Fully nonparametric survival analysis in the presence of time-dependent covariates and dependent censoring

Abstract

In the presence of informative right censoring and time-dependent covariates, we estimate the survival function in a fully nonparametric fashion. We introduce a novel method for incorporating multiple observations per subject when estimating the survival function at different covariate values and compare several competing methods via simulation. The proposed method is applied to survival data from people awaiting liver transplant.

Keywords:

• Survival analysis
• Kaplan–Meier estimation
• inverse probability censoring weighting
• dependent censoring
• organ transplantation

Disclosure statement

The authors report no potential conflict of interest.

Notes

1 In a simulation documented in the Appendix, we explore what happens when we relax this assumption by allowing individuals to have their own survival functions drawn from a common distribution.

2 The (z = 2, t = 90) case was actually based on 497 replicates because 3 had an undefined survival estimate, which happens whenever the last subject is censored before t = 90.

3 MELD is an acronym for ‘model for end-stage liver disease’.

4 In some circumstances, the MELD score is known to not accurately convey the urgency with which one needs a transplant; priority for these candidates is determined in a different manner that we do not discuss here.

Additional information

Funding

This work was supported in part by grant number R01DK111233 from the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK).