Model Transparency and Validation
This guideline explains how to make health care models trustworthy by being open about how they are built (transparency) and checking that they reflect real-world results (validation). It gives practical advice on what information to share and how to test if a model works correctly.
At a glance
Use when
Developing or reviewing health economic models, conducting HTA submissions, designing model documentation, seeking regulatory or stakeholder acceptance
Avoid when
Working outside health care modeling contexts where model credibility is not scrutinized, or when guidelines are superseded by newer standards
Inputs
Model structure, data sources, assumptions, intended use, funding sources
Outputs
Transparent documentation (nontechnical and technical), validated model performance against real-world or alternative model results
How it works
This report from the ISPOR-SMDM Modeling Good Research Practices Task Force provides structured recommendations for enhancing transparency and validation in health care modeling. It defines transparency as full disclosure of model type, structure, data sources, inputs, outputs, assumptions, and limitations, accessible to non-experts and experts alike. For validation, it outlines five types: face validity, verification (internal validity), cross validity, external validity, and predictive validity—with the latter two considered strongest. It recommends detailed technical documentation sufficient for expert evaluation or reproduction, accessible either openly or under IP-protecting agreements.
- HTA domains
- Clinical Effectiveness, Costs & Economic Evaluation, Organisational aspects
- Categories
- Model ValidationTransparency
- Assumptions
- Modelers aim to build credible models; stakeholders require access to information to assess trustworthiness; some models may require controlled access due to intellectual property
- Strengths
- Provides comprehensive, consensus-based standards for model credibility; supports reproducibility and peer review; balances openness with IP concerns; applicable across modeling types and HTA contexts
- Limitations
- Does not enforce compliance; implementation depends on modeler willingness; may be challenging for proprietary models to fully disclose
- Also known as
- ISPOR-SMDM Task Force Report 7, Model Transparency and Validation Report
Questions this answers
- › What makes a health care model trustworthy?
- › How can model transparency be achieved?
- › What types of validation should be performed on a model?
- › What information should be included in non-technical and technical model documentation?
- › How can modelers balance transparency with intellectual property protection?
- › What are best practices for verifying and validating model accuracy?
References & sources
Related methods
Similar by meaning
Beta record. Generated from the primary source via AI extraction and independent audit, pending final human review.