Exploring structural uncertainty in model-based economic evaluations.

Area of Science:

• Health economics
• Decision analysis
• Mathematical modeling

Background:

• Uncertainty assessment is crucial for model-based evaluations in decision-making.
• Structural uncertainty in decision analytic models is often overlooked in national funding body guidelines, such as the Australian Pharmaceutical Benefits Advisory Committee (PBAC).
• Lack of characterization of structural uncertainty can increase decision-making uncertainty and affect funding decisions.

Purpose of the Study:

• To summarize key elements of structural uncertainty, explaining its importance and methods for characterization.
• To review alternative approaches for characterizing structural uncertainty.
• To advocate for the consideration of structural uncertainty in submissions to national funding bodies.

Main Methods:

• Literature review and synthesis of existing approaches.
• Discussion of five alternative methods for characterizing structural uncertainty: scenario analysis, model selection, model averaging, parameterization, and discrepancy.
• Analysis of current guidelines from national funding bodies regarding structural uncertainty.

Main Results:

• Structural uncertainty significantly impacts model predictions and decision-making.
• Current guidelines from national funding bodies lack detailed descriptions and evaluations of structural uncertainty.
• Five distinct methods for characterizing structural uncertainty have been identified and discussed.

Conclusions:

• Characterizing structural uncertainty is essential for robust model-based evaluations.
• National funding body guidelines need to be updated to include detailed approaches for assessing structural uncertainty.
• Further empirical research is required to determine the most effective methods for structural sensitivity analysis in applied decision-making.