Moving beyond the hazard ratio in quantifying the between-group difference in survival analysis

Hajime Uno, Brian Claggett, Lu Tian, Eisuke Inoue, Paul Gallo, Toshio Miyata, Deborah Schrag, Masahiro Takeuchi, Yoshiaki Uyama, Lihui Zhao, Hicham Skali, Scott Solomon, Susanna Jacobus, Michael Hughes, Milton Packer, Lee Jen Wei

Research output: Contribution to journalArticle

171 Citations (Scopus)

Abstract

In a longitudinal clinical study to compare two groups, the primary end point is often the time to a specific event (eg, disease progression, death). The hazard ratio estimate is routinely used to empirically quantify the between-group difference under the assumption that the ratio of the two hazard functions is approximately constant over time. When this assumption is plausible, such a ratio estimate may capture the relative difference between two survival curves. However, the clinical meaning of such a ratio estimate is difficult, if not impossible, to interpret when the underlying proportional hazards assumption is violated (ie, the hazard ratio is not constant over time). Although this issue has been studied extensively and various alternatives to the hazard ratio estimator have been discussed in the statistical literature, such crucial information does not seem to have reached the broader community of health science researchers. In this article, we summarize several critical concerns regarding this conventional practice and discuss various well-known alternatives for quantifying the underlying differences between groups with respect to a time-to-event end point. The data from three recent cancer clinical trials, which reflect a variety of scenarios, are used throughout to illustrate our discussions. When there is not sufficient information about the profile of the between-group difference at the design stage of the study, we encourage practitioners to consider a prespecified, clinically meaningful, model-free measure for quantifying the difference and to use robust estimation procedures to draw primary inferences.

Original languageEnglish (US)
Pages (from-to)2380-2385
Number of pages6
JournalJournal of Clinical Oncology
Volume32
Issue number22
DOIs
StatePublished - Aug 1 2014

Fingerprint

Survival Analysis
Longitudinal Studies
Disease Progression
Research Personnel
Clinical Trials
Health
Neoplasms

ASJC Scopus subject areas

  • Cancer Research
  • Oncology

Cite this

Uno, H., Claggett, B., Tian, L., Inoue, E., Gallo, P., Miyata, T., ... Wei, L. J. (2014). Moving beyond the hazard ratio in quantifying the between-group difference in survival analysis. Journal of Clinical Oncology, 32(22), 2380-2385. https://doi.org/10.1200/JCO.2014.55.2208

Moving beyond the hazard ratio in quantifying the between-group difference in survival analysis. / Uno, Hajime; Claggett, Brian; Tian, Lu; Inoue, Eisuke; Gallo, Paul; Miyata, Toshio; Schrag, Deborah; Takeuchi, Masahiro; Uyama, Yoshiaki; Zhao, Lihui; Skali, Hicham; Solomon, Scott; Jacobus, Susanna; Hughes, Michael; Packer, Milton; Wei, Lee Jen.

In: Journal of Clinical Oncology, Vol. 32, No. 22, 01.08.2014, p. 2380-2385.

Research output: Contribution to journalArticle

Uno, H, Claggett, B, Tian, L, Inoue, E, Gallo, P, Miyata, T, Schrag, D, Takeuchi, M, Uyama, Y, Zhao, L, Skali, H, Solomon, S, Jacobus, S, Hughes, M, Packer, M & Wei, LJ 2014, 'Moving beyond the hazard ratio in quantifying the between-group difference in survival analysis', Journal of Clinical Oncology, vol. 32, no. 22, pp. 2380-2385. https://doi.org/10.1200/JCO.2014.55.2208
Uno, Hajime ; Claggett, Brian ; Tian, Lu ; Inoue, Eisuke ; Gallo, Paul ; Miyata, Toshio ; Schrag, Deborah ; Takeuchi, Masahiro ; Uyama, Yoshiaki ; Zhao, Lihui ; Skali, Hicham ; Solomon, Scott ; Jacobus, Susanna ; Hughes, Michael ; Packer, Milton ; Wei, Lee Jen. / Moving beyond the hazard ratio in quantifying the between-group difference in survival analysis. In: Journal of Clinical Oncology. 2014 ; Vol. 32, No. 22. pp. 2380-2385.
@article{246112c325e448eb9200c7fb684edf7a,
title = "Moving beyond the hazard ratio in quantifying the between-group difference in survival analysis",
abstract = "In a longitudinal clinical study to compare two groups, the primary end point is often the time to a specific event (eg, disease progression, death). The hazard ratio estimate is routinely used to empirically quantify the between-group difference under the assumption that the ratio of the two hazard functions is approximately constant over time. When this assumption is plausible, such a ratio estimate may capture the relative difference between two survival curves. However, the clinical meaning of such a ratio estimate is difficult, if not impossible, to interpret when the underlying proportional hazards assumption is violated (ie, the hazard ratio is not constant over time). Although this issue has been studied extensively and various alternatives to the hazard ratio estimator have been discussed in the statistical literature, such crucial information does not seem to have reached the broader community of health science researchers. In this article, we summarize several critical concerns regarding this conventional practice and discuss various well-known alternatives for quantifying the underlying differences between groups with respect to a time-to-event end point. The data from three recent cancer clinical trials, which reflect a variety of scenarios, are used throughout to illustrate our discussions. When there is not sufficient information about the profile of the between-group difference at the design stage of the study, we encourage practitioners to consider a prespecified, clinically meaningful, model-free measure for quantifying the difference and to use robust estimation procedures to draw primary inferences.",
author = "Hajime Uno and Brian Claggett and Lu Tian and Eisuke Inoue and Paul Gallo and Toshio Miyata and Deborah Schrag and Masahiro Takeuchi and Yoshiaki Uyama and Lihui Zhao and Hicham Skali and Scott Solomon and Susanna Jacobus and Michael Hughes and Milton Packer and Wei, {Lee Jen}",
year = "2014",
month = "8",
day = "1",
doi = "10.1200/JCO.2014.55.2208",
language = "English (US)",
volume = "32",
pages = "2380--2385",
journal = "Journal of Clinical Oncology",
issn = "0732-183X",
publisher = "American Society of Clinical Oncology",
number = "22",

