Nonclinical statistics groups assist pharmaceutical development scientists and engineers with experimentation, experimental design, data analysis and results interpretation.  Towards this end, members of the nonclinical statistics group may employ a range of methods including mathematical, statistical, experimental design, pattern recognition, optimization, simulation and structure-property estimation methods.  Specifically, members work with chemical process scientists and engineers to design and analyze experiments directed at obtaining the best yield of production processes for making bulk drug with high purity.  Additionally, projects are undertaken with pharmaceutical scientists and engineers to design and analyze experiments to help formulate a drug active into an appropriate dosage form. 

 

Inference for QbD enables members of nonclinical statistics groups to focus on their highest-value contribution: the creation and verification of new experimental design and analysis strategies.

 

Product R&D IT is not only challenged with getting more information to the scientists and engineers faster: the information must be of higher quality. To achieve peak productivity, scientists and engineers need accurate and readily accessible information so that they can make critical decisions quickly.  One of the biggest contributors to R&D costs and delays is poor decision making caused by poor R&D information visibility resulting in, contributes to upwardly spiraling R&D costs, and contributes to a slide in R&D productivity.

 

Inference for QbD can reverse the tide of spiraling R&D costs and hampered productivity by enabling the Pharmaceutical Product R&D IT group to: