Health economic models play a vital role in product development, providing valuable information and insights from early planning and portfolio selection to pricing analysis. Building models is complex, requiring integration of evidence from multiple sources, including results from de novo statistical analyses, into a computational framework that allows users to modify inputs and review results. The resulting modeling framework must have face validity, avoid programming errors and produce valid forecasts. Standardizing and automating model building and related statistical analyses facilitates faster model development, enhances quality and reliability, and expands accessibility to the development process.

In this presentation, we will explore approaches to automating model-building and required statistical analyses and introduce an innovative modeling platform. This platform automates model generation for a broad range of therapeutic areas and model types using a wizard for capturing model specifications and creating customized models while minimizing errors. Once the generated model is populated with inputs, the platform can run the model on a PC or a server-based environment to optimize run-times. Automation of the statistical analyses that inform model inputs will also be described and illustrated with an application for parametric fitting of time to event equations. Finally, these automation tools will be demonstrated to the audience by using the platform.

During the Q&A period, the advantages and challenges of automated model building and contrasts to AI-based model generation will be discussed.

Discover how an automated model-building tool can significantly expedite the creation of early health economic models that adhere to good practice guidelines and minimize errors. Furthermore, learn how automation tools leverage the computational power of server-based installations while preserving the Excel-based nature of models preferred by stakeholders.

Sponsor: Evidera