Frequency of Recurrent Events at Failure Time: Modeling and Inference

Yijian Huang, Mei Cheng Wang

Research output: Contribution to journalArticle

Abstract

Recurrent events arise in many longitudinal medical studies where time to a terminal event or failure is the primary endpoint. With incomplete follow-up data, the analysis of recurrent events is a challenge owing to their association with the failure, One specific quantity of interest rarely addressed in the statistical literature is the recurrence frequency at the failure time: an example is hospitalization frequency, which is often used as a rough measure of lifetime medical cost. In this article we show that a marginal model (e.g., the log-linear model) of the recurrence frequency, although desirable, is typically not identifiable. For this reason, we advocate modeling the recurrent events and the failure time jointly, and propose an approach to forming semiparametric joint models from prespecified marginal ones. We suggest two conceptually simple and nested regression models aiming at the recurrence frequency as a mark of the failure and at the process of recurrent events. We formulate monotone estimating functions and propose novel interval-estimation procedures to accommodate nonsmooth estimating functions. The resulting estimators are consistent and asymptotically normal. Simulation studies and the application to an AIDS clinical trial exhibit that these proposals are easy to implement and reliable for practical use. Finally, we generalize our proposals to marked recurrent events, and also devise a global inference procedure for recurrent events of multiple types.

Original languageEnglish (US)
Pages (from-to)663-670
Number of pages8
JournalJournal of the American Statistical Association
Volume98
Issue number463
DOIs
StatePublished - Sep 2003

Fingerprint

Recurrent Events
Failure Time
Modeling
Recurrence
Estimating Function
Nested Models
Marginal Model
Nonsmooth Function
Joint Model
Log-linear Models
Interval Estimation
Monotone Function
Semiparametric Model
Clinical Trials
Rough
Inference
Lifetime
Regression Model
Simulation Study
Estimator

Keywords

  • Accelerated failure time model
  • Censored data
  • Count data
  • Log-linear model
  • Log-rank statistic
  • Marked point process
  • Monotone estimating function
  • Multivariate failure time
  • Nonsmooth estimating function
  • Proportional rates model
  • Semi-parametric inference

ASJC Scopus subject areas

  • Mathematics(all)
  • Statistics and Probability

Cite this

Frequency of Recurrent Events at Failure Time : Modeling and Inference. / Huang, Yijian; Wang, Mei Cheng.

In: Journal of the American Statistical Association, Vol. 98, No. 463, 09.2003, p. 663-670.

Research output: Contribution to journalArticle

@article{460a36c04fd941f3a7656986eb165199,
title = "Frequency of Recurrent Events at Failure Time: Modeling and Inference",
abstract = "Recurrent events arise in many longitudinal medical studies where time to a terminal event or failure is the primary endpoint. With incomplete follow-up data, the analysis of recurrent events is a challenge owing to their association with the failure, One specific quantity of interest rarely addressed in the statistical literature is the recurrence frequency at the failure time: an example is hospitalization frequency, which is often used as a rough measure of lifetime medical cost. In this article we show that a marginal model (e.g., the log-linear model) of the recurrence frequency, although desirable, is typically not identifiable. For this reason, we advocate modeling the recurrent events and the failure time jointly, and propose an approach to forming semiparametric joint models from prespecified marginal ones. We suggest two conceptually simple and nested regression models aiming at the recurrence frequency as a mark of the failure and at the process of recurrent events. We formulate monotone estimating functions and propose novel interval-estimation procedures to accommodate nonsmooth estimating functions. The resulting estimators are consistent and asymptotically normal. Simulation studies and the application to an AIDS clinical trial exhibit that these proposals are easy to implement and reliable for practical use. Finally, we generalize our proposals to marked recurrent events, and also devise a global inference procedure for recurrent events of multiple types.",
keywords = "Accelerated failure time model, Censored data, Count data, Log-linear model, Log-rank statistic, Marked point process, Monotone estimating function, Multivariate failure time, Nonsmooth estimating function, Proportional rates model, Semi-parametric inference",
author = "Yijian Huang and Wang, {Mei Cheng}",
year = "2003",
month = "9",
doi = "10.1198/016214503000000567",
language = "English (US)",
volume = "98",
pages = "663--670",
journal = "Journal of the American Statistical Association",
issn = "0162-1459",
publisher = "Taylor and Francis Ltd.",
number = "463",

}

TY - JOUR

T1 - Frequency of Recurrent Events at Failure Time

T2 - Modeling and Inference

AU - Huang, Yijian

AU - Wang, Mei Cheng

PY - 2003/9

Y1 - 2003/9

N2 - Recurrent events arise in many longitudinal medical studies where time to a terminal event or failure is the primary endpoint. With incomplete follow-up data, the analysis of recurrent events is a challenge owing to their association with the failure, One specific quantity of interest rarely addressed in the statistical literature is the recurrence frequency at the failure time: an example is hospitalization frequency, which is often used as a rough measure of lifetime medical cost. In this article we show that a marginal model (e.g., the log-linear model) of the recurrence frequency, although desirable, is typically not identifiable. For this reason, we advocate modeling the recurrent events and the failure time jointly, and propose an approach to forming semiparametric joint models from prespecified marginal ones. We suggest two conceptually simple and nested regression models aiming at the recurrence frequency as a mark of the failure and at the process of recurrent events. We formulate monotone estimating functions and propose novel interval-estimation procedures to accommodate nonsmooth estimating functions. The resulting estimators are consistent and asymptotically normal. Simulation studies and the application to an AIDS clinical trial exhibit that these proposals are easy to implement and reliable for practical use. Finally, we generalize our proposals to marked recurrent events, and also devise a global inference procedure for recurrent events of multiple types.

AB - Recurrent events arise in many longitudinal medical studies where time to a terminal event or failure is the primary endpoint. With incomplete follow-up data, the analysis of recurrent events is a challenge owing to their association with the failure, One specific quantity of interest rarely addressed in the statistical literature is the recurrence frequency at the failure time: an example is hospitalization frequency, which is often used as a rough measure of lifetime medical cost. In this article we show that a marginal model (e.g., the log-linear model) of the recurrence frequency, although desirable, is typically not identifiable. For this reason, we advocate modeling the recurrent events and the failure time jointly, and propose an approach to forming semiparametric joint models from prespecified marginal ones. We suggest two conceptually simple and nested regression models aiming at the recurrence frequency as a mark of the failure and at the process of recurrent events. We formulate monotone estimating functions and propose novel interval-estimation procedures to accommodate nonsmooth estimating functions. The resulting estimators are consistent and asymptotically normal. Simulation studies and the application to an AIDS clinical trial exhibit that these proposals are easy to implement and reliable for practical use. Finally, we generalize our proposals to marked recurrent events, and also devise a global inference procedure for recurrent events of multiple types.

KW - Accelerated failure time model

KW - Censored data

KW - Count data

KW - Log-linear model

KW - Log-rank statistic

KW - Marked point process

KW - Monotone estimating function

KW - Multivariate failure time

KW - Nonsmooth estimating function

KW - Proportional rates model

KW - Semi-parametric inference

UR - http://www.scopus.com/inward/record.url?scp=0242510868&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0242510868&partnerID=8YFLogxK

U2 - 10.1198/016214503000000567

DO - 10.1198/016214503000000567

M3 - Article

AN - SCOPUS:0242510868

VL - 98

SP - 663

EP - 670

JO - Journal of the American Statistical Association

JF - Journal of the American Statistical Association

SN - 0162-1459

IS - 463

ER -