Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2014 Aug 1;32(22):2380-5.
doi: 10.1200/JCO.2014.55.2208. Epub 2014 Jun 30.

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

Hajime Uno  1 , Brian Claggett  1 , Lu Tian  1 , Eisuke Inoue  1 , Paul Gallo  1 , Toshio Miyata  1 , Deborah Schrag  1 , Masahiro Takeuchi  1 , Yoshiaki Uyama  1 , Lihui Zhao  1 , Hicham Skali  1 , Scott Solomon  1 , Susanna Jacobus  1 , Michael Hughes  1 , Milton Packer  1 , Lee-Jen Wei  2
Affiliations

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

Hajime Uno et al. J Clin Oncol. .

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.

PubMed Disclaimer

Conflict of interest statement

Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.

Figures

Fig 1.
Fig 1.
Estimated survival curves, hazard ratio, and restricted mean survival times with data from the Eastern Cooperative Oncology Group E4A03 study. (A) Kaplan-Meier curves for low-dose (blue) and high-dose (gold) groups. (B) Estimate of the ratio of hazard functions (low dose over high dose) over time and corresponding 0.95 point-wise confidence band. (C) Estimate of restricted mean survival time (blue area) and the restricted mean time lost (gray area) up to 40 months for the low-dose group.
Fig 2.
Fig 2.
Estimated survival curves and the hazard ratio with reconstructed data for comparing single-agent pemetrexed (P) with carboplatin plus pemetrexed (CP) in patients with advanced non–small-cell lung cancer. (A) Kaplan-Meier curves for CP (blue) and P (gold). (B) Estimate of the ratio of hazard functions (CP over P) over time and the corresponding 0.95 point-wise confidence band.
Fig 3.
Fig 3.
Estimated survival curves and the hazard ratio with reconstructed data for comparing modified FOLFOX6 (mFF6) plus bevacizumab (Bev) with mFF6 in patients with stage II to III colon cancer. (A) Estimated survival curves for mFF6 + Bev (blue) and mFF6 (gold). (B) Estimate of the ratio of hazard functions (mFF6 + Bev over mFF6) and the corresponding 0.95 point-wise confidence band.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Kalbfleisch JD, Prentice RL. The Statistical Analysis of Failure Time Data. New York, NY: John Wiley & Sons; 1980.
    1. Cox DR, Oakes D. Analysis of Survival Data. London, United Kingdom: Chapman and Hall; 1984.
    1. Cox DR. Regression models and life tables. J R Stat Soc B. 1972;34:187–220.
    1. Moher D, Hopewell S, Schulz KF, et al. CONSORT 2010 explanation and elaboration: Updated guidelines for reporting parallel group randomised trials. BMJ. 2010;340:c869. - PMC - PubMed
    1. Higgins JP, Green S, editors. Cochrane Handbook for Systematic Reviews of Interventions. New York, NY: John Wiley & Sons; 2008.

Publication types