Incorporating External Evidence Into Extrapolations: A Debate on the Trade-Offs Between Complexity and Simplicity

ISSUE: Producing plausible lifetime extrapolations is fundamental to appropriate decision-making within health technology assessment (HTA). However, the pipeline of novel therapies relying on surrogate outcomes and immature survival data continues to grow, creating challenges for modelers estimating lifetime outcomes for HTA. These challenges have led to an increase in proposed methods designed to improve extrapolation using external data such as real-world evidence (RWE) and expert elicitation. While HTA entities such as the National Institute for Health and Care Excellence (NICE) have endorsed the use of RWE and expert elicitation, modelers are left with multiple choices and little guidance on when and how these methods may be used in practice. Furthermore, the use of increasingly complex methods combined with a lack of research comparing performance across methods may lead to further scepticism of HTA results.

OVERVIEW: This panel will discuss the trade-offs between the benefits gained through the use of advanced extrapolation methods leveraging external evidence, and simplistic and transparent methods for HTA decision making. R. Brett McQueen will moderate and review the issue based on the current state of the literature. Bart Heeg will represent the perspective of a methods developer in the survival extrapolation field and articulate reasons for using more advanced methods. Dawn Lee will represent the perspective of a model reviewer for NICE submissions and articulate the problems that increasingly complexity causes, the benefits of keeping things simple where possible and the need for transparency and consideration of the manner in which data is presented in survival extrapolation methods. Finally, Ash Bullement will present the perspective of an advisor and consultant developing submissions to HTA agencies on behalf of manufacturers. All three panellists will offer contrasting perspectives on the trade-offs between simplicity and complexity of survival extrapolations for HTA decision making.