Reweighting estimators for Cox regression with missing covariate data: Analysis of insulin resistance and risk of stroke in the Northern Manhattan Study

Qiang Xu, Myunghee Cho Paik, Tatjana Rundek, Mitchell S V Elkind, Ralph L Sacco

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

Incomplete covariates often obscure analysis results from a Cox regression. In an analysis of the Northern Manhattan Study (NOMAS) to determine the influence of insulin resistance on the incidence of stroke in nondiabetic individuals, insulin level is unknown for 34.1% of the subjects. The available data suggest that the missingness mechanism depends on outcome variables, which may generate biases in estimating the parameters of interest if only using the complete observations. This article aimed to introduce practical strategies to analyze the NOMAS data and present sensitivity analyses by using the reweighting method in standard statistical packages. When the data set structure is in counting process style, the reweighting estimates can be obtained by built-in procedures with variance estimated by the jackknife method. Simulation results indicate that the jackknife variance estimate provides reasonable coverage probability in moderate sample sizes. We subsequently conducted sensitivity analyses for the NOMAS data, showing that the risk estimates are robust to a variety of missingness mechanisms. At the end of this article, we present the core SAS and R programs used in the analysis.

Original languageEnglish
Pages (from-to)3328-3340
Number of pages13
JournalStatistics in Medicine
Volume30
Issue number28
DOIs
StatePublished - Dec 10 2011

Fingerprint

Missing Covariates
Cox Regression
Insulin
Stroke
Insulin Resistance
Data analysis
Estimator
Jackknife
Estimate
Statistical package
Sample Size
Counting Process
Coverage Probability
Covariates
Incidence
Unknown
Resistance
Simulation

Keywords

  • Missing covariate
  • Proportional hazards model
  • Weighted estimating equation

ASJC Scopus subject areas

  • Epidemiology
  • Statistics and Probability

Cite this

Reweighting estimators for Cox regression with missing covariate data : Analysis of insulin resistance and risk of stroke in the Northern Manhattan Study. / Xu, Qiang; Paik, Myunghee Cho; Rundek, Tatjana; Elkind, Mitchell S V; Sacco, Ralph L.

In: Statistics in Medicine, Vol. 30, No. 28, 10.12.2011, p. 3328-3340.

Research output: Contribution to journalArticle

@article{47eca3d1a91944b0885cd13aa2909f88,
title = "Reweighting estimators for Cox regression with missing covariate data: Analysis of insulin resistance and risk of stroke in the Northern Manhattan Study",
abstract = "Incomplete covariates often obscure analysis results from a Cox regression. In an analysis of the Northern Manhattan Study (NOMAS) to determine the influence of insulin resistance on the incidence of stroke in nondiabetic individuals, insulin level is unknown for 34.1{\%} of the subjects. The available data suggest that the missingness mechanism depends on outcome variables, which may generate biases in estimating the parameters of interest if only using the complete observations. This article aimed to introduce practical strategies to analyze the NOMAS data and present sensitivity analyses by using the reweighting method in standard statistical packages. When the data set structure is in counting process style, the reweighting estimates can be obtained by built-in procedures with variance estimated by the jackknife method. Simulation results indicate that the jackknife variance estimate provides reasonable coverage probability in moderate sample sizes. We subsequently conducted sensitivity analyses for the NOMAS data, showing that the risk estimates are robust to a variety of missingness mechanisms. At the end of this article, we present the core SAS and R programs used in the analysis.",
keywords = "Missing covariate, Proportional hazards model, Weighted estimating equation",
author = "Qiang Xu and Paik, {Myunghee Cho} and Tatjana Rundek and Elkind, {Mitchell S V} and Sacco, {Ralph L}",
year = "2011",
month = "12",
day = "10",
doi = "10.1002/sim.4380",
language = "English",
volume = "30",
pages = "3328--3340",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "28",

}

TY - JOUR

T1 - Reweighting estimators for Cox regression with missing covariate data

T2 - Analysis of insulin resistance and risk of stroke in the Northern Manhattan Study

AU - Xu, Qiang

AU - Paik, Myunghee Cho

AU - Rundek, Tatjana

AU - Elkind, Mitchell S V

AU - Sacco, Ralph L

PY - 2011/12/10

Y1 - 2011/12/10

N2 - Incomplete covariates often obscure analysis results from a Cox regression. In an analysis of the Northern Manhattan Study (NOMAS) to determine the influence of insulin resistance on the incidence of stroke in nondiabetic individuals, insulin level is unknown for 34.1% of the subjects. The available data suggest that the missingness mechanism depends on outcome variables, which may generate biases in estimating the parameters of interest if only using the complete observations. This article aimed to introduce practical strategies to analyze the NOMAS data and present sensitivity analyses by using the reweighting method in standard statistical packages. When the data set structure is in counting process style, the reweighting estimates can be obtained by built-in procedures with variance estimated by the jackknife method. Simulation results indicate that the jackknife variance estimate provides reasonable coverage probability in moderate sample sizes. We subsequently conducted sensitivity analyses for the NOMAS data, showing that the risk estimates are robust to a variety of missingness mechanisms. At the end of this article, we present the core SAS and R programs used in the analysis.

AB - Incomplete covariates often obscure analysis results from a Cox regression. In an analysis of the Northern Manhattan Study (NOMAS) to determine the influence of insulin resistance on the incidence of stroke in nondiabetic individuals, insulin level is unknown for 34.1% of the subjects. The available data suggest that the missingness mechanism depends on outcome variables, which may generate biases in estimating the parameters of interest if only using the complete observations. This article aimed to introduce practical strategies to analyze the NOMAS data and present sensitivity analyses by using the reweighting method in standard statistical packages. When the data set structure is in counting process style, the reweighting estimates can be obtained by built-in procedures with variance estimated by the jackknife method. Simulation results indicate that the jackknife variance estimate provides reasonable coverage probability in moderate sample sizes. We subsequently conducted sensitivity analyses for the NOMAS data, showing that the risk estimates are robust to a variety of missingness mechanisms. At the end of this article, we present the core SAS and R programs used in the analysis.

KW - Missing covariate

KW - Proportional hazards model

KW - Weighted estimating equation

UR - http://www.scopus.com/inward/record.url?scp=81955164233&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=81955164233&partnerID=8YFLogxK

U2 - 10.1002/sim.4380

DO - 10.1002/sim.4380

M3 - Article

C2 - 21965165

AN - SCOPUS:81955164233

VL - 30

SP - 3328

EP - 3340

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 28

ER -