Models and Tools

Scan selected model-based analyses. Explore modeling methods and tools.

Image of Abstract Art with "Report" Label

Featured: Models and Guidelines

This article describes the rationale for using decision models to support the development of evidence-based clinical preventive service guidelines by the U.S. Preventive Services Task Force.

Image of Abstract Art with "Article" Label

Featured: Bayesian Calibration

This article provides a tutorial on Bayesian approaches for model calibration, describing the theoretical basis for Bayesian calibration approaches as well as pragmatic considerations.

FREQUENTLY ASKED QUESTIONS

Why are models used in decision analysis?

Models are used to understand an underlying process that may be unobserved, or to extrapolate current evidence over a longer time frame. Decisions may affect people for years to come whereas most clinical trials only run for a couple of years–modeling allows current evidence to be projected into the future.

What is a markov model?

Used in many disciplines, markov models describe a process where the probability of moving from one state of being to another only depends on your current state–in other words, history doesn’t matter. In health, markov models are typically used to understand disease progression over time.

What is model calibration?

Model calibration is a process in which we refine model parameters (i.e., inputs) so that our model output matches up with other data – so it gives us output that we expect to see. In calibration we force our model to align with empiric data describing the underlying process to make it more accurate.

Maternal Mortality Estimates

Modeling Cancer Incidence

Featured: Models and Guidelines

Featured: Bayesian Calibration