Multiple imputation for competing risks survival data via pseudo-observations

Seungbong Han; Adin Cristian Andrei; Kam Wah Tsui

doi:10.29220/CSAM.2018.25.4.385

Multiple imputation for competing risks survival data via pseudo-observations

Seungbong Han^*, Adin Cristian Andrei, Kam Wah Tsui

^*Corresponding author for this work

Preventive Medicine

Research output: Contribution to journal › Article › peer-review

2 Scopus citations

Abstract

Competing risks are commonly encountered in biomedical research. Regression models for competing risks data can be developed based on data routinely collected in hospitals or general practices. However, these data sets usually contain the covariate missing values. To overcome this problem, multiple imputation is often used to fit regression models under a MAR assumption. Here, we introduce a multivariate imputation in a chained equations algorithm to deal with competing risks survival data. Using pseudo-observations, we make use of the available outcome information by accommodating the competing risk structure. Lastly, we illustrate the practical advantages of our approach using simulations and two data examples from a coronary artery disease data and hepatocellular carcinoma data.

Original language	English (US)
Pages (from-to)	385-396
Number of pages	12
Journal	Communications for Statistical Applications and Methods
Volume	25
Issue number	4
DOIs	https://doi.org/10.29220/CSAM.2018.25.4.385
State	Published - Jul 1 2018

Funding

This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT & Future Planning (NRF-2017R1C1B 1006717).

Keywords

Competing risks
Missing data
Multiple imputation
Pseudo-observations
Random forest

ASJC Scopus subject areas

Statistics and Probability
Modeling and Simulation
Finance
Statistics, Probability and Uncertainty
Applied Mathematics

Access to Document

10.29220/CSAM.2018.25.4.385

Cite this

@article{8cc212888a5f4f8ca5b7de44e3559a69,

title = "Multiple imputation for competing risks survival data via pseudo-observations",

abstract = "Competing risks are commonly encountered in biomedical research. Regression models for competing risks data can be developed based on data routinely collected in hospitals or general practices. However, these data sets usually contain the covariate missing values. To overcome this problem, multiple imputation is often used to fit regression models under a MAR assumption. Here, we introduce a multivariate imputation in a chained equations algorithm to deal with competing risks survival data. Using pseudo-observations, we make use of the available outcome information by accommodating the competing risk structure. Lastly, we illustrate the practical advantages of our approach using simulations and two data examples from a coronary artery disease data and hepatocellular carcinoma data.",

keywords = "Competing risks, Missing data, Multiple imputation, Pseudo-observations, Random forest",

author = "Seungbong Han and Andrei, {Adin Cristian} and Tsui, {Kam Wah}",

note = "Publisher Copyright: {\textcopyright} 2018 The Korean Statistical Society, and Korean International Statistical Society.",

year = "2018",

month = jul,

day = "1",

doi = "10.29220/CSAM.2018.25.4.385",

language = "English (US)",

volume = "25",

pages = "385--396",

journal = "Communications for Statistical Applications and Methods",

issn = "2287-7843",

publisher = "Korean Statistical Society",

number = "4",

}

TY - JOUR

T1 - Multiple imputation for competing risks survival data via pseudo-observations

AU - Han, Seungbong

AU - Andrei, Adin Cristian

AU - Tsui, Kam Wah

PY - 2018/7/1

Y1 - 2018/7/1

N2 - Competing risks are commonly encountered in biomedical research. Regression models for competing risks data can be developed based on data routinely collected in hospitals or general practices. However, these data sets usually contain the covariate missing values. To overcome this problem, multiple imputation is often used to fit regression models under a MAR assumption. Here, we introduce a multivariate imputation in a chained equations algorithm to deal with competing risks survival data. Using pseudo-observations, we make use of the available outcome information by accommodating the competing risk structure. Lastly, we illustrate the practical advantages of our approach using simulations and two data examples from a coronary artery disease data and hepatocellular carcinoma data.

AB - Competing risks are commonly encountered in biomedical research. Regression models for competing risks data can be developed based on data routinely collected in hospitals or general practices. However, these data sets usually contain the covariate missing values. To overcome this problem, multiple imputation is often used to fit regression models under a MAR assumption. Here, we introduce a multivariate imputation in a chained equations algorithm to deal with competing risks survival data. Using pseudo-observations, we make use of the available outcome information by accommodating the competing risk structure. Lastly, we illustrate the practical advantages of our approach using simulations and two data examples from a coronary artery disease data and hepatocellular carcinoma data.

KW - Competing risks

KW - Missing data

KW - Multiple imputation

KW - Pseudo-observations

KW - Random forest

UR - http://www.scopus.com/inward/record.url?scp=85054015581&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85054015581&partnerID=8YFLogxK

U2 - 10.29220/CSAM.2018.25.4.385

DO - 10.29220/CSAM.2018.25.4.385

M3 - Article

AN - SCOPUS:85054015581

SN - 2287-7843

VL - 25

SP - 385

EP - 396

JO - Communications for Statistical Applications and Methods

JF - Communications for Statistical Applications and Methods

IS - 4

ER -

Multiple imputation for competing risks survival data via pseudo-observations

Abstract

Funding

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this