Statistical Science

Understanding the shape of the hazard rate: a process point of view (With comments and a rejoinder by the authors)

Odd O. Aalen and Håkon K. Gjessing

Full-text: Open access


Survival analysis as used in the medical context is focused on the concepts of survival function and hazard rate, the latter of these being the basis both for the Cox regression model and of the counting process approach. In spite of apparent simplicity, hazard rate is really an elusive concept, especially when one tries to interpret its shape considered as a function of time. It is then helpful to consider the hazard rate from a different point of view than what is common, and we will here consider survival times modeled as first passage times in stochastic processes. The concept of quasistationary distribution,which is a well-defined entity for various Markov processes, will turn out to be useful.

We study these matters for a number of Markov processes, including the following: finite Markov chains; birth-death processes; Wiener processes with and without randomization of parameters; and general diffusion processes. An example of regression of survival data with a mixed inverse Gaussian distribution is presented.

The idea of viewing survival times as first passage times has been much studied by Whitmore and others in the context of Wiener processes and inverse Gaussian distributions. These ideas have been in the background compared to more popular appoaches to survival data, at least within the field of biostatistics,but deserve more attention.

Article information

Statist. Sci., Volume 16, Number 1 (2001), 1-22.

First available in Project Euclid: 27 August 2001

Permanent link to this document

Digital Object Identifier

Mathematical Reviews number (MathSciNet)

Zentralblatt MATH identifier

First passage time hazard rate survival analysis quasistationary distribution Wiener process Markov chain


Aalen, Odd O.; Gjessing, Håkon K. Understanding the shape of the hazard rate: a process point of view (With comments and a rejoinder by the authors). Statist. Sci. 16 (2001), no. 1, 1--22. doi:10.1214/ss/998929473.