}

TY - JOUR

T1 - Moving beyond the hazard ratio in quantifying the between-group difference in survival analysis

AU - Uno, Hajime

AU - Claggett, Brian

AU - Tian, Lu

AU - Inoue, Eisuke

AU - Gallo, Paul

AU - Miyata, Toshio

AU - Schrag, Deborah

AU - Takeuchi, Masahiro

AU - Uyama, Yoshiaki

AU - Zhao, Lihui

AU - Skali, Hicham

AU - Solomon, Scott

AU - Jacobus, Susanna

AU - Hughes, Michael

AU - Packer, Milton

AU - Wei, Lee Jen

PY - 2014/8/1

Y1 - 2014/8/1

N2 - In a longitudinal clinical study to compare two groups, the primary end point is often the time to a specific event (eg, disease progression, death). The hazard ratio estimate is routinely used to empirically quantify the between-group difference under the assumption that the ratio of the two hazard functions is approximately constant over time. When this assumption is plausible, such a ratio estimate may capture the relative difference between two survival curves. However, the clinical meaning of such a ratio estimate is difficult, if not impossible, to interpret when the underlying proportional hazards assumption is violated (ie, the hazard ratio is not constant over time). Although this issue has been studied extensively and various alternatives to the hazard ratio estimator have been discussed in the statistical literature, such crucial information does not seem to have reached the broader community of health science researchers. In this article, we summarize several critical concerns regarding this conventional practice and discuss various well-known alternatives for quantifying the underlying differences between groups with respect to a time-to-event end point. The data from three recent cancer clinical trials, which reflect a variety of scenarios, are used throughout to illustrate our discussions. When there is not sufficient information about the profile of the between-group difference at the design stage of the study, we encourage practitioners to consider a prespecified, clinically meaningful, model-free measure for quantifying the difference and to use robust estimation procedures to draw primary inferences.

AB - In a longitudinal clinical study to compare two groups, the primary end point is often the time to a specific event (eg, disease progression, death). The hazard ratio estimate is routinely used to empirically quantify the between-group difference under the assumption that the ratio of the two hazard functions is approximately constant over time. When this assumption is plausible, such a ratio estimate may capture the relative difference between two survival curves. However, the clinical meaning of such a ratio estimate is difficult, if not impossible, to interpret when the underlying proportional hazards assumption is violated (ie, the hazard ratio is not constant over time). Although this issue has been studied extensively and various alternatives to the hazard ratio estimator have been discussed in the statistical literature, such crucial information does not seem to have reached the broader community of health science researchers. In this article, we summarize several critical concerns regarding this conventional practice and discuss various well-known alternatives for quantifying the underlying differences between groups with respect to a time-to-event end point. The data from three recent cancer clinical trials, which reflect a variety of scenarios, are used throughout to illustrate our discussions. When there is not sufficient information about the profile of the between-group difference at the design stage of the study, we encourage practitioners to consider a prespecified, clinically meaningful, model-free measure for quantifying the difference and to use robust estimation procedures to draw primary inferences.

UR - http://www.scopus.com/inward/record.url?scp=84905841420&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84905841420&partnerID=8YFLogxK

U2 - 10.1200/JCO.2014.55.2208

DO - 10.1200/JCO.2014.55.2208

M3 - Article

C2 - 24982461

AN - SCOPUS:84905841420

VL - 32

SP - 2380

EP - 2385

JO - Journal of Clinical Oncology

JF - Journal of Clinical Oncology

SN - 0732-183X

IS - 22

ER -