Hi,
In my research I have got exposure (drug A=1, drug B=0), gender (1=female, male=0) and outcome (1,0). I created product term gender * exposure and created Cox model (gender, exposure, product term). However, I am not sure how to interpret attached results. Product term has significant p-value. Does it mean that female sex is not independent predictor of outcomes and is modified by the use of drug A? How should I describe this results in paper?
You need a Kaplan Meier plot to see what is going on with your analysis. If you draw one things will be much easier to interpret. An example with such a plot is attached. Good luck, David Booth
BTW I forgot to ask why you added the interaction term in the first place. David
Based on your description of the problem, the outcome is significantly different by sex and not by the drug A. However, the product term that is the interaction is significant. Whenever, the interaction is significant, the main effect significance should be interpreted with caution. Here, interaction is significant means the response of drug among sex is definitely not in the same direction. While males may be showing effect being positive, the females may be showing the effect as negative or vice versa. Another possibility is effect of drug is prominent either in males or in females but not in both.
Please clarify whether the drug A is the test drug and the drug B is placebo.
- Badges
- Science topic
- Similar topics
- Chemistry
- Pharmacology
- Pharmaceuticals
- Drugs