Online Program Home
My Program

Abstract Details

Activity Number: 75
Type: Contributed
Date/Time: Sunday, July 31, 2016 : 4:00 PM to 5:50 PM
Sponsor: Biopharmaceutical Section
Abstract #319275 View Presentation
Title: Shelf Life Estimation: Bayesian Approach
Author(s): Maryna Ptukhina* and Walter Stroup
Companies: and University of Nebraska - Lincoln
Keywords: shelf life

Shelf life estimation procedures, following ICH guidelines, use multiple batch regression with fixed batch effects. This guidance specifically mandates estimates based on at least 3 batches. Shelf life estimates are required to apply to all future batches, which implies that the mixed model approach assuming random batch effects is more appropriate. Several researchers have suggested the use of mixed models; however with only 3 batches variance components are not well estimated. To alleviate that, we implement a fully Bayesian approach, where we get a posterior distribution of shelf life arising from a combination of reasonable prior information and appropriate likelihood. Our preliminary findings are that a lower percentile of this posterior distribution provides the most accurate estimate of shelf life we have found to date. In addition, using the ICH guidance that a shelf life estimate should be such that there is a reasonable likelihood that all future batches will meet acceptance criteria for at least as long as the labeled shelf life, Bayesian approach allows us to generate posterior predictive distributions of future batches and use these distributions to obtain shelf life.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2016 program

Copyright © American Statistical Association