On doubly robust estimation of the hazard difference

Publikation: Bidrag til tidsskriftTidsskriftartikelfagfællebedømt

Standard

On doubly robust estimation of the hazard difference. / Dukes, Oliver; Martinussen, Torben; Tchetgen Tchetgen, Eric J; Vansteelandt, Stijn.

I: Biometrics, Bind 75, Nr. 1, 03.2019, s. 100-109.

Publikation: Bidrag til tidsskriftTidsskriftartikelfagfællebedømt

Harvard

Dukes, O, Martinussen, T, Tchetgen Tchetgen, EJ & Vansteelandt, S 2019, 'On doubly robust estimation of the hazard difference', Biometrics, bind 75, nr. 1, s. 100-109. https://doi.org/10.1111/biom.12943

APA

Dukes, O., Martinussen, T., Tchetgen Tchetgen, E. J., & Vansteelandt, S. (2019). On doubly robust estimation of the hazard difference. Biometrics, 75(1), 100-109. https://doi.org/10.1111/biom.12943

Vancouver

Dukes O, Martinussen T, Tchetgen Tchetgen EJ, Vansteelandt S. On doubly robust estimation of the hazard difference. Biometrics. 2019 mar.;75(1):100-109. https://doi.org/10.1111/biom.12943

Author

Dukes, Oliver ; Martinussen, Torben ; Tchetgen Tchetgen, Eric J ; Vansteelandt, Stijn. / On doubly robust estimation of the hazard difference. I: Biometrics. 2019 ; Bind 75, Nr. 1. s. 100-109.

Bibtex

@article{af1579d5e2a0431886016be908268feb,
title = "On doubly robust estimation of the hazard difference",
abstract = "The estimation of conditional treatment effects in an observational study with a survival outcome typically involves fitting a hazards regression model adjusted for a high-dimensional covariate. Standard estimation of the treatment effect is then not entirely satisfactory, as the misspecification of the effect of this covariate may induce a large bias. Such misspecification is a particular concern when inferring the hazard difference, because it is difficult to postulate additive hazards models that guarantee non-negative hazards over the entire observed covariate range. We therefore consider a novel class of semiparametric additive hazards models which leave the effects of covariates unspecified. The efficient score under this model is derived. We then propose two different estimation approaches for the hazard difference (and hence also the relative chance of survival), both of which yield estimators that are doubly robust. The approaches are illustrated using simulation studies and data on right heart catheterization and mortality from the SUPPORT study.",
author = "Oliver Dukes and Torben Martinussen and {Tchetgen Tchetgen}, {Eric J} and Stijn Vansteelandt",
note = "{\textcopyright} 2018, The International Biometric Society.",
year = "2019",
month = mar,
doi = "10.1111/biom.12943",
language = "English",
volume = "75",
pages = "100--109",
journal = "Biometrics",
issn = "0006-341X",
publisher = "Wiley-Blackwell",
number = "1",

}

RIS

TY - JOUR

T1 - On doubly robust estimation of the hazard difference

AU - Dukes, Oliver

AU - Martinussen, Torben

AU - Tchetgen Tchetgen, Eric J

AU - Vansteelandt, Stijn

N1 - © 2018, The International Biometric Society.

PY - 2019/3

Y1 - 2019/3

N2 - The estimation of conditional treatment effects in an observational study with a survival outcome typically involves fitting a hazards regression model adjusted for a high-dimensional covariate. Standard estimation of the treatment effect is then not entirely satisfactory, as the misspecification of the effect of this covariate may induce a large bias. Such misspecification is a particular concern when inferring the hazard difference, because it is difficult to postulate additive hazards models that guarantee non-negative hazards over the entire observed covariate range. We therefore consider a novel class of semiparametric additive hazards models which leave the effects of covariates unspecified. The efficient score under this model is derived. We then propose two different estimation approaches for the hazard difference (and hence also the relative chance of survival), both of which yield estimators that are doubly robust. The approaches are illustrated using simulation studies and data on right heart catheterization and mortality from the SUPPORT study.

AB - The estimation of conditional treatment effects in an observational study with a survival outcome typically involves fitting a hazards regression model adjusted for a high-dimensional covariate. Standard estimation of the treatment effect is then not entirely satisfactory, as the misspecification of the effect of this covariate may induce a large bias. Such misspecification is a particular concern when inferring the hazard difference, because it is difficult to postulate additive hazards models that guarantee non-negative hazards over the entire observed covariate range. We therefore consider a novel class of semiparametric additive hazards models which leave the effects of covariates unspecified. The efficient score under this model is derived. We then propose two different estimation approaches for the hazard difference (and hence also the relative chance of survival), both of which yield estimators that are doubly robust. The approaches are illustrated using simulation studies and data on right heart catheterization and mortality from the SUPPORT study.

U2 - 10.1111/biom.12943

DO - 10.1111/biom.12943

M3 - Journal article

C2 - 30133696

VL - 75

SP - 100

EP - 109

JO - Biometrics

JF - Biometrics

SN - 0006-341X

IS - 1

ER -

ID: 223255961