Export citation


  • Aalen, O. O. (1992). Modellingheterogeneity in survival analysis by the compound Poisson distribution. Ann. Appl. Probab. 2 951-972.
  • Aalen, O. O. (1994). Effects of frailty in survival analysis. Statist. Methods Medical Res. 3 227-243.
  • Aalen, O. O. (1995). Phase type distributions in survival analysis. Scand. J. Statist. 22 447-463. Aalen, O. O., Farewell, V. T., DeAngelis, D., Day, N. E. and
  • Gill, O. N. (1997). A Markov model for HIV disease progression including the effect of HIV diagnosis and treatment: application to AIDS prediction in England and Wales. Statistics in Medicine 16 2191-2210.
  • Cavender, J. A. (1978). Quasi-stationary distributions of birthand-death processes. Adv. Appl. Probab. 10 570-586.
  • Chhikara, R. S. and Folks, J. L. (1989). The Inverse Gaussian Distribution: Theory, Methodology and Applications. Dekker, New York.
  • Cox, D. R. and Miller, H. D. (1965). The Theory of Stochastic Processes. Methuen, London.
  • Doksum, K. A. and Høyland, A. (1992). Models for variablestress accelerated life testingexperiments based on Wiener processes and the inverse gaussian distribution. Technometrics 34 74-82.
  • Eaton, W. W. and Whitmore, G. A. (1977). Length of stay as a stochastic process: a general approach and application to hospitalization for schizophrenia. J. Math. Sociol. 5 273-292.
  • Goel, N. S. and Richter-Dyn, N. (1974). Stochastic Models in Biology. Academic Press, New York.
  • Griffiths, W. S. (1988). Shock models. In Encyclopedia of Statistical Sciences 8 442-446. Wiley, New York.
  • Gross, D. and Harris, C. M. (1985). Fundamentals of Queuing Theory, 2nd ed. Wiley, New York.
  • Hougaard, P. (1986). Survival models for heterogeneous populations derived from stable distributions. Biometrika 73 387-396.
  • Jewell, N. P. and Kalbfleisch, J. D. (1996). Marker processes in survival analysis. Lifetime Data Anal. 2 15-29.
  • Kalbfleisch, J. D. and Prentice, R. L. (1980). The Statistical Analysis of Failure Time Data. Wiley, New York.
  • Karlin, S. and Taylor, H. M. (1981). A Second Course in Stochastic Processes. Academic Press, New York.
  • Keiding, N., Andersen, P. K. and Klein, J. P. (1997). The role of frailty models and accelerated failure time models in describingheterogeneity due to omitted covariates. Statistics in Medicine 16 215-224.
  • Keilson, J. (1979). Markov Chain Models-Rarity and Exponentiality. Springer, New York.
  • Lancaster, D. J. (1982). Stochastic Models for Social Processes, 3rd ed. Wiley, New York. Longini, I. M., Clark, W. S., Byers, R. H., Ward, J. W., Dar
  • Row, W. W., Lemp, G. F. and Hethcote, H. W. (1989). Statistical analysis of the stages of HIV-infections using a Markov model. Statistics in Medicine 8 831-843.
  • Lynn, N. J. and Singpurwalla, N. D. (1997). "Burn-in" makes us feel good. Comment on "Burn-in," by H. W. Block and T. H. Savits. Statist. Sci. 12 13-19.
  • Martinez, S. and San Martin, J. (1994). Quasi-stationary distributions for a Brownian motion with drift and associated limit laws. J. Appl. Probab. 31 911-920.
  • Nielsen, J. P. and Linton, O. B. (1995). Kernel estimation in a nonparametric marker dependent hazard model. Ann. Statist. 23 1735-1748.
  • O'Cinneide, C. A. (1990). Characterization of phase type distributions. Comm. Statist. Stochastic Models 6 1-57.
  • Øksendal, B. (1998). Stochastic Differential Equations, 5th ed. Springer, New York.
  • Patel, J. K. (1983). Hazard rate and other classifications of distributions. In Encyclopedia of Statistical Sciences 3. Wiley, New York. pages: 590-594.
  • Shaked, M. and Shanthikumar, J. G. (1991). Shock models with MIFRA time to failure distributions. J. Statist. Plann. Inference 29 157-169.
  • Schwarz, W. (1992). The Wiener process between a reflecting and an absorbingbarrier. J. Appl. Probab. 29 597-604.
  • Seshadri, V. (1998). The Inverse Gaussian Distribution: Statistical Theory and Applications. Springer, New York. Whitmore, G. A. (1986a). First-passage-time models for duration data: regression structures and competing risks. The Statistician 35 207-219. Whitmore, G. A. (1986b). Normal-gamma mixtures of inverse Gaussian distributions. Scand. J. Statist. 13 211-220.
  • Whitmore, G. A. (1995). Estimatingdeg radation by a Wiener diffusion process. Lifetime Data Anal. 1 307-319.
  • Whitmore, G. A., Crowder, M. J. and Lawless, J. F. (1998). Failure inference from a marker process based on a bivariate Wiener model. Lifetime Data Anal. 4 229-251.
  • Bagdonavicius, V. B. and Nikulin, M. S. (1999). Generalized proportional hazards model based on modified partial likelihood. Lifetime Data Anal. 5 329-350.
  • Barlow, R. E. and Proschan, F. (1975). Statistical Theory of Reliability. Holt, Rinehart & Winston, New York.
  • Block, H. and Joe, H. (1997). Tale behavior of the failure rate functions of mixtures. Lifetime Data Anal. 3 269-288.
  • Block, H. W., Mi, J. and Savits, T. H. (1993). Burn-in and mixed populations. J. Appl. Probab. 30 692-702.
  • Cox, D. R. (1972). Regression models and life tables (with discussion). J. Roy. Statist. Soc. Ser. B 34 187-220.
  • Finkelstein, M. S. and Essaoulova, V. (2000). Modelinga failure rate for a mixture of distribution functions. Technical Report 279, Univ. Orange Free State.
  • Gurland, J. and Sethuraman, J. (1994). Reversal of increasingfailure rates when poolingfailure data. Technometrics 36 416-418.
  • Gurland, J. and Sethuraman, J. (1995). How poolingfailure data may reverse increasingfailure rates. J. Amer. Statist. Assoc. 90 1416-1423.
  • Hougaard, P., Myglegaard, P. and Borch-Johnsen, K. (1994). Heterogeneity models of disease susceptibility, with application to diabetic nephropathy. Biometrics 50 1178-1188.
  • Proschan, F. (1963). Theoretical explanation of observed decreasingfailure rate. Technometrics 5 373-383.
  • Robins, J. and Greenland, S. (1989). The probability of causation under a stochastic model for individual risk. Biometrics 45 1125-1138.
  • Robins, J. and Greenland, S. (1991). Estimability and estimation of expected years of life lost due to a hazardous exposure. Statistics in Medicine 10 79-93.
  • Scheike, T. H., Petersen, J. H. and Martinussen, T. (1999). Retrospective ascertainment of recurrent events: an application to time to pregnancy. J. Amer. Statist. Assoc. 94 713-725.
  • Tetens, J. N. (1786). Einleitung zur Berechnung der Leibrenten und Anwarschaften II. Weidmanns Erben und Reich, Leipzig.
  • Vaupel, J. W., Manton, K. G. and Stallard, E. (1979). The impact of heterogeneity in individual frailty on the dynamics of mortality. Demography 16 439-454.
  • Vaupel, T. W. and Yashin, A. I. (1985). Heterogeneity's ruse: some surprisingeffects of selection on population dynamics. Amer. Statist. 39 176-185.
  • Wang, J., Muller, H. and Capra, W. B. (1998). Analysis of oldest-old mortality: lifetimes revisited. Ann. Statist. 26 126-133.