Project Details
Projekt
Assessment of time-varying long-term effects of therapies and prognostic factors (T: Time-varying)
Applicant
Professor Dr. Wilhelm Sauerbrei
Subject Area
Epidemiology and Medical Biometry/Statistics
Term
from 2004 to 2010
Project identifier
Deutsche Forschungsgemeinschaft (DFG) - Project number 5470786
The availability of large datasets with long-term follow-up from different breast cancer studies has opened a controversial discussion on possible time-varying effects and their implication on clinical practice. Several approaches for the modeling of such time-varying effects have been proposed, but up to now, no standard procedure has been established. This project will compare some of these methods and assess their characteristics to establish these methods to practical application. We will consider proposals that build on the semi parametric Cox model, but relax the assumption of proportional hazards (PH). A special focus will lie on fractional polynomials. Other possibilities evaluated will include parametric survival and frailty models. Within breast cancer studies with long-term follow-up we will assess the different approaches. The analyses and interpretation of results will help to better understand the complex mechanisms affecting treatment response and prognosis in the framework of breast cancer.
DFG Programme
Research Units
