Abstract
The hazard ratio is one of the most commonly reported measures of treatment effect in randomised trials, yet the source of much misinterpretation. This point was made clear by Hernán (Epidemiology (Cambridge, Mass) 21(1):13–15, 2010) in a commentary, which emphasised that the hazard ratio contrasts populations of treated and untreated individuals who survived a given period of time, populations that will typically fail to be comparable—even in a randomised trial—as a result of different pressures or intensities acting on different populations. The commentary has been very influential, but also a source of surprise and confusion. In this note, we aim to provide more insight into the subtle interpretation of hazard ratios and differences, by investigating in particular what can be learned about a treatment effect from the hazard ratio becoming 1 (or the hazard difference 0) after a certain period of time. We further define a hazard ratio that has a causal interpretation and study its relationship to the Cox hazard ratio, and we also define a causal hazard difference. These quantities are of theoretical interest only, however, since they rely on assumptions that cannot be empirically evaluated. Throughout, we will focus on the analysis of randomised experiments.
Similar content being viewed by others
References
Aalen OO, Cook RJ, Røysland K (2015) Does cox analysis of a randomized survival study yield a causal treatment effect? Lifetime Data Anal 21(4):579–93
Bartolucci F, Grilli L (2011) Modeling partial compliance through copulas in a principal stratification framework. J Am Stat Assoc 106(494):469–479
Collett D (2015) Modelling survival data in medical research. CRC Press, Boca Raton
Frangakis CE, Rubin DB (2002) Principal stratification in causal inference. Biometrics 58(1):21–9
Hernán M (2010) The hazards of hazard ratios. Epidemiology (Cambridge, Mass) 21(1):13–15
Lederle FA, Kyriakides TC, Stroupe KT, Freischlag JA, Padberg FT, Matsumura JS, Huo Z, Johnson GR (2019) Open versus endovascular repair of abdominal aortic aneurysm. N Engl J Med 380(22):2126–2135 PMID: 31141634
Lin DY, Wei LJ, Ying Z (1993) Checking the Cox model with cumulative sums of martingale-based residuals. Biometrika 80:557–572
Martinussen T, Scheike TH (2006) Dynamic regression models for survival data. Springer, New York
Martinussen T, Vansteelandt S (2013) On collapsibility and confounding bias in cox and aalen regression models. Lifetime Data Anal 19(3):279–96
Nelsen RB (2006) An introduction to copulas. Springer, New York
Oakes D (1989) Bivariate survival models induced by frailties. J Am Stat Assoc 84(406):487–493
Primrose JN, Fox RP, Palmer DH, Malik HZ, Prasad R, Mirza D, Anthony A, Corrie P, Falk S, Finch-Jones M, Wasan H, Ross P, Wall L, Wadsley J, Evans JTR, Stocken D, Praseedom R, Ma YT, Davidson B, Neoptolemos JP, Iveson T, Raftery J, Zhu S, Cunningham D, Garden OJ, Stubbs C, Valle JW, Bridgewater J, Primrose J, Fox R, Morement H, Chan O, Rees C, Ma Y, Hickish T, Falk S, Finch-Jones M, Pope I, Corrie P, Crosby T, Sothi S, Sharkland K, Adamson D, Wall L, Evans J, Dent J, Hombaiah U, Iwuji C, Anthoney A, Bridgewater J, Cunningham D, Gillmore R, Ross P, Slater S, Wasan H, Waters J, Valle J, Palmer D, Malik H, Neoptolemos J, Faluyi O, Sumpter K, Dernedde U, Maduhusudan S, Cogill G, Archer C, Iveson T, Wadsley J, Darby S, Peterson M, Mukhtar A, Thorpe J, Bateman A, Tsang D, Cummins S, Nolan L, Beaumont E, Prasad R, Mirza D, Stocken D, Praseedom R, Davidson B, Raftery J, Zhu S, Garden J, Stubbs C, Coxon F (2019) Capecitabine compared with observation in resected biliary tract cancer (bilcap): a randomised, controlled, multicentre, phase 3 study. Lancet Oncol 20(5):663–673
Rubin DB (2000) Causal inference without counterfactuals: comment. J Am Stat Assoc 95(450):435–438
Stablein DM, Koutrouvelis IA (1985) A two-sample test sensitive to crossing hazards in uncensored and singly censored data. Biometrics 41(3):643–52
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 L-J (2014) Moving beyond the hazard ratio in quantifying the between-group difference in survival analysis. J Clin Oncol 32(22):2380–5
Vansteelandt S, Martinussen T, Tchetgen ET (2014) On adjustment for auxiliary covariates in additive hazard models for the analysis of randomized experiments. Biometrika 101(1):237–244
White IR, Royston P (2009) Imputing missing covariate values for the cox model. Stat Med 28(15):1982–98
Zhao L, Claggett B, Tian L, Uno H, Pfeffer MA, Solomon SD, Trippa L, Wei LJ (2016) On the restricted mean survival time curve in survival analysis. Biometrics 72(1):215–21
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Electronic supplementary material
Below is the link to the electronic supplementary material.
Appendix A
Appendix A
1.1 A.1 Binary frailty variable
Let Z be binary, for instance \(P(Z=0.2)=0.2\) and \(P(Z=1.2)=0.8\), low and high risk groups, then \(E(Z)=1\), and similar results as in the Gamma-distribution case (Sect. 3.2) are obtained. In this binary case, the Laplace transform is
Therefore
where
which is an increasing function. Hence, if we take \(\beta _1<0\) and \(\beta _2=0\), then again
for all t.
1.2 A.2 Frailty model arising by marginalization
The DGP (4) given by \(\lambda (t;A,Z)\) with \(\nu =\infty \) induces the marginal Cox model for the observed data where we do not get access to Z. It is tempting to interpret \(\text{ HR}_{Z}(t)\) as a causal hazard ratio, but this only holds under further untestable assumptions as shown in Sect. 4.1. Below, we formulate a more general DGP involving an additional frailty variable so that both the marginal Cox model and the model \(\lambda (t;A,Z)\) are correctly specified.
Rename Z to \(Z_1\). We then show that similarly we can also pick a DGP \(\lambda (t;A,Z_1,Z_2)\) so that it marginalizes to \(\lambda (t;A,Z_1)\) that further marginalizes to \(\lambda (t;A)\), the latter being the Cox model. For ease of calculations, let
with \(Z_2\) being Gamma distributed with mean and variance equal to 1, and independent of \(Z_1\) and A. Similar calculations as those in Sect. 3.1 gives the following expression
Hence, if the DGP is governed by (18) then model (4), with \(\nu =\infty \), and the marginal Cox model, only conditioning on A, are also correctly specified.
1.3 A.3 Hazard differences
and
and therefore
1.4 A.4 \(\text{ HR }(t)\) under copulas not obeying (11)
We outline now how to generate samples from the copula \(C^{*}\) to gain further insight, Nelsen (2006), pp. 40–41,
-
1.
Generate a standard uniformly distributed random variable u;
-
2.
Set
$$\begin{aligned} v = \left\{ \begin{array}{ll} u&{} \quad \text {if}\ u\le \xi { or}u\ge 1-\xi \\ 1-u &{} \quad \text {if}\ \xi<u<1-\xi \end{array}\right. \end{aligned}$$ -
3.
Set \(T^0=\Lambda _0^{-1}\{-\log {(u)}\}\) and \(T^1=\Lambda _0^{-1}\{-\log {(v)}/\phi \}\) (assuming \(\Lambda _0\) is strictly increasing).
As an illustration, we took \(\Lambda _0(t)=t\), \(\phi =2/3\) and \(\xi =0.251\) and generated \(n=2,000,000\) samples. For these simulated data we obtained an estimated Cox HR of 0.666 and Spearman’s \(\rho \) was estimated to equal 0.753, which fits nicely with the theoretical values. As mentioned, if we follow the path \((\exp {\{-\Lambda _0(t)\}},\exp {\{-\phi \Lambda _0(t)\}})\) (starting at (1,1) when \(t=0\)) in \([0,1]^2\) in such a way that we pass through the upper triangle of \(J_2\) then \(\text{ HR }(t)\) will be equal to 0 before we reach \(J_2\) but then it jumps to \(k(u,v)\phi \), where \(k(u,v)>1\). To illustrate this, we estimated \(\text{ HR }(t)\) non-parametrically in time-points t until we cross the line \(v=1-u\) in \(J_2\). This was done by taking the observed frequency of \(T^1\)-failures in the time interval \([t,t+\delta ]\), \(\delta =0.01\), among those where both \(T^0\ge t\) and \(T^1\ge t\); and similarly for \(T^0\); and then taking the ratio of the two. This is seen in Fig. 3, where the theoretical values \( \phi \exp {\{-t(\phi -1) \}}\) are superimposed. The dashed line corresponds to \(\phi =2/3\). We see that the theoretical values fit nicely to the estimated ones. For this copula it can be shown that \(\text{ HR }(t)\) is always below 1, but as we have just seen it can also be larger than \(\phi \)! Hence, to have that \(\text{ HR }(t)<\phi \) we need more than just positive correlation between \(T^0\) and \(T^1\) as this example illustrates.
1.5 A.5 Monotonicity and \(\text{ HR }(t)\)
Assume that \(T^1\ge T^0\) a.s. and that the distribution of \((T^0,T^1)\) is absolutely continuous with density \(f(t_0,t_1)\) with respect to the Lebesgue measure. Monotonicity is attained if \(f(t_0,t_1)=0\) for \(t_0>t_1\). We now look at
Hence
where
But this means \(\lambda _{T^1}(t|T^0\ge t,T^1\ge t)=0\), and therefore \(\text{ HR }(t)=0\). To avoid having \(\lambda _{T^1}(t|T^0\ge t,T^1\ge t)\) equal to zero we need for instance \(T^0\) given \(T^1\) to have a discrete distribution, but this seems unappealing.
Rights and permissions
About this article
Cite this article
Martinussen, T., Vansteelandt, S. & Andersen, P.K. Subtleties in the interpretation of hazard contrasts. Lifetime Data Anal 26, 833–855 (2020). https://doi.org/10.1007/s10985-020-09501-5
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10985-020-09501-5