ASA Traveling Course: Survival Analysis with Correlated or Repeated Endpoints

Survival Analysis with Correlated or Repeated Endpoints

 Presented by: Terry M Therneau, Department of Health Science Research, Mayo Clinic

Sponsored by: The Philadelphia, Delaware, and Northern New Jersey Chapters of the American Statistical Association and an award from the American Statistical Association.

Abstract: The Cox model has become the standard for analysis of time-to-event data in the medical sciences. In this lecture we will look at the extension of the model to cases where the endpoints are correlated, either due to multiple endpoints per subject (such as repeated disease episodes) or interrelated subjects such as family studies. The two primary tools are the use of GEE like models and random effects models.

The material for the first of these overlaps substantially with that in the book by Therneau and Grambsch, that for random effects overlaps somewhat but is largely new. In particular the software has made great strides in the last several years, allowing for a much wider range of models. The primary focus of the course will be a set of examples, many from the presenter's ongoing consulting experience, which highlight some of the potential as well as the challenges of such data. More time will be spent on the random effects models, since the standard multiple events approach is now fairly well understood.

Location: The University of Pennsylvania, Houston Hall, 2nd Floor, in the Golkin Room.

Schedule: 

8:30AM to 9:00AM     Registration : Coffee/muffins available

9:00AM to 12:00PM    Morning Session (tea/coffee available at 10:30AM)

12:00PM to 1:30PM    Lunch (on your own)

1:30PM to 5:00PM      Afternoon Session (snacks available at 3PM)

For more details visit http://www.amstatphilly.org/events/spring/2009/short_course/ASAPSpringCourse.pdf