Trusting the Results of Model-Based Economic Analyses: Is there a Pragmatic Validation Solution?

Area of Science:

• Health economics
• Decision science
• Health technology assessment

Background:

• Decision-analytic models are essential for health technology assessment due to limitations in clinical trial data for long-term impact and subgroup analyses.
• Despite decades of use, decision-makers frequently encounter poorly validated models, leading to impaired trust in their results.
• Challenges include the artificial nature of models hindering validation, complexity, data scarcity, and resource constraints.

Purpose of the Study:

• To address the critical issue of inadequate model validation in health technology assessment.
• To highlight the deficiencies in current validation and transparency practices.
• To propose requirements for improvement and pragmatic solutions for trustworthy models.

Main Methods:

• The article synthesizes perspectives and experiences from authors with diverse backgrounds in health technology assessment.
• It involves characterizing the current challenges in model validation.
• It outlines requirements for enhanced validation and transparency.

Main Results:

• Current health technology assessment models often suffer from poor validation, impacting decision-maker trust.
• Key issues identified include model artificiality, complexity, data limitations, and resource constraints.
• There is a recognized deficiency in validation and transparency practices within the field.

Conclusions:

• Improving the validation and transparency of decision-analytic models is crucial for reliable health technology assessment.
• Addressing the identified challenges requires a concerted effort to develop pragmatic solutions.
• Enhanced validation will foster greater trust and confidence in the use of models for healthcare decision-making.