Time-to-event (survival) analysis is an important element in many economic analyses of health care technologies. This is particularly true in oncology given the requirement to estimate lifetime costs and outcomes (i.e. extrapolate) beyond the follow-up typically observed in clinical trials. Cost-effectiveness estimates can be sensitive to the methods applied in modelling survival data. Recommendations for selecting a parametric survival model have been recently been published; following a review of extrapolation modelling in National Institute for Health and Care Excellence (NICE) technology appraisals.
The purpose of this course is to provide participants with an understanding of the fundamentals of survival analysis and key issues to be considered when comparing alternative survival models for inclusion in cost-effectiveness analysis. This will include an understanding of the differences between partitioned survival and Markov-based approaches. The course runs 2 days for 2 hrs/day.