Variability and uncertainty: interpretation and usage of pharmacometric simulations and intervals

Pharmacometric modeling and simulation are being used increasingly in all stages of drug development. The simulations can produce a variety of probabilistic intervals representing different combinations of variability and uncertainty. Pharmacometricians are often unclear on when would be best to use which interval. The problem is exacerbated by often-unrealized confusions on the concepts of confidence interval and prediction interval among pharmacometricians. Consequently, pharmacometric intervals may often be inappropriately constructed or misinterpreted, and drug development decisions may be jeopardized.This talk aims to clarify some of the important underlying concepts and the appropriate pharmacometric simulation focus, develop a classification of common simulation applications and properties, and provide a practically convenient guideline on simulation conduct and interval interpretation. Some common misconceptions and misusages will also be discussed.