Doubly robust testing and estimation of model-adjusted effect-measure modification with complex survey data

Hao W. Zheng, Babette A. Brumback, Xiaomin Lu, Erin D. Bouldin, Michael B. Cannell, Elena Andresen

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Model-based standardization enables adjustment for confounding of a population-averaged exposure effect on an outcome. It requires either a model for the probability of the exposure conditional on the confounders (an exposure model) or a model for the expectation of the outcome conditional on the exposure and the confounders (an outcome model). The methodology can also be applied to estimate averaged exposure effects within categories of an effect modifier and to test whether these effects differ or not. Recently, we extended that methodology for use with complex survey data, to estimate the effects of disability status on cost barriers to health care within three age categories and to test for differences. We applied the methodology to data from the 2007 Florida Behavioral Risk Factor Surveillance System Survey (BRFSS). The exposure modeling and outcome modeling approaches yielded two contrasting sets of results. In the present paper, we develop and apply to the BRFSS example two doubly robust approaches to testing and estimating effect modification with complex survey data; these approaches require that only one of these two models be correctly specified. Furthermore, assuming that at least one of the models is correctly specified, we can use the doubly robust approaches to develop and apply goodness-of-fit tests for the exposure and outcome models. We compare the exposure modeling, outcome modeling, and doubly robust approaches in terms of a simulation study and the BRFSS example.

Original languageEnglish (US)
Pages (from-to)673-684
Number of pages12
JournalStatistics in Medicine
Volume32
Issue number4
DOIs
StatePublished - Feb 20 2013

Fingerprint

Survey Data
Behavioral Risk Factor Surveillance System
Testing
Risk Factors
Surveillance
Modeling
Model
Methodology
Surveys and Questionnaires
Confounding
Disability
Goodness of Fit Test
Delivery of Health Care
Standardization
Costs and Cost Analysis
Estimate
Healthcare
Adjustment
Population
Simulation Study

Keywords

  • Causal inference
  • Complex survey data
  • Confounding
  • Doubly robust
  • Heterogeneity
  • Interaction
  • Model-based standardization

ASJC Scopus subject areas

  • Epidemiology
  • Statistics and Probability

Cite this

Doubly robust testing and estimation of model-adjusted effect-measure modification with complex survey data. / Zheng, Hao W.; Brumback, Babette A.; Lu, Xiaomin; Bouldin, Erin D.; Cannell, Michael B.; Andresen, Elena.

In: Statistics in Medicine, Vol. 32, No. 4, 20.02.2013, p. 673-684.

Research output: Contribution to journalArticle

Zheng, Hao W. ; Brumback, Babette A. ; Lu, Xiaomin ; Bouldin, Erin D. ; Cannell, Michael B. ; Andresen, Elena. / Doubly robust testing and estimation of model-adjusted effect-measure modification with complex survey data. In: Statistics in Medicine. 2013 ; Vol. 32, No. 4. pp. 673-684.
@article{2a3f2d1e1d5846088ea36be07495ff9a,
title = "Doubly robust testing and estimation of model-adjusted effect-measure modification with complex survey data",
abstract = "Model-based standardization enables adjustment for confounding of a population-averaged exposure effect on an outcome. It requires either a model for the probability of the exposure conditional on the confounders (an exposure model) or a model for the expectation of the outcome conditional on the exposure and the confounders (an outcome model). The methodology can also be applied to estimate averaged exposure effects within categories of an effect modifier and to test whether these effects differ or not. Recently, we extended that methodology for use with complex survey data, to estimate the effects of disability status on cost barriers to health care within three age categories and to test for differences. We applied the methodology to data from the 2007 Florida Behavioral Risk Factor Surveillance System Survey (BRFSS). The exposure modeling and outcome modeling approaches yielded two contrasting sets of results. In the present paper, we develop and apply to the BRFSS example two doubly robust approaches to testing and estimating effect modification with complex survey data; these approaches require that only one of these two models be correctly specified. Furthermore, assuming that at least one of the models is correctly specified, we can use the doubly robust approaches to develop and apply goodness-of-fit tests for the exposure and outcome models. We compare the exposure modeling, outcome modeling, and doubly robust approaches in terms of a simulation study and the BRFSS example.",
keywords = "Causal inference, Complex survey data, Confounding, Doubly robust, Heterogeneity, Interaction, Model-based standardization",
author = "Zheng, {Hao W.} and Brumback, {Babette A.} and Xiaomin Lu and Bouldin, {Erin D.} and Cannell, {Michael B.} and Elena Andresen",
year = "2013",
month = "2",
day = "20",
doi = "10.1002/sim.5532",
language = "English (US)",
volume = "32",
pages = "673--684",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "4",

