Time-to-event data with time-varying biomarkers measured only at study entry, with applications to Alzheimer's disease

for the Alzheimer's Disease Neuroimaging Initiative

Research output: Contribution to journalArticle

Abstract

Relating time-varying biomarkers of Alzheimer's disease to time-to-event using a Cox model is complicated by the fact that Alzheimer's disease biomarkers are sparsely collected, typically only at study entry; this is problematic since Cox regression with time-varying covariates requires observation of the covariate process at all failure times. The analysis might be simplified by using study entry as the time origin and treating the time-varying covariate measured at study entry as a fixed baseline covariate. In this paper, we first derive conditions under which using an incorrect time origin of study entry results in consistent estimation of regression parameters when the time-varying covariate is continuous and fully observed. We then derive conditions under which treating the time-varying covariate as fixed at study entry results in consistent estimation. We provide methods for estimating the regression parameter when a functional form can be assumed for the time-varying biomarker, which is measured only at study entry. We demonstrate our analytical results in a simulation study and apply our methods to data from the Rush Religious Orders Study and Memory and Aging Project and data from the Alzheimer's Disease Neuroimaging Initiative.

Original languageEnglish (US)
Pages (from-to)914-932
Number of pages19
JournalStatistics in Medicine
Volume37
Issue number6
DOIs
StatePublished - Mar 15 2018

Fingerprint

Time-varying Covariates
Alzheimer's Disease
Biomarkers
Time-varying
Alzheimer Disease
Consistent Estimation
Covariates
Regression
Cox Regression
Neuroimaging
Cox Model
Failure Time
Baseline
Simulation Study
Demonstrate
Proportional Hazards Models
Observation

Keywords

  • Cox model
  • delayed entry
  • left truncation
  • survival analysis
  • time origin
  • time-dependent covariates

ASJC Scopus subject areas

  • Epidemiology
  • Statistics and Probability

Cite this

Time-to-event data with time-varying biomarkers measured only at study entry, with applications to Alzheimer's disease. / for the Alzheimer's Disease Neuroimaging Initiative.

In: Statistics in Medicine, Vol. 37, No. 6, 15.03.2018, p. 914-932.

Research output: Contribution to journalArticle

for the Alzheimer's Disease Neuroimaging Initiative. / Time-to-event data with time-varying biomarkers measured only at study entry, with applications to Alzheimer's disease. In: Statistics in Medicine. 2018 ; Vol. 37, No. 6. pp. 914-932.
@article{b22830758168494b9e6b2734b66e60aa,
title = "Time-to-event data with time-varying biomarkers measured only at study entry, with applications to Alzheimer's disease",
abstract = "Relating time-varying biomarkers of Alzheimer's disease to time-to-event using a Cox model is complicated by the fact that Alzheimer's disease biomarkers are sparsely collected, typically only at study entry; this is problematic since Cox regression with time-varying covariates requires observation of the covariate process at all failure times. The analysis might be simplified by using study entry as the time origin and treating the time-varying covariate measured at study entry as a fixed baseline covariate. In this paper, we first derive conditions under which using an incorrect time origin of study entry results in consistent estimation of regression parameters when the time-varying covariate is continuous and fully observed. We then derive conditions under which treating the time-varying covariate as fixed at study entry results in consistent estimation. We provide methods for estimating the regression parameter when a functional form can be assumed for the time-varying biomarker, which is measured only at study entry. We demonstrate our analytical results in a simulation study and apply our methods to data from the Rush Religious Orders Study and Memory and Aging Project and data from the Alzheimer's Disease Neuroimaging Initiative.",
keywords = "Cox model, delayed entry, left truncation, survival analysis, time origin, time-dependent covariates",
author = "{for the Alzheimer's Disease Neuroimaging Initiative} and Catherine Lee and Rebecca Betensky",
year = "2018",
month = "3",
day = "15",
doi = "10.1002/sim.7547",
language = "English (US)",
volume = "37",
pages = "914--932",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "6",

}

TY - JOUR

T1 - Time-to-event data with time-varying biomarkers measured only at study entry, with applications to Alzheimer's disease

AU - for the Alzheimer's Disease Neuroimaging Initiative

AU - Lee, Catherine

AU - Betensky, Rebecca

PY - 2018/3/15

Y1 - 2018/3/15

N2 - Relating time-varying biomarkers of Alzheimer's disease to time-to-event using a Cox model is complicated by the fact that Alzheimer's disease biomarkers are sparsely collected, typically only at study entry; this is problematic since Cox regression with time-varying covariates requires observation of the covariate process at all failure times. The analysis might be simplified by using study entry as the time origin and treating the time-varying covariate measured at study entry as a fixed baseline covariate. In this paper, we first derive conditions under which using an incorrect time origin of study entry results in consistent estimation of regression parameters when the time-varying covariate is continuous and fully observed. We then derive conditions under which treating the time-varying covariate as fixed at study entry results in consistent estimation. We provide methods for estimating the regression parameter when a functional form can be assumed for the time-varying biomarker, which is measured only at study entry. We demonstrate our analytical results in a simulation study and apply our methods to data from the Rush Religious Orders Study and Memory and Aging Project and data from the Alzheimer's Disease Neuroimaging Initiative.

AB - Relating time-varying biomarkers of Alzheimer's disease to time-to-event using a Cox model is complicated by the fact that Alzheimer's disease biomarkers are sparsely collected, typically only at study entry; this is problematic since Cox regression with time-varying covariates requires observation of the covariate process at all failure times. The analysis might be simplified by using study entry as the time origin and treating the time-varying covariate measured at study entry as a fixed baseline covariate. In this paper, we first derive conditions under which using an incorrect time origin of study entry results in consistent estimation of regression parameters when the time-varying covariate is continuous and fully observed. We then derive conditions under which treating the time-varying covariate as fixed at study entry results in consistent estimation. We provide methods for estimating the regression parameter when a functional form can be assumed for the time-varying biomarker, which is measured only at study entry. We demonstrate our analytical results in a simulation study and apply our methods to data from the Rush Religious Orders Study and Memory and Aging Project and data from the Alzheimer's Disease Neuroimaging Initiative.

KW - Cox model

KW - delayed entry

KW - left truncation

KW - survival analysis

KW - time origin

KW - time-dependent covariates

UR - http://www.scopus.com/inward/record.url?scp=85038418239&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85038418239&partnerID=8YFLogxK

U2 - 10.1002/sim.7547

DO - 10.1002/sim.7547

M3 - Article

VL - 37

SP - 914

EP - 932

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 6

ER -