A comparison of the conditional inference survival forest model to random survival forests based on a simulation study as well as on two applications with time-to-event data

Justine B Nasejje,Maia Lesosky,Keertan Dheda,Henry Mwambi

doi:10.1186/s12874-017-0383-8

Justine B Nasejje, Maia Lesosky + Show 2 more

Open Access

https://doi.org/10.1186/s12874-017-0383-8

Copy DOI

Abstract

BackgroundRandom survival forest (RSF) models have been identified as alternative methods to the Cox proportional hazards model in analysing time-to-event data. These methods, however, have been criticised for the bias that results from favouring covariates with many split-points and hence conditional inference forests for time-to-event data have been suggested. Conditional inference forests (CIF) are known to correct the bias in RSF models by separating the procedure for the best covariate to split on from that of the best split point search for the selected covariate.MethodsIn this study, we compare the random survival forest model to the conditional inference model (CIF) using twenty-two simulated time-to-event datasets. We also analysed two real time-to-event datasets. The first dataset is based on the survival of children under-five years of age in Uganda and it consists of categorical covariates with most of them having more than two levels (many split-points). The second dataset is based on the survival of patients with extremely drug resistant tuberculosis (XDR TB) which consists of mainly categorical covariates with two levels (few split-points).ResultsThe study findings indicate that the conditional inference forest model is superior to random survival forest models in analysing time-to-event data that consists of covariates with many split-points based on the values of the bootstrap cross-validated estimates for integrated Brier scores. However, conditional inference forests perform comparably similar to random survival forests models in analysing time-to-event data consisting of covariates with fewer split-points.ConclusionAlthough survival forests are promising methods in analysing time-to-event data, it is important to identify the best forest model for analysis based on the nature of covariates of the dataset in question.

Highlights

Random survival forest (RSF) models have been identified as alternative methods to the Cox proportional hazards model in analysing time-to-event data
There are some unique differences in the predictive performance between two random survival forests and the Conditional inference forests (CIF) model which can not be ignored
The prediction error values for the CIF model appear to be at par with those of RSF2 on Data 1, the model has the lowest error values compared to RSF1 and RSF2 on the remaining five datasets

Summary

Introduction

Random survival forest (RSF) models have been identified as alternative methods to the Cox proportional hazards model in analysing time-to-event data. These methods, have been criticised for the bias that results from favouring covariates with many split-points and conditional inference forests for time-to-event data have been suggested. Survival trees and random survival forests (RSF) are an attractive alternative approach to the Cox proportional hazards models when the PH assumption is violated [8] These methods are extensions of classification and regression trees and random forests (RF) [9, 10] for time-to-event data. Conditional inference forests (CIF) are known to reduce this selection bias by separating the algorithm for selecting the best covariate to split on from that of the best split point search [15, 17, 18]

Methods

Results

Conclusion