Survival modeling techniques are commonly used to extrapolate clinical trial outcomes like overall survival to a time horizon that is appropriate for health economic evaluations. Standard parametric distributions, such as the exponential and Weibull, have been the de-facto standard for conducting such extrapolations but, with the advent of novel potentially curative therapies, these standard parametric distributions fail to capture the underlying survival trend. Newer techniques like parametric mixture, mixture-cure, and non-mixture cure models are among novel ways to capture these more complex survival patterns. In addition, the incorporation of external evidence has gained prominence. The purpose of this course is to enable participants to identify which methods are most appropriate in a specific context, considering underlying structural assumptions, and discuss how modeling choices propagate into health economic evaluations. To gain a more in-depth understanding of the impact of the choice for a specific method, there will be walkthroughs of exercises which participants will be able to practice in their own time.

PREREQUISITE: Participants must have a basic understanding of Kaplan-Meier curves, standard survival modeling techniques, such as a Weibull distribution in a partitioned survival framework, and R in order to follow along with the exercise walkthroughs.