Handling time-varying confounding in state transition models for dynamic optimization of adaptive interdisciplinary pain management

Aera LeBoulluec, Nilabh Ohol, Victoria Chen, Li Zeng, Jay Rosenberger, Robert Joseph Gatchel

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Interdisciplinary pain management combines multiple disciplines of professionals to understand the biological and psychosocial factors causing a patient's pain and to determine the best treatments among many to administer. To improve current and future pain outcomes, the developed adaptive interdisciplinary pain management framework employs approximate dynamic programming with state transition and outcome models estimated from actual patient data. The sequential treatment structure and observational nature of the data lead to a form of endogeneity, which results in biased model parameter estimates when developing state transition and outcome models. This research develops a process based on the inverse probability of treatment weighted method to address the endogeneity in estimating state transition and outcome models. This article discusses a general method developed for independent treatments. The proposed approach can potentially be employed for adaptive treatment in other sequential health care applications based on observational data. Our approach is demonstrated using data from the Eugene McDermott Center for Pain Management at the University of Texas Southwestern Medical Center at Dallas.

Original languageEnglish
Pages (from-to)83-92
Number of pages10
JournalIISE Transactions on Healthcare Systems Engineering
Volume8
Issue number1
DOIs
StatePublished - 2 Jan 2018

Fingerprint

Pain Management
pain
management
Therapeutics
Pain
biological factors
Health care
Dynamic programming
psychosocial factors
Biological Factors
programming
Handling (Psychology)
time
health care
Psychology
Delivery of Health Care
Research

Keywords

  • Dynamic optimization
  • endogeneity
  • linear regression
  • outpatient clinics
  • time-varying confounding

Cite this

LeBoulluec, Aera ; Ohol, Nilabh ; Chen, Victoria ; Zeng, Li ; Rosenberger, Jay ; Gatchel, Robert Joseph. / Handling time-varying confounding in state transition models for dynamic optimization of adaptive interdisciplinary pain management. In: IISE Transactions on Healthcare Systems Engineering. 2018 ; Vol. 8, No. 1. pp. 83-92.
@article{70cd7208fd1d42b68de66a0ad60c4757,
title = "Handling time-varying confounding in state transition models for dynamic optimization of adaptive interdisciplinary pain management",
abstract = "Interdisciplinary pain management combines multiple disciplines of professionals to understand the biological and psychosocial factors causing a patient's pain and to determine the best treatments among many to administer. To improve current and future pain outcomes, the developed adaptive interdisciplinary pain management framework employs approximate dynamic programming with state transition and outcome models estimated from actual patient data. The sequential treatment structure and observational nature of the data lead to a form of endogeneity, which results in biased model parameter estimates when developing state transition and outcome models. This research develops a process based on the inverse probability of treatment weighted method to address the endogeneity in estimating state transition and outcome models. This article discusses a general method developed for independent treatments. The proposed approach can potentially be employed for adaptive treatment in other sequential health care applications based on observational data. Our approach is demonstrated using data from the Eugene McDermott Center for Pain Management at the University of Texas Southwestern Medical Center at Dallas.",
keywords = "Dynamic optimization, endogeneity, linear regression, outpatient clinics, time-varying confounding",
author = "Aera LeBoulluec and Nilabh Ohol and Victoria Chen and Li Zeng and Jay Rosenberger and Gatchel, {Robert Joseph}",
year = "2018",
month = "1",
day = "2",
doi = "10.1080/24725579.2017.1418770",
language = "English",
volume = "8",
pages = "83--92",
journal = "IISE Transactions on Healthcare Systems Engineering",
issn = "2472-5579",
publisher = "Taylor and Francis Ltd.",
number = "1",

}

Handling time-varying confounding in state transition models for dynamic optimization of adaptive interdisciplinary pain management. / LeBoulluec, Aera; Ohol, Nilabh; Chen, Victoria; Zeng, Li; Rosenberger, Jay; Gatchel, Robert Joseph.

In: IISE Transactions on Healthcare Systems Engineering, Vol. 8, No. 1, 02.01.2018, p. 83-92.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Handling time-varying confounding in state transition models for dynamic optimization of adaptive interdisciplinary pain management

AU - LeBoulluec, Aera

AU - Ohol, Nilabh

AU - Chen, Victoria

AU - Zeng, Li

AU - Rosenberger, Jay

AU - Gatchel, Robert Joseph

PY - 2018/1/2

Y1 - 2018/1/2

N2 - Interdisciplinary pain management combines multiple disciplines of professionals to understand the biological and psychosocial factors causing a patient's pain and to determine the best treatments among many to administer. To improve current and future pain outcomes, the developed adaptive interdisciplinary pain management framework employs approximate dynamic programming with state transition and outcome models estimated from actual patient data. The sequential treatment structure and observational nature of the data lead to a form of endogeneity, which results in biased model parameter estimates when developing state transition and outcome models. This research develops a process based on the inverse probability of treatment weighted method to address the endogeneity in estimating state transition and outcome models. This article discusses a general method developed for independent treatments. The proposed approach can potentially be employed for adaptive treatment in other sequential health care applications based on observational data. Our approach is demonstrated using data from the Eugene McDermott Center for Pain Management at the University of Texas Southwestern Medical Center at Dallas.

AB - Interdisciplinary pain management combines multiple disciplines of professionals to understand the biological and psychosocial factors causing a patient's pain and to determine the best treatments among many to administer. To improve current and future pain outcomes, the developed adaptive interdisciplinary pain management framework employs approximate dynamic programming with state transition and outcome models estimated from actual patient data. The sequential treatment structure and observational nature of the data lead to a form of endogeneity, which results in biased model parameter estimates when developing state transition and outcome models. This research develops a process based on the inverse probability of treatment weighted method to address the endogeneity in estimating state transition and outcome models. This article discusses a general method developed for independent treatments. The proposed approach can potentially be employed for adaptive treatment in other sequential health care applications based on observational data. Our approach is demonstrated using data from the Eugene McDermott Center for Pain Management at the University of Texas Southwestern Medical Center at Dallas.

KW - Dynamic optimization

KW - endogeneity

KW - linear regression

KW - outpatient clinics

KW - time-varying confounding

UR - http://www.scopus.com/inward/record.url?scp=85041136117&partnerID=8YFLogxK

U2 - 10.1080/24725579.2017.1418770

DO - 10.1080/24725579.2017.1418770

M3 - Article

AN - SCOPUS:85041136117

VL - 8

SP - 83

EP - 92

JO - IISE Transactions on Healthcare Systems Engineering

JF - IISE Transactions on Healthcare Systems Engineering

SN - 2472-5579

IS - 1

ER -