A0297
Title: Diffusion piecewise exponential models for survival extrapolation using piecewise deterministic Monte Carlo
Authors: Luke Hardcastle - University College London (United Kingdom) [presenting]
Samuel Livingstone - University College London (United Kingdom)
Gianluca Baio - University College London (United Kingdom)
Abstract: The purpose is to introduce the diffusion piecewise exponential model. Piecewise exponential models are a flexible non-parametric approach for time-to-event data, but extrapolation beyond final observation times typically relies on random walk priors and deterministic knot locations, resulting in unrealistic long-term hazards. The diffusion piecewise exponential model is a prior framework that consists of a discretized diffusion for the hazard and can encode a wide variety of information about the long-term behavior of the hazard, with time changed by a Poisson process prior for knot locations. This allows the behavior of the hazard in the observation period to be combined with prior information to inform extrapolations. Efficient posterior sampling is achieved using piecewise deterministic Markov processes, whereby we extend existing approaches using sticky dynamics from sampling spike-and-slab distributions to more general trans-dimensional posteriors. The focus is on applications in health-technology assessment, where the need to compute mean survival requires hazard functions to be extrapolated beyond the observation period, showcasing performance on datasets for Colon cancer patients.
