8&qid=1486967455&sr=8-2&keywords=lee+survival+data+analysis
http://www4.stat.ncsu.edu/~boos/var.select/lasso.adaptive.html
https://www.amazon.com/Everyone-Advanced-Analytics-Graphics-Addison-Wesley/dp/0321888030/ref=sr_1_1?s=books&ie=UTF8&qid=1486967895&sr=1-1&keywords=jared+lander
You really should not need to wonder about this as you should have explored the data sufficiently to understand at least the direction of association. I agree with David that at the very least you should just run kaplan meier kurves to understand your data. Also I think your question should be "if diabetes is a predictor for TIME to MI" if you simply want to know if it predicts (or rather is associated with) MI why would you run a cox regression that takes into consideration time? For stata you should stsset your data with diabetes as the panel variable and then run stsgraph to get the kaplan meier plots.
- Badges
- Science topic
- Similar topics
- Medicine
- Disease
- Metabolic Diseases
- Glucose Metabolism Disorders
- Diabetes