Weighted hurdle regression method for joint modeling of cardiovascular events likelihood and rate in the US dialysis population

Damla Şentürk, Lorien Dalrymple, Yi Mu, Danh V. Nguyen

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

We propose a new weighted hurdle regression method for modeling count data, with particular interest in modeling cardiovascular events in patients on dialysis. Cardiovascular disease remains one of the leading causes of hospitalization and death in this population. Our aim is to jointly model the relationship/association between covariates and (i) the probability of cardiovascular events, a binary process, and (ii) the rate of events once the realization is positive-when the 'hurdle' is crossed-using a zero-truncated Poisson distribution. When the observation period or follow-up time, from the start of dialysis, varies among individuals, the estimated probability of positive cardiovascular events during the study period will be biased. Furthermore, when the model contains covariates, then the estimated relationship between the covariates and the probability of cardiovascular events will also be biased. These challenges are addressed with the proposed weighted hurdle regression method. Estimation for the weighted hurdle regression model is a weighted likelihood approach, where standard maximum likelihood estimation can be utilized. The method is illustrated with data from the United States Renal Data System. Simulation studies show the ability of proposed method to successfully adjust for differential follow-up times and incorporate the effects of covariates in the weighting.

Original languageEnglish (US)
Pages (from-to)4387-4401
Number of pages15
JournalStatistics in Medicine
Volume33
Issue number25
DOIs
StatePublished - 2014

Fingerprint

Joint Modeling
Dialysis
Likelihood
Joints
Regression
Covariates
Population
Biased
Poisson Distribution
Weighted Likelihood
Truncated Distributions
Information Systems
Count Data
Poisson distribution
Cause of Death
Hospitalization
Cardiovascular Diseases
Modeling
Maximum Likelihood Estimation
Observation

Keywords

  • Cardiovascular outcomes
  • Dialysis
  • End-stage renal disease
  • Hurdle model
  • Infection
  • Poisson regression
  • United States renal data system

ASJC Scopus subject areas

  • Epidemiology
  • Statistics and Probability

Cite this

Weighted hurdle regression method for joint modeling of cardiovascular events likelihood and rate in the US dialysis population. / Şentürk, Damla; Dalrymple, Lorien; Mu, Yi; Nguyen, Danh V.

In: Statistics in Medicine, Vol. 33, No. 25, 2014, p. 4387-4401.

Research output: Contribution to journalArticle

Şentürk, Damla ; Dalrymple, Lorien ; Mu, Yi ; Nguyen, Danh V. / Weighted hurdle regression method for joint modeling of cardiovascular events likelihood and rate in the US dialysis population. In: Statistics in Medicine. 2014 ; Vol. 33, No. 25. pp. 4387-4401.
@article{0f7ea72952bf46d887210f7044bafdcf,
title = "Weighted hurdle regression method for joint modeling of cardiovascular events likelihood and rate in the US dialysis population",
abstract = "We propose a new weighted hurdle regression method for modeling count data, with particular interest in modeling cardiovascular events in patients on dialysis. Cardiovascular disease remains one of the leading causes of hospitalization and death in this population. Our aim is to jointly model the relationship/association between covariates and (i) the probability of cardiovascular events, a binary process, and (ii) the rate of events once the realization is positive-when the 'hurdle' is crossed-using a zero-truncated Poisson distribution. When the observation period or follow-up time, from the start of dialysis, varies among individuals, the estimated probability of positive cardiovascular events during the study period will be biased. Furthermore, when the model contains covariates, then the estimated relationship between the covariates and the probability of cardiovascular events will also be biased. These challenges are addressed with the proposed weighted hurdle regression method. Estimation for the weighted hurdle regression model is a weighted likelihood approach, where standard maximum likelihood estimation can be utilized. The method is illustrated with data from the United States Renal Data System. Simulation studies show the ability of proposed method to successfully adjust for differential follow-up times and incorporate the effects of covariates in the weighting.",
keywords = "Cardiovascular outcomes, Dialysis, End-stage renal disease, Hurdle model, Infection, Poisson regression, United States renal data system",
author = "Damla Şent{\"u}rk and Lorien Dalrymple and Yi Mu and Nguyen, {Danh V.}",
year = "2014",
doi = "10.1002/sim.6232",
language = "English (US)",
volume = "33",
pages = "4387--4401",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "25",

}

TY - JOUR

T1 - Weighted hurdle regression method for joint modeling of cardiovascular events likelihood and rate in the US dialysis population

AU - Şentürk, Damla

AU - Dalrymple, Lorien

AU - Mu, Yi

AU - Nguyen, Danh V.

PY - 2014

Y1 - 2014

N2 - We propose a new weighted hurdle regression method for modeling count data, with particular interest in modeling cardiovascular events in patients on dialysis. Cardiovascular disease remains one of the leading causes of hospitalization and death in this population. Our aim is to jointly model the relationship/association between covariates and (i) the probability of cardiovascular events, a binary process, and (ii) the rate of events once the realization is positive-when the 'hurdle' is crossed-using a zero-truncated Poisson distribution. When the observation period or follow-up time, from the start of dialysis, varies among individuals, the estimated probability of positive cardiovascular events during the study period will be biased. Furthermore, when the model contains covariates, then the estimated relationship between the covariates and the probability of cardiovascular events will also be biased. These challenges are addressed with the proposed weighted hurdle regression method. Estimation for the weighted hurdle regression model is a weighted likelihood approach, where standard maximum likelihood estimation can be utilized. The method is illustrated with data from the United States Renal Data System. Simulation studies show the ability of proposed method to successfully adjust for differential follow-up times and incorporate the effects of covariates in the weighting.

AB - We propose a new weighted hurdle regression method for modeling count data, with particular interest in modeling cardiovascular events in patients on dialysis. Cardiovascular disease remains one of the leading causes of hospitalization and death in this population. Our aim is to jointly model the relationship/association between covariates and (i) the probability of cardiovascular events, a binary process, and (ii) the rate of events once the realization is positive-when the 'hurdle' is crossed-using a zero-truncated Poisson distribution. When the observation period or follow-up time, from the start of dialysis, varies among individuals, the estimated probability of positive cardiovascular events during the study period will be biased. Furthermore, when the model contains covariates, then the estimated relationship between the covariates and the probability of cardiovascular events will also be biased. These challenges are addressed with the proposed weighted hurdle regression method. Estimation for the weighted hurdle regression model is a weighted likelihood approach, where standard maximum likelihood estimation can be utilized. The method is illustrated with data from the United States Renal Data System. Simulation studies show the ability of proposed method to successfully adjust for differential follow-up times and incorporate the effects of covariates in the weighting.

KW - Cardiovascular outcomes

KW - Dialysis

KW - End-stage renal disease

KW - Hurdle model

KW - Infection

KW - Poisson regression

KW - United States renal data system

UR - http://www.scopus.com/inward/record.url?scp=84927581178&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84927581178&partnerID=8YFLogxK

U2 - 10.1002/sim.6232

DO - 10.1002/sim.6232

M3 - Article

VL - 33

SP - 4387

EP - 4401

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 25

ER -