Leiden University Scholarly Publications

Persistent URL of this record https://hdl.handle.net/1887/3304235

In Collections

This item can be found in the following collections:

Geloven, N. van; Giardiello, D.; Bonneville, E.F.; Teece, L.; Ramspek, C.L.; Smeden, M. van; ... ; Steyerberg, E. (2022)

Validation of prediction models in the presence of competing risks: a guide through modern methods

Article / Letter to editor
Thorough validation is pivotal for any prediction model before it can be advocated for use in medical  practice. For time-to-event outcomes such as breast cancer recurrence, death from other causes is a competing risk. Model performance measures must account for such competing events. In this article, we present a comprehensive yet accessible overview of performance measures for this competing event
setting, including the calculation and interpretation of statistical measures for calibration, discrimination, overall prediction error, and clinical usefulness by decision curve analysis. All methods are illustrated for patients with breast cancer, with publicly available data and R code.
All authors
Geloven, N. van; Giardiello, D.; Bonneville, E.F.; Teece, L.; Ramspek, C.L.; Smeden, M. van; Snell, K.I.E.; Calster, B. van; Pohar-Perme, M.; Riley, R.D.; Putter, H.; Steyerberg, E.
Date
2022
Journal
BRITISH MEDICAL JOURNAL