DIA 2017 Annual Meeting

Component Type: Workshop
Level: Basic
CE: CME 1.25; IACET 1.25; RN 1.25

In drug development, we are regularly required to make decisions (e.g. whether to progress a new molecular entity or new biologic to the next phase of development or what doses of a compound to test) or to design a clinical trial in the face of imperfect evidence about the efficacy and safety of the asset.This can result in making incorrect decisions or assumptions which can lead to failure, false conclusions and/or wasted resources. In these situations, one option is to ask the experts what they believe, and to formally quantify these beliefs and the associated uncertainty.

Prior elicitation is a scientific, structured methodology which to translate expert knowledge and judgment into a quantitative belief distribution that captures the best estimate of what is known “right now” about the true value of the unknown quantity(s) (e.g. true treatment effect of a particular asset), as well as quantifying the uncertainty around the estimate of the true value. The resulting belief distribution can then be used to calculate the probability of success of a new study(s), aid in the evaluation and development of study design and/or integrate into the interpretation of study results (i.e. Bayesian approach).

In this workshop, participants will be introduced to the concepts of prior elicitation, the Sheffield Elicitation Framework (O’Hagan and Oakley) and some of the benefits and challenges of conducting an elicitation. The concepts will be further illustrated through participation in a mock elicitation about a hypothetical new drug.

Learning Objectives

Discuss a structured methodology (Prior Elicitation) to improve the development of reliable assumptions based on expert knowledge; Describe what a prior elicitation is and the benefits and challenges (which will be illustrated through participating in mock elicitation); Identify where a prior elicitation may have value in their work.

Chair

Timothy H Montague

Speaker

What Do the Experts Believe? Translating Expert Knowledge and Judgment Into a Quantitative Belief Distribution
Timothy H Montague

Facilitator
Michael Sonksen