Assessing the sensitivity of decision-analytic results to unobserved markers of risk: Defining the effects of heterogeneity bias

Karen M. Kuntz, Sue J. Goldie

Research output: Contribution to journalArticle

28 Citations (Scopus)

Abstract

An important assumption made when constructing a Markov model is that all persons residing in a health state are identical. Failure to adjust for population heterogeneity caused by unobserved variables can therefore cause bias in model results. The authors used a simple model to evaluate the potential impact of heterogeneity bias, defined as the percentage change in the life expectancy gain with an intervention predicted by a model that does not adjust for heterogeneity (unadjusted model) compared to one that does (adjusted model). The life expectancy gains were consistently greater in the unadjusted model compared to the adjusted model (positive bias). For an annual probability of developing disease of 1%, the heterogeneity bias exceeded 50% when the relative risk of disease with the heterogeneity factor versus without the factor was greater than 15 and the prevalence of the heterogeneity factor was between 5% and 25%. When constructing decision-analytic models, analysts need to be cognizant of unobserved factors that introduce heterogeneity into the cohart. This analysis provides a general framework to determine when issues of heterogeneity may be important.

Original languageEnglish (US)
Pages (from-to)218-227
Number of pages10
JournalMedical Decision Making
Volume22
Issue number3
DOIs
StatePublished - May 30 2002

Fingerprint

Life Expectancy
Population Characteristics
Health

Keywords

  • Decision analysis
  • Heterogeneity
  • Markov model

Cite this

Assessing the sensitivity of decision-analytic results to unobserved markers of risk : Defining the effects of heterogeneity bias. / Kuntz, Karen M.; Goldie, Sue J.

In: Medical Decision Making, Vol. 22, No. 3, 30.05.2002, p. 218-227.

Research output: Contribution to journalArticle

@article{cca6f438a5c94ce99b4cae7f155711ae,
title = "Assessing the sensitivity of decision-analytic results to unobserved markers of risk: Defining the effects of heterogeneity bias",
abstract = "An important assumption made when constructing a Markov model is that all persons residing in a health state are identical. Failure to adjust for population heterogeneity caused by unobserved variables can therefore cause bias in model results. The authors used a simple model to evaluate the potential impact of heterogeneity bias, defined as the percentage change in the life expectancy gain with an intervention predicted by a model that does not adjust for heterogeneity (unadjusted model) compared to one that does (adjusted model). The life expectancy gains were consistently greater in the unadjusted model compared to the adjusted model (positive bias). For an annual probability of developing disease of 1{\%}, the heterogeneity bias exceeded 50{\%} when the relative risk of disease with the heterogeneity factor versus without the factor was greater than 15 and the prevalence of the heterogeneity factor was between 5{\%} and 25{\%}. When constructing decision-analytic models, analysts need to be cognizant of unobserved factors that introduce heterogeneity into the cohart. This analysis provides a general framework to determine when issues of heterogeneity may be important.",
keywords = "Decision analysis, Heterogeneity, Markov model",
author = "Kuntz, {Karen M.} and Goldie, {Sue J.}",
year = "2002",
month = "5",
day = "30",
doi = "10.1177/02789X02022003004",
language = "English (US)",
volume = "22",
pages = "218--227",
journal = "Medical Decision Making",
issn = "0272-989X",
publisher = "SAGE Publications Inc.",
number = "3",

}

TY - JOUR

T1 - Assessing the sensitivity of decision-analytic results to unobserved markers of risk

T2 - Defining the effects of heterogeneity bias

AU - Kuntz, Karen M.

AU - Goldie, Sue J.

PY - 2002/5/30

Y1 - 2002/5/30

N2 - An important assumption made when constructing a Markov model is that all persons residing in a health state are identical. Failure to adjust for population heterogeneity caused by unobserved variables can therefore cause bias in model results. The authors used a simple model to evaluate the potential impact of heterogeneity bias, defined as the percentage change in the life expectancy gain with an intervention predicted by a model that does not adjust for heterogeneity (unadjusted model) compared to one that does (adjusted model). The life expectancy gains were consistently greater in the unadjusted model compared to the adjusted model (positive bias). For an annual probability of developing disease of 1%, the heterogeneity bias exceeded 50% when the relative risk of disease with the heterogeneity factor versus without the factor was greater than 15 and the prevalence of the heterogeneity factor was between 5% and 25%. When constructing decision-analytic models, analysts need to be cognizant of unobserved factors that introduce heterogeneity into the cohart. This analysis provides a general framework to determine when issues of heterogeneity may be important.

AB - An important assumption made when constructing a Markov model is that all persons residing in a health state are identical. Failure to adjust for population heterogeneity caused by unobserved variables can therefore cause bias in model results. The authors used a simple model to evaluate the potential impact of heterogeneity bias, defined as the percentage change in the life expectancy gain with an intervention predicted by a model that does not adjust for heterogeneity (unadjusted model) compared to one that does (adjusted model). The life expectancy gains were consistently greater in the unadjusted model compared to the adjusted model (positive bias). For an annual probability of developing disease of 1%, the heterogeneity bias exceeded 50% when the relative risk of disease with the heterogeneity factor versus without the factor was greater than 15 and the prevalence of the heterogeneity factor was between 5% and 25%. When constructing decision-analytic models, analysts need to be cognizant of unobserved factors that introduce heterogeneity into the cohart. This analysis provides a general framework to determine when issues of heterogeneity may be important.

KW - Decision analysis

KW - Heterogeneity

KW - Markov model

UR - http://www.scopus.com/inward/record.url?scp=0035999536&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0035999536&partnerID=8YFLogxK

U2 - 10.1177/02789X02022003004

DO - 10.1177/02789X02022003004

M3 - Article

C2 - 12058779

AN - SCOPUS:0035999536

VL - 22

SP - 218

EP - 227

JO - Medical Decision Making

JF - Medical Decision Making

SN - 0272-989X

IS - 3

ER -