The relative efficiency of time-to-threshold and rate of change in longitudinal data

M. C. Donohue, A. C. Gamst, R. G. Thomas, R. Xu, Laurel A Beckett, R. C. Petersen, M. W. Weiner, P. Aisen

Research output: Contribution to journalArticle

18 Citations (Scopus)

Abstract

Randomized, placebo-controlled trials often use time-to-event as the primary endpoint, even when a continuous measure of disease severity is available. We compare the power to detect a treatment effect using either rate of change, as estimated by linear models of longitudinal continuous data, or time-to-event estimated by Cox proportional hazards models. We propose an analytic inflation factor for comparing the two types of analyses assuming that the time-to-event can be expressed as a time-to-threshold of the continuous measure. We conduct simulations based on a publicly available Alzheimer's disease data set in which the time-to-event is algorithmically defined based on a battery of assessments. A Cox proportional hazards model of the time-to-event endpoint is compared to a linear model of a single assessment from the battery. The simulations also explore the impact of baseline covariates in either analysis.

Original languageEnglish (US)
Pages (from-to)685-693
Number of pages9
JournalContemporary Clinical Trials
Volume32
Issue number5
DOIs
StatePublished - Sep 2011

Fingerprint

Proportional Hazards Models
Linear Models
Economic Inflation
Alzheimer Disease
Randomized Controlled Trials
Placebos
Therapeutics
Power (Psychology)
Datasets

Keywords

  • Linear mixed models
  • Longitudinal data
  • Marginal linear models
  • Power
  • Survival analysis

ASJC Scopus subject areas

  • Pharmacology (medical)
  • Medicine(all)

Cite this

The relative efficiency of time-to-threshold and rate of change in longitudinal data. / Donohue, M. C.; Gamst, A. C.; Thomas, R. G.; Xu, R.; Beckett, Laurel A; Petersen, R. C.; Weiner, M. W.; Aisen, P.

In: Contemporary Clinical Trials, Vol. 32, No. 5, 09.2011, p. 685-693.

Research output: Contribution to journalArticle

Donohue, MC, Gamst, AC, Thomas, RG, Xu, R, Beckett, LA, Petersen, RC, Weiner, MW & Aisen, P 2011, 'The relative efficiency of time-to-threshold and rate of change in longitudinal data', Contemporary Clinical Trials, vol. 32, no. 5, pp. 685-693. https://doi.org/10.1016/j.cct.2011.04.007
Donohue, M. C. ; Gamst, A. C. ; Thomas, R. G. ; Xu, R. ; Beckett, Laurel A ; Petersen, R. C. ; Weiner, M. W. ; Aisen, P. / The relative efficiency of time-to-threshold and rate of change in longitudinal data. In: Contemporary Clinical Trials. 2011 ; Vol. 32, No. 5. pp. 685-693.
@article{d5490ced8fbd4ebea837395fca3f2fc4,
title = "The relative efficiency of time-to-threshold and rate of change in longitudinal data",
abstract = "Randomized, placebo-controlled trials often use time-to-event as the primary endpoint, even when a continuous measure of disease severity is available. We compare the power to detect a treatment effect using either rate of change, as estimated by linear models of longitudinal continuous data, or time-to-event estimated by Cox proportional hazards models. We propose an analytic inflation factor for comparing the two types of analyses assuming that the time-to-event can be expressed as a time-to-threshold of the continuous measure. We conduct simulations based on a publicly available Alzheimer's disease data set in which the time-to-event is algorithmically defined based on a battery of assessments. A Cox proportional hazards model of the time-to-event endpoint is compared to a linear model of a single assessment from the battery. The simulations also explore the impact of baseline covariates in either analysis.",
keywords = "Linear mixed models, Longitudinal data, Marginal linear models, Power, Survival analysis",
author = "Donohue, {M. C.} and Gamst, {A. C.} and Thomas, {R. G.} and R. Xu and Beckett, {Laurel A} and Petersen, {R. C.} and Weiner, {M. W.} and P. Aisen",
year = "2011",
month = "9",
doi = "10.1016/j.cct.2011.04.007",
language = "English (US)",
volume = "32",
pages = "685--693",
journal = "Contemporary Clinical Trials",
issn = "1551-7144",
publisher = "Elsevier Inc.",
number = "5",

}

TY - JOUR

T1 - The relative efficiency of time-to-threshold and rate of change in longitudinal data

AU - Donohue, M. C.

AU - Gamst, A. C.

AU - Thomas, R. G.

AU - Xu, R.

AU - Beckett, Laurel A

AU - Petersen, R. C.

AU - Weiner, M. W.

AU - Aisen, P.

PY - 2011/9

Y1 - 2011/9

N2 - Randomized, placebo-controlled trials often use time-to-event as the primary endpoint, even when a continuous measure of disease severity is available. We compare the power to detect a treatment effect using either rate of change, as estimated by linear models of longitudinal continuous data, or time-to-event estimated by Cox proportional hazards models. We propose an analytic inflation factor for comparing the two types of analyses assuming that the time-to-event can be expressed as a time-to-threshold of the continuous measure. We conduct simulations based on a publicly available Alzheimer's disease data set in which the time-to-event is algorithmically defined based on a battery of assessments. A Cox proportional hazards model of the time-to-event endpoint is compared to a linear model of a single assessment from the battery. The simulations also explore the impact of baseline covariates in either analysis.

AB - Randomized, placebo-controlled trials often use time-to-event as the primary endpoint, even when a continuous measure of disease severity is available. We compare the power to detect a treatment effect using either rate of change, as estimated by linear models of longitudinal continuous data, or time-to-event estimated by Cox proportional hazards models. We propose an analytic inflation factor for comparing the two types of analyses assuming that the time-to-event can be expressed as a time-to-threshold of the continuous measure. We conduct simulations based on a publicly available Alzheimer's disease data set in which the time-to-event is algorithmically defined based on a battery of assessments. A Cox proportional hazards model of the time-to-event endpoint is compared to a linear model of a single assessment from the battery. The simulations also explore the impact of baseline covariates in either analysis.

KW - Linear mixed models

KW - Longitudinal data

KW - Marginal linear models

KW - Power

KW - Survival analysis

UR - http://www.scopus.com/inward/record.url?scp=79960894260&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79960894260&partnerID=8YFLogxK

U2 - 10.1016/j.cct.2011.04.007

DO - 10.1016/j.cct.2011.04.007

M3 - Article

VL - 32

SP - 685

EP - 693

JO - Contemporary Clinical Trials

JF - Contemporary Clinical Trials

SN - 1551-7144

IS - 5

ER -