Learn more about our 100% virtual Home-Based Pulmonary Rehabilitation Program

Rating the reliability of predictive models and artificial intelligence (AI) in healthcare

Systematic Review of Prediction Models

  • Targeted screening for unrecognized dementia with a prediction model
    Questions: Although the US Preventive Services Task Force does not recommend screening older adults in primary care for dementia, it may be possible to conduct targeted screening in the highest-risk patients (e.g., those with a dementia prevalence ³90th percentile in a health plan’s population). How effectively would KP Washington’s prediction model, eRADAR, improve targeted screening? … Read more

How and why we systematically review models

Northwest Permanente epidemiologists review prediction models to assess their risk of bias and applicability for clinical use by Northwest Permanente physicians. Systematic reviewers assesses models’ risk of bias and applicability with the PROBAST tool.[1] The epidemiologists’ reviews are presented to Northwest Permanente’s Prediction Modeling Governance Committee, which decides which models will be put into practice. The Committee also decides whether models require further validation to assess their bias and applicability for KP Northwest.

Morlen Health posts the epidemiologists’ reviews for other healthcare decision-makers.

1. Moons KGM, Wolff RF, Riley RD, et al. PROBAST: A tool to assess the risk of bias and applicability of prediction model studies: explanation and elaboration. Ann Intern Med 2019;170:W1-W33.