Dynamic clinical prediction models for discrete time‐to‐event data with competing risks—A case study on the OUTCOMEREA database

2018 ◽  
Vol 61 (3) ◽  
pp. 514-534 ◽  
Author(s):  
Rachel Heyard ◽  
Jean‐François Timsit ◽  
Wafa Ibn Essaied ◽  
Leonhard Held ◽  
Keyword(s):  
Competing Risks ◽  
Discrete Time ◽  
Prediction Models ◽  
Event Data ◽  
Clinical Prediction ◽  
Time To Event ◽  
Clinical Prediction Models ◽  
Time To Event Data

scholarly journals Modeling potential time to event data with competing risks

2013 ◽  
Vol 20 (2) ◽  
pp. 316-334 ◽  
Author(s):  
Liang Li ◽  
Bo Hu ◽  
Michael W. Kattan
Keyword(s):  
Competing Risks ◽  
Event Data ◽  
Time To Event ◽  
Time To Event Data

scholarly journals A nonparametric updating method to correct clinical prediction model drift

2019 ◽  
Vol 26 (12) ◽  
pp. 1448-1457 ◽  
Author(s):  
Sharon E Davis ◽  
Robert A Greevy ◽  
Christopher Fonnesbeck ◽  
Thomas A Lasko ◽  
Colin G Walsh ◽  
...  
Keyword(s):  
Prediction Models ◽  
Testing Procedure ◽  
Learning Approaches ◽  
Department Of Veterans Affairs ◽  
Clinical Prediction ◽  
Clinical Prediction Models ◽  
Subsequent Performance ◽  
Nonparametric Models ◽  
Population Shift

Abstract Objective Clinical prediction models require updating as performance deteriorates over time. We developed a testing procedure to select updating methods that minimizes overfitting, incorporates uncertainty associated with updating sample sizes, and is applicable to both parametric and nonparametric models. Materials and Methods We describe a procedure to select an updating method for dichotomous outcome models by balancing simplicity against accuracy. We illustrate the test’s properties on simulated scenarios of population shift and 2 models based on Department of Veterans Affairs inpatient admissions. Results In simulations, the test generally recommended no update under no population shift, no update or modest recalibration under case mix shifts, intercept correction under changing outcome rates, and refitting under shifted predictor-outcome associations. The recommended updates provided superior or similar calibration to that achieved with more complex updating. In the case study, however, small update sets lead the test to recommend simpler updates than may have been ideal based on subsequent performance. Discussion Our test’s recommendations highlighted the benefits of simple updating as opposed to systematic refitting in response to performance drift. The complexity of recommended updating methods reflected sample size and magnitude of performance drift, as anticipated. The case study highlights the conservative nature of our test. Conclusions This new test supports data-driven updating of models developed with both biostatistical and machine learning approaches, promoting the transportability and maintenance of a wide array of clinical prediction models and, in turn, a variety of applications relying on modern prediction tools.

scholarly journals Marginal Models for Clustered Time-to-Event Data with Competing Risks Using Pseudovalues

Biometrics ◽  
2010 ◽  
Vol 67 (1) ◽  
pp. 1-7 ◽  
Author(s):  
Brent R. Logan ◽  
Mei-Jie Zhang ◽  
John P. Klein
Keyword(s):  
Competing Risks ◽  
Event Data ◽  
Marginal Models ◽  
Time To Event ◽  
Time To Event Data

Updating Clinical Prediction Models: An Illustrative Case Study

2021 ◽  
pp. 109-113
Author(s):  
Hendrik-Jan Mijderwijk ◽  
Stefan van Beek ◽  
Daan Nieboer
Keyword(s):  
Prediction Models ◽  
Illustrative Case ◽  
Clinical Prediction ◽  
Clinical Prediction Models ◽  
Illustrative Case Study
Sign in / Sign up

Export Citation Format

Share Document