Performance of standard imputation methods for missing quality of life data as covariate in survival analysis based on simulations from the International Breast Cancer Study Group Trials VI and VII*

Marion Procter, Chris Robertson

Research output: Contribution to journalArticle

Abstract

Imputation methods for missing data on a time-dependent variable within time-dependent Cox models are investigated in a simulation study. Quality of life (QoL) assessments were removed from the complete simulated datasets, which have a positive relationship between QoL and disease-free survival (DFS) and delayed chemotherapy and DFS, by missing at random and missing not at random (MNAR) mechanisms. Standard imputation methods were applied before analysis. Method performance was influenced by missing data mechanism, with one exception for simple imputation. The greatest bias occurred under MNAR and large effect sizes. It is important to carefully investigate the missing data mechanism.

Original languageEnglish
Pages (from-to)3063-3077
Number of pages15
JournalCommunications in Statistics: Simulation and Computation
Volume48
Issue number10
Early online date13 Aug 2018
DOIs
Publication statusPublished - 26 Nov 2019

Fingerprint

Missing at Random
Quality of Life
Survival Analysis
Imputation
Breast Cancer
Missing Data Mechanism
Covariates
Chemotherapy
Cox Model
Simulation
Effect Size
Missing Data
Exception
Simulation Study
Standards

Keywords

  • imputation methods
  • missing data mechanism
  • quality of life
  • simulation study
  • time-dependent Cox model

Cite this

@article{65c21a5b484d4704b693736f51aea5be,
title = "Performance of standard imputation methods for missing quality of life data as covariate in survival analysis based on simulations from the International Breast Cancer Study Group Trials VI and VII*",
abstract = "Imputation methods for missing data on a time-dependent variable within time-dependent Cox models are investigated in a simulation study. Quality of life (QoL) assessments were removed from the complete simulated datasets, which have a positive relationship between QoL and disease-free survival (DFS) and delayed chemotherapy and DFS, by missing at random and missing not at random (MNAR) mechanisms. Standard imputation methods were applied before analysis. Method performance was influenced by missing data mechanism, with one exception for simple imputation. The greatest bias occurred under MNAR and large effect sizes. It is important to carefully investigate the missing data mechanism.",
keywords = "imputation methods, missing data mechanism, quality of life, simulation study, time-dependent Cox model",
author = "Marion Procter and Chris Robertson",
year = "2019",
month = "11",
day = "26",
doi = "10.1080/03610918.2018.1473587",
language = "English",
volume = "48",
pages = "3063--3077",
journal = "Communications in Statistics - Simulation and Computation",
issn = "0361-0918",
number = "10",

}

TY - JOUR

T1 - Performance of standard imputation methods for missing quality of life data as covariate in survival analysis based on simulations from the International Breast Cancer Study Group Trials VI and VII*

AU - Procter, Marion

AU - Robertson, Chris

PY - 2019/11/26

Y1 - 2019/11/26

N2 - Imputation methods for missing data on a time-dependent variable within time-dependent Cox models are investigated in a simulation study. Quality of life (QoL) assessments were removed from the complete simulated datasets, which have a positive relationship between QoL and disease-free survival (DFS) and delayed chemotherapy and DFS, by missing at random and missing not at random (MNAR) mechanisms. Standard imputation methods were applied before analysis. Method performance was influenced by missing data mechanism, with one exception for simple imputation. The greatest bias occurred under MNAR and large effect sizes. It is important to carefully investigate the missing data mechanism.

AB - Imputation methods for missing data on a time-dependent variable within time-dependent Cox models are investigated in a simulation study. Quality of life (QoL) assessments were removed from the complete simulated datasets, which have a positive relationship between QoL and disease-free survival (DFS) and delayed chemotherapy and DFS, by missing at random and missing not at random (MNAR) mechanisms. Standard imputation methods were applied before analysis. Method performance was influenced by missing data mechanism, with one exception for simple imputation. The greatest bias occurred under MNAR and large effect sizes. It is important to carefully investigate the missing data mechanism.

KW - imputation methods

KW - missing data mechanism

KW - quality of life

KW - simulation study

KW - time-dependent Cox model

UR - http://www.scopus.com/inward/record.url?scp=85051949762&partnerID=8YFLogxK

U2 - 10.1080/03610918.2018.1473587

DO - 10.1080/03610918.2018.1473587

M3 - Article

AN - SCOPUS:85051949762

VL - 48

SP - 3063

EP - 3077

JO - Communications in Statistics - Simulation and Computation

JF - Communications in Statistics - Simulation and Computation

SN - 0361-0918

IS - 10

ER -