}

TY - JOUR

T1 - Doubly robust testing and estimation of model-adjusted effect-measure modification with complex survey data

AU - Zheng, Hao W.

AU - Brumback, Babette A.

AU - Lu, Xiaomin

AU - Bouldin, Erin D.

AU - Cannell, Michael B.

AU - Andresen, Elena

PY - 2013/2/20

Y1 - 2013/2/20

N2 - Model-based standardization enables adjustment for confounding of a population-averaged exposure effect on an outcome. It requires either a model for the probability of the exposure conditional on the confounders (an exposure model) or a model for the expectation of the outcome conditional on the exposure and the confounders (an outcome model). The methodology can also be applied to estimate averaged exposure effects within categories of an effect modifier and to test whether these effects differ or not. Recently, we extended that methodology for use with complex survey data, to estimate the effects of disability status on cost barriers to health care within three age categories and to test for differences. We applied the methodology to data from the 2007 Florida Behavioral Risk Factor Surveillance System Survey (BRFSS). The exposure modeling and outcome modeling approaches yielded two contrasting sets of results. In the present paper, we develop and apply to the BRFSS example two doubly robust approaches to testing and estimating effect modification with complex survey data; these approaches require that only one of these two models be correctly specified. Furthermore, assuming that at least one of the models is correctly specified, we can use the doubly robust approaches to develop and apply goodness-of-fit tests for the exposure and outcome models. We compare the exposure modeling, outcome modeling, and doubly robust approaches in terms of a simulation study and the BRFSS example.

AB - Model-based standardization enables adjustment for confounding of a population-averaged exposure effect on an outcome. It requires either a model for the probability of the exposure conditional on the confounders (an exposure model) or a model for the expectation of the outcome conditional on the exposure and the confounders (an outcome model). The methodology can also be applied to estimate averaged exposure effects within categories of an effect modifier and to test whether these effects differ or not. Recently, we extended that methodology for use with complex survey data, to estimate the effects of disability status on cost barriers to health care within three age categories and to test for differences. We applied the methodology to data from the 2007 Florida Behavioral Risk Factor Surveillance System Survey (BRFSS). The exposure modeling and outcome modeling approaches yielded two contrasting sets of results. In the present paper, we develop and apply to the BRFSS example two doubly robust approaches to testing and estimating effect modification with complex survey data; these approaches require that only one of these two models be correctly specified. Furthermore, assuming that at least one of the models is correctly specified, we can use the doubly robust approaches to develop and apply goodness-of-fit tests for the exposure and outcome models. We compare the exposure modeling, outcome modeling, and doubly robust approaches in terms of a simulation study and the BRFSS example.

KW - Causal inference

KW - Complex survey data

KW - Confounding

KW - Doubly robust

KW - Heterogeneity

KW - Interaction

KW - Model-based standardization

UR - http://www.scopus.com/inward/record.url?scp=84872802276&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84872802276&partnerID=8YFLogxK

U2 - 10.1002/sim.5532

DO - 10.1002/sim.5532

M3 - Article

C2 - 22833449

AN - SCOPUS:84872802276

VL - 32

SP - 673

EP - 684

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 4

ER -