Obstructive lung disease models: what is valid? | JoVE Visualize

Area of Science:

Health outcomes research
Computational epidemiology
Biostatistics

Background:

Disease simulation models are increasingly used to evaluate health outcomes and inform public health interventions.
There is a growing demand for evidence demonstrating the credibility and accuracy of these simulation models.
Scrutiny of model methodologies necessitates a clear understanding of their validation processes.

Purpose of the Study:

To systematically review recent obstructive lung disease simulation models and their validation strategies.
To assess the types and extent of validation employed in these models.
To identify challenges and propose future directions for improving model validation.

Main Methods:

A systematic literature search was conducted using Medline and EMBASE for obstructive lung disease simulation models published between January 2000 and June 2006.
Reviewed publications to assess model attributes and categorize validation into first-order (verification), second-order (internal validation), third-order (external validation), and predictive validity.
Analyzed validation endpoints such as disease prevalence, exacerbation rates, and all-cause mortality.

Main Results:

Identified six asthma and seven chronic obstructive pulmonary disease (COPD) simulation models.
Over half of the models (54%) reported second-order and third-order validation, often comparing simulated outcomes to source study cohorts or independent studies.
Validation was frequently described as acceptable, yet criteria for judging adequacy were largely absent, and reporting was inconsistent.

Conclusions:

While over half of recent obstructive lung disease simulation models report validation, inconsistencies in methods and reporting impede thorough assessment.
Standardized validation criteria and quantitative reporting are crucial for establishing simulation models as reliable tools for clinical and public health decision-making.
Further development is needed to define required validation levels and provide guidance for quantitative assessment to promote simulation models as practical decision tools.