Cox Regression in a Markov Renewal Model: An Application to the Analysis of Bone Marrow Transplant Data

By Dabrowska, Dorota M.; Sun, Guo-wen et al. | Journal of the American Statistical Association, September 1994 | Go to article overview

Cox Regression in a Markov Renewal Model: An Application to the Analysis of Bone Marrow Transplant Data


Dabrowska, Dorota M., Sun, Guo-wen, Horowitz, Mary M., Journal of the American Statistical Association


1. INTRODUCTION

We consider estimation and prediction in a Cox proportional hazard model where observations on each individual under study form a Markov renewal process with a finite state space, say {1, . . . , r}. In the absence of covariates, for each subject we observe a (possibly censored) process (T, X) = {([T.sub.n], [X.sub.n]): n [greater than or equal to] 0}, where 0 = [T.sub.0] < [T.sub.1] < [T.sub.2]... are consecutive times of entrances into the states [X.sub.0], [X.sub.1], [X.sub.2], . .. , [X.sub.n] E {1, . . . , r}. Here the sequence X = {[X.sub.n],: n [greater than or equal to] 0} of states visited forms an embedded Markov chain and, given X, the sojourn times [T.sub.1], [T.sub.2] - [T.sub.1], . . . are independent with distributions depending on the adjoining states only. More precisely, (T, X) is a Markov renewal process if the distribution of the sojourn times [T.sub.n+1] - [T.sub.n], n [greater than or equal to] 0 satisfies

(1) Pr([T.sub.n+1] - [T.sub.n] [less than or equal to]X,[X.sub.n+1] =j|[X.sub.0], [T.sub.0],[X.sub.1],[T.sub.1],. . . [X.sub.n], [T.sub.n])

= Pr([T.sub.n+1] - ([T.sub.n] [less than or equal to][chi],[X.sub.n+1] =j|[X.sub.n]).

In medical applications, such a process can be used to model the patient's experience following a treatment in cases when the outcome of the treatment is affected by internal covariates whose values carry information about the patient's survival status. The state space {1, . . . , r} may correspond then to specific stages of the disease or to levels of covariates such as temperature or blood count. By way of (1), the Markov renewal model puts emphasis on the duration times of the consecutive episodes in the patient's post-treatment experience. Conditionally on the sequence of states visited X, these duration times are independent, whereas the intradependence among the times of the occurrence of the successive events is determined by the Markov chain assumption on the X process. Extensions of this model to the Cox regression setting fall into the framework of modulated renewal processes (Cox 1973; Kalbfleisch and Prentice 1980) and assume that the one-step transitions are determined by hazard functions of the form [[lambda].sub.ij](t, Z(t)) = [[alpha].sub.0,ij](L(t))exp{[[beta].sup.T][Z.sub.ij](t)}, where L(t) is the backwards recurrence time (i.e., the time elapsed between t and the last event prior to t) and Z(t) = {[Z.sub.ij](t): i, j [less than or equal to] r} is a vector of transition-specific time-dependent covanates. This model is more parsimonious than the usual Cox gegression model in that it accommodates both the calendar time t and the nonchronological duration time scale corresponding to the backwards recurrence time L(t). Along with Markov renewal processes, special cases of this model include homogeneous Markov chains, Cox regression for multiple decrement models (Aalen 1978), and variants of the Fix and Neyman (1951) model for recovery, relapse, and death due to cancer (Keiding 1990; McKeague and Utikal 1990). The dependence of the baseline hazard functions on the backwards recurrence time L(t) makes this model more difficult to analyze in that the nonchronological duration time clock used leads outside the framework of Aalen's multiplicative intensity models and invalidates the usual martingale methods. Gill (1980), Voelkel and Crowley (1984), Oakes and Cui (1992), and Andersen, Borgan, Gill, and Keiding (1993) have discussed this point in more detail. Here we assume that the covariates Z(t) depend on the backwards recurrence time only. In this case, estimation can be based on Jacod's (1975) likelihood for counting processes and leads to a partial likelihood that functionally is the same as in the case of ordinary Cox regression. In Section 3, a modification of Andersen and Gill's (1982) results is given to justify the asymptotic structure of the estimates. We further consider prediction of the realization of a process for a new subject. …

The rest of this article is only available to active members of Questia

Sign up now for a free, 1-day trial and receive full access to:

  • Questia's entire collection
  • Automatic bibliography creation
  • More helpful research tools like notes, citations, and highlights
  • Ad-free environment

Already a member? Log in now.

Notes for this article

Add a new note
If you are trying to select text to create highlights or citations, remember that you must now click or tap on the first word, and then click or tap on the last word.
One moment ...
Default project is now your active project.
Project items

Items saved from this article

This article has been saved
Highlights (0)
Some of your highlights are legacy items.

Highlights saved before July 30, 2012 will not be displayed on their respective source pages.

You can easily re-create the highlights by opening the book page or article, selecting the text, and clicking “Highlight.”

Citations (0)
Some of your citations are legacy items.

Any citation created before July 30, 2012 will labeled as a “Cited page.” New citations will be saved as cited passages, pages or articles.

We also added the ability to view new citations from your projects or the book or article where you created them.

Notes (0)
Bookmarks (0)

You have no saved items from this article

Project items include:
  • Saved book/article
  • Highlights
  • Quotes/citations
  • Notes
  • Bookmarks
Notes
Cite this article

Cited article

Style
Citations are available only to our active members.
Sign up now to cite pages or passages in MLA, APA and Chicago citation styles.

(Einhorn, 1992, p. 25)

(Einhorn 25)

1

1. Lois J. Einhorn, Abraham Lincoln, the Orator: Penetrating the Lincoln Legend (Westport, CT: Greenwood Press, 1992), 25, http://www.questia.com/read/27419298.

Cited article

Cox Regression in a Markov Renewal Model: An Application to the Analysis of Bone Marrow Transplant Data
Settings

Settings

Typeface
Text size Smaller Larger Reset View mode
Search within

Search within this article

Look up

Look up a word

  • Dictionary
  • Thesaurus
Please submit a word or phrase above.
Print this page

Print this page

Why can't I print more than one page at a time?

Full screen

matching results for page

Cited passage

Style
Citations are available only to our active members.
Sign up now to cite pages or passages in MLA, APA and Chicago citation styles.

"Portraying himself as an honest, ordinary person helped Lincoln identify with his audiences." (Einhorn, 1992, p. 25).

"Portraying himself as an honest, ordinary person helped Lincoln identify with his audiences." (Einhorn 25)

"Portraying himself as an honest, ordinary person helped Lincoln identify with his audiences."1

1. Lois J. Einhorn, Abraham Lincoln, the Orator: Penetrating the Lincoln Legend (Westport, CT: Greenwood Press, 1992), 25, http://www.questia.com/read/27419298.

Cited passage

Welcome to the new Questia Reader

The Questia Reader has been updated to provide you with an even better online reading experience.  It is now 100% Responsive, which means you can read our books and articles on any sized device you wish.  All of your favorite tools like notes, highlights, and citations are still here, but the way you select text has been updated to be easier to use, especially on touchscreen devices.  Here's how:

1. Click or tap the first word you want to select.
2. Click or tap the last word you want to select.

OK, got it!

Thanks for trying Questia!

Please continue trying out our research tools, but please note, full functionality is available only to our active members.

Your work will be lost once you leave this Web page.

For full access in an ad-free environment, sign up now for a FREE, 1-day trial.

Already a member? Log in now.