Multiple imputation for interval censored data with auxiliary variables

Chiu-Hsieh Hsu, Jeremy M G Taylor, Susan Murray, Daniel Commenges

Research output: Contribution to journalArticle

18 Citations (Scopus)

Abstract

We propose a non-parametric multiple imputation scheme, NPMLE imputation, for the analysis of interval censored survival data. Features of the method are that it converts interval-censored data problems to complete data or right censored data problems to which many standard approaches can be used, and that measures of uncertainty are easily obtained. In addition to the event time of primary interest, there are frequently other auxiliary variables that are associated with the event time. For the goal of estimating the marginal survival distribution, these auxiliary variables may provide some additional information about the event time for the interval censored observations. We extend the imputation methods to incorporate information from auxiliary variables with potentially complex structures. To conduct the imputation, we use a working failure-time proportional hazards model to define an imputing risk set for each censored observation. The imputation schemes consist of using the data in the imputing risk sets to create an exact event time for each interval censored observation. In simulation studies we show that the use of multiple imputation methods can improve the efficiency of estimators and reduce the effect of missing visits when compared to simpler approaches. We apply the approach to cytomegalovirus shedding data from an AIDS clinical trial, in which CD4 count is the auxiliary variable.

Original languageEnglish (US)
Pages (from-to)769-781
Number of pages13
JournalStatistics in Medicine
Volume26
Issue number4
DOIs
StatePublished - Feb 20 2007

Fingerprint

Interval-censored Data
Multiple Imputation
Auxiliary Variables
Imputation
Censored Observations
Censored Survival Data
Survival Distribution
Right-censored Data
Interval
Proportional Hazards Model
Observation
Failure Time
Marginal Distribution
Complex Structure
Clinical Trials
Convert
CD4 Lymphocyte Count
Count
Cytomegalovirus
Proportional Hazards Models

Keywords

  • Auxiliary variables
  • Interval censored
  • Multiple imputation

ASJC Scopus subject areas

  • Epidemiology

Cite this

Multiple imputation for interval censored data with auxiliary variables. / Hsu, Chiu-Hsieh; Taylor, Jeremy M G; Murray, Susan; Commenges, Daniel.

In: Statistics in Medicine, Vol. 26, No. 4, 20.02.2007, p. 769-781.

Research output: Contribution to journalArticle

Hsu, Chiu-Hsieh ; Taylor, Jeremy M G ; Murray, Susan ; Commenges, Daniel. / Multiple imputation for interval censored data with auxiliary variables. In: Statistics in Medicine. 2007 ; Vol. 26, No. 4. pp. 769-781.
@article{db1806420fad497095ee068b0499c5bc,
title = "Multiple imputation for interval censored data with auxiliary variables",
abstract = "We propose a non-parametric multiple imputation scheme, NPMLE imputation, for the analysis of interval censored survival data. Features of the method are that it converts interval-censored data problems to complete data or right censored data problems to which many standard approaches can be used, and that measures of uncertainty are easily obtained. In addition to the event time of primary interest, there are frequently other auxiliary variables that are associated with the event time. For the goal of estimating the marginal survival distribution, these auxiliary variables may provide some additional information about the event time for the interval censored observations. We extend the imputation methods to incorporate information from auxiliary variables with potentially complex structures. To conduct the imputation, we use a working failure-time proportional hazards model to define an imputing risk set for each censored observation. The imputation schemes consist of using the data in the imputing risk sets to create an exact event time for each interval censored observation. In simulation studies we show that the use of multiple imputation methods can improve the efficiency of estimators and reduce the effect of missing visits when compared to simpler approaches. We apply the approach to cytomegalovirus shedding data from an AIDS clinical trial, in which CD4 count is the auxiliary variable.",
keywords = "Auxiliary variables, Interval censored, Multiple imputation",
author = "Chiu-Hsieh Hsu and Taylor, {Jeremy M G} and Susan Murray and Daniel Commenges",
year = "2007",
month = "2",
day = "20",
doi = "10.1002/sim.2581",
language = "English (US)",
volume = "26",
pages = "769--781",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "4",

}

TY - JOUR

T1 - Multiple imputation for interval censored data with auxiliary variables

AU - Hsu, Chiu-Hsieh

AU - Taylor, Jeremy M G

AU - Murray, Susan

AU - Commenges, Daniel

PY - 2007/2/20

Y1 - 2007/2/20

N2 - We propose a non-parametric multiple imputation scheme, NPMLE imputation, for the analysis of interval censored survival data. Features of the method are that it converts interval-censored data problems to complete data or right censored data problems to which many standard approaches can be used, and that measures of uncertainty are easily obtained. In addition to the event time of primary interest, there are frequently other auxiliary variables that are associated with the event time. For the goal of estimating the marginal survival distribution, these auxiliary variables may provide some additional information about the event time for the interval censored observations. We extend the imputation methods to incorporate information from auxiliary variables with potentially complex structures. To conduct the imputation, we use a working failure-time proportional hazards model to define an imputing risk set for each censored observation. The imputation schemes consist of using the data in the imputing risk sets to create an exact event time for each interval censored observation. In simulation studies we show that the use of multiple imputation methods can improve the efficiency of estimators and reduce the effect of missing visits when compared to simpler approaches. We apply the approach to cytomegalovirus shedding data from an AIDS clinical trial, in which CD4 count is the auxiliary variable.

AB - We propose a non-parametric multiple imputation scheme, NPMLE imputation, for the analysis of interval censored survival data. Features of the method are that it converts interval-censored data problems to complete data or right censored data problems to which many standard approaches can be used, and that measures of uncertainty are easily obtained. In addition to the event time of primary interest, there are frequently other auxiliary variables that are associated with the event time. For the goal of estimating the marginal survival distribution, these auxiliary variables may provide some additional information about the event time for the interval censored observations. We extend the imputation methods to incorporate information from auxiliary variables with potentially complex structures. To conduct the imputation, we use a working failure-time proportional hazards model to define an imputing risk set for each censored observation. The imputation schemes consist of using the data in the imputing risk sets to create an exact event time for each interval censored observation. In simulation studies we show that the use of multiple imputation methods can improve the efficiency of estimators and reduce the effect of missing visits when compared to simpler approaches. We apply the approach to cytomegalovirus shedding data from an AIDS clinical trial, in which CD4 count is the auxiliary variable.

KW - Auxiliary variables

KW - Interval censored

KW - Multiple imputation

UR - http://www.scopus.com/inward/record.url?scp=33846820182&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33846820182&partnerID=8YFLogxK

U2 - 10.1002/sim.2581

DO - 10.1002/sim.2581

M3 - Article

C2 - 16755528

AN - SCOPUS:33846820182

VL - 26

SP - 769

EP - 781

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 4

ER -