Sign in to confirm you’re not a bot
This helps protect our community. Learn more

Sarabh Mukhopadhyay - A Novel Bayesian Evaluation of Benefits and Risks for Medical Products -

R in Pharma
3.55K subscribers
42 views 8 months ago
A Novel yet Practical Bayesian Evaluation of Benefits and Risks for Medical Products with a Companion R-Shiny Application - Sarabh Mukhopadhyay Abstract: Evaluation of the benefit-risk profile is a key element throughout the life cycle of a medical product. There is an increasing emphasis on quantitative benefit-risk assessments, as it aims to define explicit, transparent, and evidence-based criteria to compare benefit-risk profiles of different treatments that can inform sponsors, regulators, payers, and other stakeholders at different stages of its entire life cycle. One challenge in implementing quantitative benefit risk is the elicitation of utility functions to map observed outcomes to numeric scores. In this presentation, we first discuss a Bayesian framework that provides an explicit and efficient way to estimate preferences using trade-off data from stakeholders. A key gap remains in how to effectively combine preferences elicited with  …...more
...more
Sarabh Mukhopadhyay - A Novel Bayesian Evaluation of Benefits and Risks for Medical Products -
0Likes
42Views
Nov 252024
A Novel yet Practical Bayesian Evaluation of Benefits and Risks for Medical Products with a Companion R-Shiny Application - Sarabh Mukhopadhyay Abstract: Evaluation of the benefit-risk profile is a key element throughout the life cycle of a medical product. There is an increasing emphasis on quantitative benefit-risk assessments, as it aims to define explicit, transparent, and evidence-based criteria to compare benefit-risk profiles of different treatments that can inform sponsors, regulators, payers, and other stakeholders at different stages of its entire life cycle. One challenge in implementing quantitative benefit risk is the elicitation of utility functions to map observed outcomes to numeric scores. In this presentation, we first discuss a Bayesian framework that provides an explicit and efficient way to estimate preferences using trade-off data from stakeholders. A key gap remains in how to effectively combine preferences elicited with data from clinical trials. Therefore, we next characterize the uncertainty of the performance of a treatment based on clinical trial data. Finally, we discuss how to combine preferences with outcomes of treatment to define an overall benefit-risk utility score of a treatment. The methodology also allows stochastic comparisons between two or more treatments or doses to understand relative preferences and associated uncertainties. This comprehensive approach is also easy to implement in practice and can leverage aggregate-level clinical trial data to estimate treatment performance. We demonstrate the entire process using simulated data and an R-shiny app. Resources mentioned in the talk: Presented at the 2024 R/Pharma Conference

Follow along using the transcript.

R in Pharma

3.55K subscribers