Analysis of interval-censored disease progression data via multi-state models under a nonignorable inspection process

Baojiang Chen, Grace Y. Yi, Richard J. Cook

Research output: Contribution to journalArticle

20 Citations (Scopus)

Abstract

Irreversible multi-state models provide a convenient framework for characterizing disease processes that arise when the states represent the degree of organ or tissue damage incurred by a progressive disease. In many settings, however, individuals are only observed at periodic clinic visits and so the precise times of the transitions are not observed. If the life history and observation processes are not independent, the observation process contains information about the life history process, and more importantly, likelihoods based on the disease process alone are invalid. With interval-censored failure time data, joint models are nonidentifiable and data analysts must rely on sensitivity analyses to assess the effect of the dependent observation times. This paper is concerned, however, with the analysis of data from progressive multi-state disease processes in which individuals are scheduled to be seen at periodic pre-scheduled assessment times. We cast the problem in the framework used for incomplete longitudinal data problems. Maximum likelihood estimation via an EM algorithm is advocated for parameter estimation. Simulation studies demonstrate that the proposed method works well under a variety of situations. Data from a cohort of patients with psoriatic arthritis are analyzed for illustration.

Original languageEnglish (US)
Pages (from-to)1175-1189
Number of pages15
JournalStatistics in Medicine
Volume29
Issue number11
DOIs
StatePublished - May 20 2010

Fingerprint

Multi-state Model
Progression
Data Model
Disease Progression
Inspection
Interval
Observation
Psoriatic Arthritis
Ambulatory Care
Dependent Observations
Interval-censored Data
Arthritis
Failure Time Data
Joint Model
Multi-state
Joints
Incomplete Data
Longitudinal Data
EM Algorithm
Maximum Likelihood Estimation

Keywords

  • EM algorithm
  • Longitudinal data
  • Maximum likelihood
  • Progressive models

ASJC Scopus subject areas

  • Epidemiology
  • Statistics and Probability

Cite this

Analysis of interval-censored disease progression data via multi-state models under a nonignorable inspection process. / Chen, Baojiang; Yi, Grace Y.; Cook, Richard J.

In: Statistics in Medicine, Vol. 29, No. 11, 20.05.2010, p. 1175-1189.

Research output: Contribution to journalArticle

Chen, Baojiang ; Yi, Grace Y. ; Cook, Richard J. / Analysis of interval-censored disease progression data via multi-state models under a nonignorable inspection process. In: Statistics in Medicine. 2010 ; Vol. 29, No. 11. pp. 1175-1189.
@article{5890d2aeaf534bd0873b166d81456f31,
title = "Analysis of interval-censored disease progression data via multi-state models under a nonignorable inspection process",
abstract = "Irreversible multi-state models provide a convenient framework for characterizing disease processes that arise when the states represent the degree of organ or tissue damage incurred by a progressive disease. In many settings, however, individuals are only observed at periodic clinic visits and so the precise times of the transitions are not observed. If the life history and observation processes are not independent, the observation process contains information about the life history process, and more importantly, likelihoods based on the disease process alone are invalid. With interval-censored failure time data, joint models are nonidentifiable and data analysts must rely on sensitivity analyses to assess the effect of the dependent observation times. This paper is concerned, however, with the analysis of data from progressive multi-state disease processes in which individuals are scheduled to be seen at periodic pre-scheduled assessment times. We cast the problem in the framework used for incomplete longitudinal data problems. Maximum likelihood estimation via an EM algorithm is advocated for parameter estimation. Simulation studies demonstrate that the proposed method works well under a variety of situations. Data from a cohort of patients with psoriatic arthritis are analyzed for illustration.",
keywords = "EM algorithm, Longitudinal data, Maximum likelihood, Progressive models",
author = "Baojiang Chen and Yi, {Grace Y.} and Cook, {Richard J.}",
year = "2010",
month = "5",
day = "20",
doi = "10.1002/sim.3804",
language = "English (US)",
volume = "29",
pages = "1175--1189",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "11",

}

TY - JOUR

T1 - Analysis of interval-censored disease progression data via multi-state models under a nonignorable inspection process

AU - Chen, Baojiang

AU - Yi, Grace Y.

AU - Cook, Richard J.

PY - 2010/5/20

Y1 - 2010/5/20

N2 - Irreversible multi-state models provide a convenient framework for characterizing disease processes that arise when the states represent the degree of organ or tissue damage incurred by a progressive disease. In many settings, however, individuals are only observed at periodic clinic visits and so the precise times of the transitions are not observed. If the life history and observation processes are not independent, the observation process contains information about the life history process, and more importantly, likelihoods based on the disease process alone are invalid. With interval-censored failure time data, joint models are nonidentifiable and data analysts must rely on sensitivity analyses to assess the effect of the dependent observation times. This paper is concerned, however, with the analysis of data from progressive multi-state disease processes in which individuals are scheduled to be seen at periodic pre-scheduled assessment times. We cast the problem in the framework used for incomplete longitudinal data problems. Maximum likelihood estimation via an EM algorithm is advocated for parameter estimation. Simulation studies demonstrate that the proposed method works well under a variety of situations. Data from a cohort of patients with psoriatic arthritis are analyzed for illustration.

AB - Irreversible multi-state models provide a convenient framework for characterizing disease processes that arise when the states represent the degree of organ or tissue damage incurred by a progressive disease. In many settings, however, individuals are only observed at periodic clinic visits and so the precise times of the transitions are not observed. If the life history and observation processes are not independent, the observation process contains information about the life history process, and more importantly, likelihoods based on the disease process alone are invalid. With interval-censored failure time data, joint models are nonidentifiable and data analysts must rely on sensitivity analyses to assess the effect of the dependent observation times. This paper is concerned, however, with the analysis of data from progressive multi-state disease processes in which individuals are scheduled to be seen at periodic pre-scheduled assessment times. We cast the problem in the framework used for incomplete longitudinal data problems. Maximum likelihood estimation via an EM algorithm is advocated for parameter estimation. Simulation studies demonstrate that the proposed method works well under a variety of situations. Data from a cohort of patients with psoriatic arthritis are analyzed for illustration.

KW - EM algorithm

KW - Longitudinal data

KW - Maximum likelihood

KW - Progressive models

UR - http://www.scopus.com/inward/record.url?scp=77951612162&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77951612162&partnerID=8YFLogxK

U2 - 10.1002/sim.3804

DO - 10.1002/sim.3804

M3 - Article

C2 - 20437455

AN - SCOPUS:77951612162

VL - 29

SP - 1175

EP - 1189

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 11

ER -