Header

UZH-Logo

Maintenance Infos

Survival forests under test: Impact of the proportional hazards assumption on prognostic and predictive forests for amyotrophic lateral sclerosis survival


Korepanova, Natalia; Seibold, Heidi; Steffen, Verena; Hothorn, Torsten (2020). Survival forests under test: Impact of the proportional hazards assumption on prognostic and predictive forests for amyotrophic lateral sclerosis survival. Statistical Methods in Medical Research, 29(5):1403-1419.

Abstract

We investigate the effect of the proportional hazards assumption on prognostic and predictive models of the survival time of patients suffering from amyotrophic lateral sclerosis. We theoretically compare the underlying model formulations of several variants of survival forests and implementations thereof, including random forests for survival, conditional inference forests, Ranger, and survival forests with L<jats:sub>1</jats:sub> splitting, with two novel variants, namely distributional and transformation survival forests. Theoretical considerations explain the low power of log-rank-based splitting in detecting patterns in non-proportional hazards situations in survival trees and corresponding forests. This limitation can potentially be overcome by the alternative split procedures suggested herein. We empirically investigated this effect using simulation experiments and a re-analysis of the Pooled Resource Open-Access ALS Clinical Trials database of amyotrophic lateral sclerosis survival, giving special emphasis to both prognostic and predictive models.

Abstract

We investigate the effect of the proportional hazards assumption on prognostic and predictive models of the survival time of patients suffering from amyotrophic lateral sclerosis. We theoretically compare the underlying model formulations of several variants of survival forests and implementations thereof, including random forests for survival, conditional inference forests, Ranger, and survival forests with L<jats:sub>1</jats:sub> splitting, with two novel variants, namely distributional and transformation survival forests. Theoretical considerations explain the low power of log-rank-based splitting in detecting patterns in non-proportional hazards situations in survival trees and corresponding forests. This limitation can potentially be overcome by the alternative split procedures suggested herein. We empirically investigated this effect using simulation experiments and a re-analysis of the Pooled Resource Open-Access ALS Clinical Trials database of amyotrophic lateral sclerosis survival, giving special emphasis to both prognostic and predictive models.

Statistics

Citations

Dimensions.ai Metrics
4 citations in Web of Science®
4 citations in Scopus®
Google Scholar™

Altmetrics

Additional indexing

Item Type:Journal Article, refereed, original work
Communities & Collections:04 Faculty of Medicine > Epidemiology, Biostatistics and Prevention Institute (EBPI)
Dewey Decimal Classification:610 Medicine & health
Scopus Subject Areas:Health Sciences > Epidemiology
Physical Sciences > Statistics and Probability
Health Sciences > Health Information Management
Uncontrolled Keywords:Statistics and Probability, Health Information Management, Epidemiology
Language:English
Date:1 May 2020
Deposited On:25 Jan 2021 15:41
Last Modified:26 Jan 2021 21:01
Publisher:Sage Publications
ISSN:0962-2802
OA Status:Closed
Publisher DOI:https://doi.org/10.1177/0962280219862586
PubMed ID:31304888
Project Information:
  • : FunderSNSF
  • : Grant ID200021_184603
  • : Project TitleA Lego System for Transformation Inference
  • : FunderSNSF
  • : Grant ID100015-100276
  • : Project TitleEine neue deutsche Geschichte der Literatur Polens. Externe Perspektiven und Nationalliteratur.

Download

Full text not available from this repository.
View at publisher

Get full-text in a library