I have a continuous variable X, a continuous mediator M, and a time-dependent outcome Y. Can mediation analyses based on the principles of Baron & Kenny be applied, using Cox-proportional hazards models rather than linear regression models?
Any other statistical procedures that I could use when assessing mediation of a time-dependent outcome?
Thanks!
I think the attached articles can help, as both of them followed Baron & Kenny technique , as well as other methods of statistical test for mediation. hope they help. good luck
I reviewed the articles you kindly provided, however, noted that the independent variable was not time-dependent. My independent variable is a composite of mortality and cardiovascular events, hence my question about using cox-proportional hazard models instead.
Dear Thais Coutinho , did you find out how to do it? (in R preferably). Thanks in advance!
- Badges
- Science topic
- Similar topics
- Geoscience
- Asia
- India