Sensitivity of the discrete-time Kaplan-Meier estimate to nonignorable censoring: Application in a clinical trial

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

Untestable assumptions about the association between survival and censoring times can affect the validity of estimates of the survival distribution including the Kaplan-Meier (KM) nonparametric maximum likelihood estimate (MLE). This paper explores the sensitivity of the KM curve to nonignorable censoring by extending the index of local sensitivity to nonignorability (ISNI; Troxel et al., Statistica Sinica, 14, 1221-1237, 2004; Zhang and Heitjan, Biometrics, 62, 1260-1268, 2006) to the case of a nonparametric survival model. The method involves, first, specifying a coarse-data selection model to describe the association between the failure and censoring processes and then evaluating the slope of the nonparametric survival MLE ordinate with respect to a nonignorability parameter in the neighborhood of the ignorable model. We define the nonparametric MLE of the survival curve for a fixed value of the nonignorability parameter and show in a simulation that ISNI analysis effectively captures local sensitivity to nonignorability. The method measures sensitivity in the sense of identifying functionals of the nonparametric MLE that nonignorability, if present, can affect substantially. We demonstrate the method with an application to a trial comparing mechanical assistance to optimal medical management in the treatment of end-stage heart failure.

Original languageEnglish (US)
Pages (from-to)2998-3010
Number of pages13
JournalStatistics in Medicine
Volume31
Issue number25
DOIs
StatePublished - Nov 10 2012

Fingerprint

Kaplan-Meier Estimate
Likelihood Functions
Nonparametric Maximum Likelihood
Censoring
Maximum Likelihood Estimate
Clinical Trials
Discrete-time
Kaplan-Meier
Survival Distribution
Heart Failure
Ordinate
Survival Model
Curve
Selection Model
Nonparametric Model
Biometrics
Data Model
Slope
Estimate
Demonstrate

Keywords

  • Coarse-data model
  • Ignorability
  • Informative censoring
  • ISNI
  • Sensitivity analysis

ASJC Scopus subject areas

  • Epidemiology
  • Statistics and Probability

Cite this

Sensitivity of the discrete-time Kaplan-Meier estimate to nonignorable censoring : Application in a clinical trial. / Liu, Tao; Heitjan, Daniel F.

In: Statistics in Medicine, Vol. 31, No. 25, 10.11.2012, p. 2998-3010.

Research output: Contribution to journalArticle

@article{d08ab3634cd9426e9dfb7466ce68e3b8,
title = "Sensitivity of the discrete-time Kaplan-Meier estimate to nonignorable censoring: Application in a clinical trial",
abstract = "Untestable assumptions about the association between survival and censoring times can affect the validity of estimates of the survival distribution including the Kaplan-Meier (KM) nonparametric maximum likelihood estimate (MLE). This paper explores the sensitivity of the KM curve to nonignorable censoring by extending the index of local sensitivity to nonignorability (ISNI; Troxel et al., Statistica Sinica, 14, 1221-1237, 2004; Zhang and Heitjan, Biometrics, 62, 1260-1268, 2006) to the case of a nonparametric survival model. The method involves, first, specifying a coarse-data selection model to describe the association between the failure and censoring processes and then evaluating the slope of the nonparametric survival MLE ordinate with respect to a nonignorability parameter in the neighborhood of the ignorable model. We define the nonparametric MLE of the survival curve for a fixed value of the nonignorability parameter and show in a simulation that ISNI analysis effectively captures local sensitivity to nonignorability. The method measures sensitivity in the sense of identifying functionals of the nonparametric MLE that nonignorability, if present, can affect substantially. We demonstrate the method with an application to a trial comparing mechanical assistance to optimal medical management in the treatment of end-stage heart failure.",
keywords = "Coarse-data model, Ignorability, Informative censoring, ISNI, Sensitivity analysis",
author = "Tao Liu and Heitjan, {Daniel F.}",
year = "2012",
month = "11",
day = "10",
doi = "10.1002/sim.5454",
language = "English (US)",
volume = "31",
pages = "2998--3010",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "25",

}

TY - JOUR

T1 - Sensitivity of the discrete-time Kaplan-Meier estimate to nonignorable censoring

T2 - Application in a clinical trial

AU - Liu, Tao

AU - Heitjan, Daniel F.

PY - 2012/11/10

Y1 - 2012/11/10

N2 - Untestable assumptions about the association between survival and censoring times can affect the validity of estimates of the survival distribution including the Kaplan-Meier (KM) nonparametric maximum likelihood estimate (MLE). This paper explores the sensitivity of the KM curve to nonignorable censoring by extending the index of local sensitivity to nonignorability (ISNI; Troxel et al., Statistica Sinica, 14, 1221-1237, 2004; Zhang and Heitjan, Biometrics, 62, 1260-1268, 2006) to the case of a nonparametric survival model. The method involves, first, specifying a coarse-data selection model to describe the association between the failure and censoring processes and then evaluating the slope of the nonparametric survival MLE ordinate with respect to a nonignorability parameter in the neighborhood of the ignorable model. We define the nonparametric MLE of the survival curve for a fixed value of the nonignorability parameter and show in a simulation that ISNI analysis effectively captures local sensitivity to nonignorability. The method measures sensitivity in the sense of identifying functionals of the nonparametric MLE that nonignorability, if present, can affect substantially. We demonstrate the method with an application to a trial comparing mechanical assistance to optimal medical management in the treatment of end-stage heart failure.

AB - Untestable assumptions about the association between survival and censoring times can affect the validity of estimates of the survival distribution including the Kaplan-Meier (KM) nonparametric maximum likelihood estimate (MLE). This paper explores the sensitivity of the KM curve to nonignorable censoring by extending the index of local sensitivity to nonignorability (ISNI; Troxel et al., Statistica Sinica, 14, 1221-1237, 2004; Zhang and Heitjan, Biometrics, 62, 1260-1268, 2006) to the case of a nonparametric survival model. The method involves, first, specifying a coarse-data selection model to describe the association between the failure and censoring processes and then evaluating the slope of the nonparametric survival MLE ordinate with respect to a nonignorability parameter in the neighborhood of the ignorable model. We define the nonparametric MLE of the survival curve for a fixed value of the nonignorability parameter and show in a simulation that ISNI analysis effectively captures local sensitivity to nonignorability. The method measures sensitivity in the sense of identifying functionals of the nonparametric MLE that nonignorability, if present, can affect substantially. We demonstrate the method with an application to a trial comparing mechanical assistance to optimal medical management in the treatment of end-stage heart failure.

KW - Coarse-data model

KW - Ignorability

KW - Informative censoring

KW - ISNI

KW - Sensitivity analysis

UR - http://www.scopus.com/inward/record.url?scp=84867472078&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84867472078&partnerID=8YFLogxK

U2 - 10.1002/sim.5454

DO - 10.1002/sim.5454

M3 - Article

C2 - 22806989

AN - SCOPUS:84867472078

VL - 31

SP - 2998

EP - 3010

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 25

ER -