A parametric dynamic survival model applied to breast cancer survival times

Much current analysis of cancer registry data uses the semiparametric proportional hazards Cox model. In this paper, the time-dependent effect of various prognostic indicators on breast cancer survival times from the West Midlands Cancer Intelligence Unit are investigated. Using Bayesian methodology and Markov chain Monte Carlo estimation methods, we develop a parametric dynamic survival model which avoids the proportional hazards assumption. The model has close links to that developed by both Gamerman and Sinha and co-workers: the log-base-line hazard and covariate effects are piecewise constant functions, related between intervals by a simple stochastic evolution process. Here this evolution is assigned a parametric distribution, with a variance that is further included as a hyperparameter. To avoid problems of convergence within the Gibbs sampler, we consider using a reparameterization. It is found that, for some of the prognostic indicators considered, the estimated effects change with increasing follow-up time. In general those prognostic indicators which are thought to be representative of the most hazardous groups (late-staged tumour and oldest age group) have a declining effect. Copyright 2002 Royal Statistical Society.

[1]  Joseph G. Ibrahim,et al.  Bayesian Survival Analysis , 2004 .

[2]  I. Bleiweiss,et al.  Stage 0 to stage III breast cancer in young women. , 2000, Journal of the American College of Surgeons.

[3]  M. Coleman,et al.  Completeness of cancer registration: a new method for routine use , 2000, British Journal of Cancer.

[4]  Karla Hemming,et al.  Parametric dynamic survival models , 2000 .

[5]  D Sinha,et al.  Bayesian Analysis and Model Selection for Interval‐Censored Survival Data , 1999, Biometrics.

[6]  Eric R. Ziegel,et al.  Practical Nonparametric and Semiparametric Bayesian Statistics , 1998, Technometrics.

[7]  D. Gamerman.  Markov chain Monte Carlo for dynamic generalised linear models , 1998 .

[8]  Sylvia Richardson,et al.  Markov Chain Monte Carlo in Practice , 1997 .

[9]  D J Sargent,et al.  A Flexible Approach to Time-varying Coefficients in the Cox Regression Setting , 1997, Lifetime data analysis.

[10]  A. Gulsvik,et al.  The semi-proportional hazards model revisited: practical reparametrizations. , 1996, Statistics in medicine.

[11]  D. S. James,et al.  Information on death certificates: cause for concern? , 1996, Journal of clinical pathology.

[12]  J. Heikkinen,et al.  An algorithm for nonparametric Bayesian estimation of a Poisson intensity , 1996 .

[13]  H C van Houwelingen,et al.  Time-dependent effects of fixed covariates in Cox regression. , 1995, Biometrics.

[14]  A. Gelfand,et al.  Bayesian analysis of proportional hazards models built from monotone functions. , 1995, Biometrics.

[15]  R. Tibshirani,et al.  Generalized additive models for medical research , 1995, Statistical methods in medical research.

[16]  L. Tierney.  Markov Chains for Exploring Posterior Distributions , 1994 .

[17]  P. Grambsch,et al.  Proportional hazards tests and diagnostics based on weighted residuals , 1994 .

[18]  M. Newton.  Approximate Bayesian-inference With the Weighted Likelihood Bootstrap , 1994 .

[19]  Niels Keiding,et al.  Statistical Models Based on Counting Processes , 1993 .

[20]  Robert Gray,et al.  Flexible Methods for Analyzing Survival Data Using Splines, with Applications to Breast Cancer Prognosis , 1992 .

[21]  D G Clayton,et al.  A Monte Carlo method for Bayesian inference in frailty models. , 1991, Biometrics.

[22]  D. Gamerman.  Dynamic Bayesian Models for Survival Data , 1991 .

[23]  J. Estève,et al.  Relative survival and the estimation of net survival: elements for further discussion. , 1990, Statistics in medicine.

[24]  P. Townsend,et al.  Health and Deprivation: Inequality and the North , 1987 .

[25]  C. N. Morris,et al.  The calculation of posterior distributions by data augmentation , 1987 .

[26]  M. West,et al.  An application of dynamic survival models in unemployment studies , 1987 .

[27]  Dani Gamerman,et al.  Dynamic analysis of survival models and related processes , 1987 .

[28]  Antonio Ciampi,et al.  Extended hazard regression for censored survival data with covariates : a spline approximation for the baseline hazard function , 1987 .

[29]  G. Casella,et al.  Reconciling Bayesian and Frequentist Evidence in the One-Sided Testing Problem , 1987 .

[30]  J. Haybittle,et al.  Death certification in cancer of the breast. , 1984, British medical journal.

[31]  J. Grant,et al.  Intracranial infection due to mycobacterium bovis in Hodgkin's disease. , 1984, British medical journal.

[32]  Sheila M. Gore,et al.  Statistical models and methods for lifetime data, Jerald F. Lawless, Wiley, New York,1982. Price £D27.25. No. of pages: 580 , 1982 .

[33]  S J Pocock,et al.  Long term survival analysis: the curability of breast cancer. , 1982, Statistics in medicine.

[34]  J. Kalbfleisch.  Non‐Parametric Bayesian Analysis of Survival Time Data , 1978 .

[35]  Tom Leonard.  Density Estimation, Stochastic Processes and Prior Information , 1978 .

[36]  David R. Cox,et al.  Regression models and life tables (with discussion , 1972 .

[37]  F. Ederer,et al.  The relative survival rate: a statistical methodology. , 1961, National Cancer Institute monograph.