ICH Q14, ANALYTICAL PROCEDURE DEVELOPMENT, describes science and risk-based approaches for developing and maintaining analytical procedures. In this session, the usefulness of several statistical methods as key enablers of these goals are shown by example. These examples display the how the powerful combination of statistics, risk management and scientific understanding can most efficiently and accurately develop robust methods (Design of Experiments), identify the relative influence of multiple sources of variability (variance components analysis), and most quickly identify changes in the analytical process (statistical process control). Using data visualization for improvement and acceleration of root cause investigations ("is it the lab or the process?" ) will also be shown. In addition to the method examples, the nuances of pharmaceutical analytical data when applying these methods will be described.
Learning Objectives:
Describe how statistical methods enable adoption of the principles of ICH Q14.
List 4 common statistical methods that can be leveraged across the analytical procedure lifecycle
Describe specific robustness and efficiency gains that are possible when design of experiments are leveraged
Describe statistical methods that can be applied to current anaytical validation procedures to gain addtional information
Describe how control charts enable a prospective approach to ongoing monitoring and accelerate root cause investigations