Literature DB >> 35227445

A Framework for Using Real-World Data and Health Outcomes Modeling to Evaluate Machine Learning-Based Risk Prediction Models.

Patricia J Rodriguez1, David L Veenstra2, Patrick J Heagerty3, Christopher H Goss4, Kathleen J Ramos5, Aasthaa Bansal6.   

Abstract

OBJECTIVES: We propose a framework of health outcomes modeling with dynamic decision making and real-world data (RWD) to evaluate the potential utility of novel risk prediction models in clinical practice. Lung transplant (LTx) referral decisions in cystic fibrosis offer a complex case study.
METHODS: We used longitudinal RWD for a cohort of adults (n = 4247) from the Cystic Fibrosis Foundation Patient Registry to compare outcomes of an LTx referral policy based on machine learning (ML) mortality risk predictions to referral based on (1) forced expiratory volume in 1 second (FEV1) alone and (2) heterogenous usual care (UC). We then developed a patient-level simulation model to project number of patients referred for LTx and 5-year survival, accounting for transplant availability, organ allocation policy, and heterogenous treatment effects.
RESULTS: Only 12% of patients (95% confidence interval 11%-13%) were referred for LTx over 5 years under UC, compared with 19% (18%-20%) under FEV1 and 20% (19%-22%) under ML. Of 309 patients who died before LTx referral under UC, 31% (27%-36%) would have been referred under FEV1 and 40% (35%-45%) would have been referred under ML. Given a fixed supply of organs, differences in referral time did not lead to significant differences in transplants, pretransplant or post-transplant deaths, or overall survival in 5 years.
CONCLUSIONS: Health outcomes modeling with RWD may help to identify novel ML risk prediction models with high potential real-world clinical utility and rule out further investment in models that are unlikely to offer meaningful real-world benefits.
Copyright © 2021 International Society for Pharmacoeconomics and Outcomes Research, Inc. Published by Elsevier Inc. All rights reserved.

Entities:  

Keywords:  machine learning; microsimulation; real-world data

Mesh:

Year:  2021        PMID: 35227445      PMCID: PMC9311314          DOI: 10.1016/j.jval.2021.11.1360

Source DB:  PubMed          Journal:  Value Health        ISSN: 1098-3015            Impact factor:   5.101


  48 in total

Review 1.  Risk prediction models: II. External validation, model updating, and impact assessment.

Authors:  Karel G M Moons; Andre Pascal Kengne; Diederick E Grobbee; Patrick Royston; Yvonne Vergouwe; Douglas G Altman; Mark Woodward
Journal:  Heart       Date:  2012-03-07       Impact factor: 5.994

Review 2.  Traditional statistical methods for evaluating prediction models are uninformative as to clinical value: towards a decision analytic framework.

Authors:  Andrew J Vickers; Angel M Cronin
Journal:  Semin Oncol       Date:  2010-02       Impact factor: 4.929

3.  Super learner.

Authors:  Mark J van der Laan; Eric C Polley; Alan E Hubbard
Journal:  Stat Appl Genet Mol Biol       Date:  2007-09-16

4.  Prognosis and prognostic research: application and impact of prognostic models in clinical practice.

Authors:  Karel G M Moons; Douglas G Altman; Yvonne Vergouwe; Patrick Royston
Journal:  BMJ       Date:  2009-06-04

5.  Referral to lung transplantation--too little, too late.

Authors:  Shawn D Aaron; Cecilia Chaparro
Journal:  J Cyst Fibros       Date:  2016-03       Impact factor: 5.482

6.  Machine learning and artificial intelligence research for patient benefit: 20 critical questions on transparency, replicability, ethics, and effectiveness.

Authors:  Sebastian Vollmer; Bilal A Mateen; Gergo Bohner; Franz J Király; Rayid Ghani; Pall Jonsson; Sarah Cumbers; Adrian Jonas; Katherine S L McAllister; Puja Myles; David Granger; Mark Birse; Richard Branson; Karel G M Moons; Gary S Collins; John P A Ioannidis; Chris Holmes; Harry Hemingway
Journal:  BMJ       Date:  2020-03-20

7.  Decision curve analysis: a novel method for evaluating prediction models.

Authors:  Andrew J Vickers; Elena B Elkin
Journal:  Med Decis Making       Date:  2006 Nov-Dec       Impact factor: 2.583

8.  Cystic fibrosis physicians' perspectives on the timing of referral for lung transplant evaluation: a survey of physicians in the United States.

Authors:  Kathleen J Ramos; Ranjani Somayaji; Erika D Lease; Christopher H Goss; Moira L Aitken
Journal:  BMC Pulm Med       Date:  2017-01-19       Impact factor: 3.317

9.  Bridging the survival gap in cystic fibrosis: An investigation of lung transplant outcomes in Canada and the United States.

Authors:  Anne L Stephenson; Kathleen J Ramos; Jenna Sykes; Xiayi Ma; Sanja Stanojevic; Bradley S Quon; Bruce C Marshall; Kristofer Petren; Joshua S Ostrenga; Aliza K Fink; Albert Faro; Alexander Elbert; Cecilia Chaparro; Christopher H Goss
Journal:  J Heart Lung Transplant       Date:  2020-12-07       Impact factor: 10.247

10.  Key challenges for delivering clinical impact with artificial intelligence.

Authors:  Christopher J Kelly; Alan Karthikesalingam; Mustafa Suleyman; Greg Corrado; Dominic King
Journal:  BMC Med       Date:  2019-10-29       Impact factor: 8.775

View more
  1 in total

1.  The lung allocation score and other available models lack predictive accuracy for post-lung transplant survival.

Authors:  Jay M Brahmbhatt; Travis Hee Wai; Christopher H Goss; Erika D Lease; Christian A Merlo; Siddhartha G Kapnadak; Kathleen J Ramos
Journal:  J Heart Lung Transplant       Date:  2022-05-20       Impact factor: 13.569

  1 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.