Predictive averaging and Rubin’s rule-based model pooling to predict survival risk with imputations in the presence of missing patient data: methodology and verification using two case studies and simulations
Bart J A Mertens, Lars L J van der Burg, Erwan Donal, Nina Ajmone Marsan, Liesbeth C de Wreede · 2026
Read Predictive averaging and Rubin’s rule-based model pooling to predict survival risk with imputations in the presence of missing patient data: methodology and verification using two case studies and simulations by Bart J A Mertens, Lars L J van der Burg, Erwan Donal, Nina Ajmone Marsan, Liesbeth C de Wreede (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13474813.txt.
Description
Bart J A Mertens and colleagues study in 2026 how to predict patient survival when key data are absent. Missing values plague both the past records used to build a model and the new records it must judge. Two ways to use filled-in data are compared. One averages survival guesses from models built on each filled set, while the other pools model parts first and predicts once.
Two real data sets plus wide simulations tested both routes. Accuracy and bias looked much alike across methods. Gaps did appear in how stable and how honest the uncertainty around guesses was. The authors walk through which route suits registry work and bedside risk tools.
Catalog record
| CREATOR | Bart J A Mertens, Lars L J van der Burg, Erwan Donal, Nina Ajmone Marsan, Liesbeth C de Wreede |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 75623 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |