Abstract
Time-to-event data often violate the proportional hazards assumption inherent in the popular Cox regression model. Such violations are especially common in the sphere of biological and medical data where latent heterogeneity due to unmeasured covariates or time varying effects are common. A variety of parametric survival models have been proposed in the literature which make more appropriate assumptions on the hazard function, at least for certain applications. One such model is derived from the First Hitting Time (FHT) paradigm which assumes that a subject’s event time is determined by a latent stochastic process reaching a threshold value. Several random effects specifications of the FHT model have also been proposed which allow for better modeling of data with unmeasured covariates. While often appropriate, these methods often display limited flexibility due to their inability to model a wide range of heterogeneities. To address this issue, we propose a Bayesian model which loosens assumptions on the mixing distribution inherent in the random effects FHT models currently in use. We demonstrate via simulation study that the proposed model greatly improves both survival and parameter estimation in the presence of latent heterogeneity. We also apply the proposed methodology to data from a toxicology/carcinogenicity study which exhibits nonproportional hazards and contrast the results with both the Cox model and two popular FHT models.
Similar content being viewed by others
References
Aalen O, Gjessing H (2001) Understanding the shape of the hazard rate: a process point of view (with comments and a rejoinder by the authors). Statist Sci 16(1):1–22. https://doi.org/10.1214/ss/998929473
Aalen O, Borgan O, Gjessing H (2008) Survival and event history analysis: a process point of view. Statistics for biology and health. Springer, New York
Caroni C (2017) First hitting time regression models. Wiley-Blackwell. https://doi.org/10.1002/9781119437260.ch2
Chhikara RS, Folks LJ (1989) The inverse gaussian distribution: theory, methodology, and applications. Marcel Dekker Inc, New York
Choi S, Huang X, Cormier J, Doksum K (2014) A semiparametric inverse-Gaussian model and inference for survival data with a cured proportion. Can J Stat 42(4):635–649. https://doi.org/10.1002/cjs.11226
Cox DR (1972) Regression models and life-tables. J R Stat Soc Ser B (Methodological) 34(2):187–220
Eaton W, Whitmore G (1977) Length of stay as a stochastic process: a general approach and application to hospitalization for schizophrenia. J Math Soc 5(2):273–292. https://doi.org/10.1080/0022250X.1977.9989877
Economou P, Malefaki S, Caroni C (2015) Bayesian threshold regression model with random effects for recurrent events. Methodol Comput Appl Prob 17(4):871–898. https://doi.org/10.1007/s11009-015-9445-8
Erich R, Pennell M (2015) Ornstein–Uhlenbeck threshold regression for time-to-event data with and without a cure fraction. Lifetime Data Anal 21(1):1–19
Gelfand A, Dey D (1994) Bayesian model choice: asymptotics and exact calculations. J R Stat Soc Ser B (Methodological) 56(3):501–514
Hastie T, Tibshirani R (1990) Exploring the nature of covariate effect in the proportional hazards model. Biometrics 46:1005–1016
Hougaard P (1991) Modeling heterogeneity in survival data. J Appl Probab. https://doi.org/10.2307/3214503
Ishwaran H, James LF (2001) Gibbs sampling methods for stick-breaking priors. J Am Stat Assoc 96(453):161–173. https://doi.org/10.1198/016214501750332758
Ishwaran H, Zarepour M (2000) Markov chain Monte Carlo in approximate Dirichlet and beta two-parameter process hierarchical models. Biometrika 87(2):371–390. https://doi.org/10.1093/biomet/87.2.371
Ishwaran H, Zarepour M (2002) Exact and approximate sum representations for the Dirichlet process. Can J Stat 30(2):269–283
Kass R, Raftery A (1995) Bayes factors. J Am Stat Assoc 90(430):773–795. https://doi.org/10.1080/01621459.1995.10476572
Keiding N, Andersen P, Klein J (1997) The role of frailty models and accelerated failure time models in describing heterogeneity due to omitted covariates. Stat Med 16(2):215–224
Klein JP, Moeschberger ML (2003) Survival analysis: techniques for censored and truncated data. Springer, New York
Lancaster T (1972) A stochastic model for the duration of a strike. J R Stat Soc Ser A (General) 135(2):257–271
Lee M, Whitmore G (2006) Threshold regression for survival analysis: modeling event times by a stochastic process reaching a boundary. Stat Sci 21(4):501–513. https://doi.org/10.1214/088342306000000330
Lee M, Chang M, Whitmore G (2008) A threshold regression mixture model for assessing treatment efficacy in a multiple myeloma clinical trial. J Biopharm Stat 18(6):1136–1149
Li J, Lee M (2011) Analysis of failure time using threshold regression with semi-parametric varying coefficients. Stat Neerl 65(2):164–182. https://doi.org/10.1111/j.1467-9574.2011.00481.x
National Toxicology Program (2004) Technical report on the toxicology and carcinogenesis studies of urethane, ethanol, and urethane/ethanol in B6CF3F1. Department of Health & Human Services, Public Health Service, Public Health Service, National Institutes of Health
Pennell M, Whitmore GA, Lee MLT (2010) Bayesian random-effects threshold regression with application to survival data with nonproportional hazards. Biostatistics 11(1):111–126. https://doi.org/10.1093/biostatistics/kxp041
Stogiannis D, Caroni C (2013) Issues in fitting inverse Gaussian first hitting time regression models for lifetime data. Commun Stat Simul Comput 42(9):1948–1960. https://doi.org/10.1080/03610918.2012.687061
Stone M (1977) An asymptotic equivalence of choice of model by cross-validation and Akaike’s criterion. J R Stat Soc Ser B (Methodological) 39(1):44–47
Tian L, Zucker D, Wei LJ (2005) On the Cox model with time-varying regression coefficients. J Am Stat Assoc 100:172–183
Tong X, He X, Sun J, Lee ML (2008) Joint analysis of current status and marker data: an extension of a bivariate threshold model. Int J Biostat 4:1122. https://doi.org/10.2202/1557-4679.1122
Tweedie M (1957) Statistical properties of inverse Gaussian distributions. i. Ann Math Stat 28(2):362–377. https://doi.org/10.1214/aoms/1177706964
Wang L, Dunson D (2011) Fast Bayesian inference in Dirichlet process mixture models. J Comput Graph Stat 20:196–216
West M, Mueller P, Escobar MD (1994) Hierarchical priors and mixture models, with application in regression and density estimation. In: Freeman PR, Smith AFM (eds) Aspects of uncertainty: a tribute to D.V. Lindley. John Wiley and Sons, Chichester, p 363–386
Whitmore G (1975) The inverse Gaussian distribution as a model of hospital stay. Health Serv Res 10(3):297–302
Whitmore G, Su Y (2007) Modeling low birth weights using threshold regression: results for US birth data. Lifetime Data Anal 13(2):161–190. https://doi.org/10.1007/s10985-006-9032-y
Whitmore G, Crowder M, Lawless J (1998) Failure inference from a marker process based on a bivariate Wiener model. Lifetime Data Anal 4(3):229–251. https://doi.org/10.1023/A:1009617814586
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.
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
About this article
Cite this article
Race, J.A., Pennell, M.L. Semi-parametric survival analysis via Dirichlet process mixtures of the First Hitting Time model. Lifetime Data Anal 27, 177–194 (2021). https://doi.org/10.1007/s10985-020-09514-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10985-020-09514-0