Better Decision Making in Drug Development Through Adoption of Formal Prior Elicitation
Nigel Dallow, Nicky Best, Timothy Montague

TL;DR
This paper discusses how GSK integrates formal expert elicitation methods, specifically the SHELF framework, into drug development to improve decision making, transparency, and robustness of prior distributions in Bayesian analyses.
Contribution
It presents a structured approach and real-world case studies demonstrating the integration of prior elicitation within a large pharmaceutical company's decision processes.
Findings
Elicitation provides a quantitative basis for decision making.
It enhances transparency and understanding of risks.
Supports better portfolio and project decisions.
Abstract
With the continued increase in the use of Bayesian methods in drug development, there is a need for statisticians to have tools to develop robust and defensible informative prior distributions. Whilst relevant empirical data should, where possible, provide the basis for such priors, it is often the case that limitations in data and/or our understanding may preclude direct construction of a data-based prior. Formal expert elicitation methods are a key technique that can be used to determine priors in these situations. Within GlaxoSmithKline (GSK), we have adopted a structured approach to prior elicitation based on the SHELF elicitation framework, and routinely use this in conjunction with calculation of probability of success (assurance) of the next study(s) to inform internal decision making at key project milestones. The aim of this paper is to share our experiences of embedding the…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
