I'm interested in analyzing two time-to-event variables simultaneously, using a survival model. Each one of the subjects would be given the drug, and each one would have two time-to event variables measured (For example time to analgesia and time to improvement of disability). The outcome would not be a survival curve but a 'survival plane'. Has anybody worked on a similar model, or at least in clinical trials looking at outcomes with bivariate distributions ? Thank you F. Diaz Sent via Deja.com http://www.deja.com/ Before you buy.