How to Address Uncertainty in Health Economic Discrete-Event Simulation Models: An Illustration for Chronic...

Area of Science:

• Health Economics
• Computational Modeling
• Biostatistics

Background:

• Personalized treatment evaluation necessitates health economic models incorporating patient characteristics.
• Patient-level discrete-event simulation (DES) models are suitable for simulating diverse patient profiles and treatment pathways.
• A scarcity of published DES models and missing methodological details hinder their application.

Purpose of the Study:

• To describe challenges in developing DES models, focusing on heterogeneity and uncertainty (structural, stochastic, parameter).
• To provide guidance on correctly implementing uncertainty in DES models.
• To illustrate the impact of modeling choices using a chronic obstructive pulmonary disease (COPD) case study.

Main Methods:

• Description of four key challenges in DES model development: heterogeneity, structural, stochastic, and parameter uncertainty.
• Explanation of the importance and correct implementation of these uncertainty types.
• Application of a DES model for COPD to demonstrate the impact of modeling choices.

Main Results:

• Correct implementation of model uncertainty showed a hypothetical intervention could be cost-effective.
• Incorrect modeling of uncertainty led to a 50% to 14-fold increase in the incremental cost-effectiveness ratio.
• Inaccurate uncertainty modeling resulted in an extended life expectancy of 1.4 years and vastly increased outcome uncertainty.

Conclusions:

• Guidance is provided for implementing uncertainty in DES models, enhancing transparency in reporting methods.
• The COPD case study effectively illustrates the practical implications of uncertainty modeling.
• Availability of R code and pseudo-code facilitates understanding and replication for other model developers.