Marginal structural models for case-cohort study designs to estimate the association of antiretroviral therapy initiation with incident AIDS or death

Stephen R. Cole, Michael G. Hudgens, Phyllis C. Tien, Kathryn Anastos, Lawrence Kingsley, Joan S. Chmiel, Lisa P. Jacobson

Research output: Contribution to journalArticle

10 Citations (Scopus)

Abstract

To estimate the association of antiretroviral therapy initiation with incident acquired immunodeficiency syndrome (AIDS) or death while accounting for time-varying confounding in a cost-efficient manner, the authors combined a case-cohort study design with inverse probability-weighted estimation of a marginal structural Cox proportional hazards model. A total of 950 adults who were positive for human immunodeficiency virus type 1 were followed in 2 US cohort studies between 1995 and 2007. In the full cohort, 211 AIDS cases or deaths occurred during 4,456 person-years. In an illustrative 20% random subcohort of 190 participants, 41 AIDS cases or deaths occurred during 861 person-years. Accounting for measured confounders and determinants of dropout by inverse probability weighting, the full cohort hazard ratio was 0.41 (95% confidence interval: 0.26, 0.65) and the case-cohort hazard ratio was 0.47 (95% confidence interval: 0.26, 0.83). Standard multivariable-adjusted hazard ratios were closer to the null, regardless of study design. The precision lost with the case-cohort design was modest given the cost savings. Results from Monte Carlo simulations demonstrated that the proposed approach yields approximately unbiased estimates of the hazard ratio with appropriate confidence interval coverage. Marginal structural model analysis of case-cohort study designs provides a cost-efficient design coupled with an accurate analytic method for research settings in which there is time-varying confounding.

Original languageEnglish (US)
Pages (from-to)381-390
Number of pages10
JournalAmerican Journal of Epidemiology
Volume175
Issue number5
DOIs
StatePublished - Mar 1 2012

Fingerprint

Structural Models
Acquired Immunodeficiency Syndrome
Cohort Studies
Confidence Intervals
Costs and Cost Analysis
Cost Savings
Proportional Hazards Models
HIV-1
Therapeutics
Research

Keywords

  • acquired immunodeficiency syndrome
  • case-cohort studies
  • cohort studies
  • confounding bias
  • HIV
  • pharmacoepidemiology
  • selection bias

ASJC Scopus subject areas

  • Epidemiology

Cite this

Marginal structural models for case-cohort study designs to estimate the association of antiretroviral therapy initiation with incident AIDS or death. / Cole, Stephen R.; Hudgens, Michael G.; Tien, Phyllis C.; Anastos, Kathryn; Kingsley, Lawrence; Chmiel, Joan S.; Jacobson, Lisa P.

In: American Journal of Epidemiology, Vol. 175, No. 5, 01.03.2012, p. 381-390.

Research output: Contribution to journalArticle

Cole, Stephen R. ; Hudgens, Michael G. ; Tien, Phyllis C. ; Anastos, Kathryn ; Kingsley, Lawrence ; Chmiel, Joan S. ; Jacobson, Lisa P. / Marginal structural models for case-cohort study designs to estimate the association of antiretroviral therapy initiation with incident AIDS or death. In: American Journal of Epidemiology. 2012 ; Vol. 175, No. 5. pp. 381-390.
@article{c5ec89ad729347758fdb708d8eb974cf,
title = "Marginal structural models for case-cohort study designs to estimate the association of antiretroviral therapy initiation with incident AIDS or death",
abstract = "To estimate the association of antiretroviral therapy initiation with incident acquired immunodeficiency syndrome (AIDS) or death while accounting for time-varying confounding in a cost-efficient manner, the authors combined a case-cohort study design with inverse probability-weighted estimation of a marginal structural Cox proportional hazards model. A total of 950 adults who were positive for human immunodeficiency virus type 1 were followed in 2 US cohort studies between 1995 and 2007. In the full cohort, 211 AIDS cases or deaths occurred during 4,456 person-years. In an illustrative 20{\%} random subcohort of 190 participants, 41 AIDS cases or deaths occurred during 861 person-years. Accounting for measured confounders and determinants of dropout by inverse probability weighting, the full cohort hazard ratio was 0.41 (95{\%} confidence interval: 0.26, 0.65) and the case-cohort hazard ratio was 0.47 (95{\%} confidence interval: 0.26, 0.83). Standard multivariable-adjusted hazard ratios were closer to the null, regardless of study design. The precision lost with the case-cohort design was modest given the cost savings. Results from Monte Carlo simulations demonstrated that the proposed approach yields approximately unbiased estimates of the hazard ratio with appropriate confidence interval coverage. Marginal structural model analysis of case-cohort study designs provides a cost-efficient design coupled with an accurate analytic method for research settings in which there is time-varying confounding.",
keywords = "acquired immunodeficiency syndrome, case-cohort studies, cohort studies, confounding bias, HIV, pharmacoepidemiology, selection bias",
author = "Cole, {Stephen R.} and Hudgens, {Michael G.} and Tien, {Phyllis C.} and Kathryn Anastos and Lawrence Kingsley and Chmiel, {Joan S.} and Jacobson, {Lisa P.}",
year = "2012",
month = "3",
day = "1",
doi = "10.1093/aje/kwr346",
language = "English (US)",
volume = "175",
pages = "381--390",
journal = "American Journal of Epidemiology",
issn = "0002-9262",
publisher = "Oxford University Press",
number = "5",

}

TY - JOUR

T1 - Marginal structural models for case-cohort study designs to estimate the association of antiretroviral therapy initiation with incident AIDS or death

AU - Cole, Stephen R.

AU - Hudgens, Michael G.

AU - Tien, Phyllis C.

AU - Anastos, Kathryn

AU - Kingsley, Lawrence

AU - Chmiel, Joan S.

AU - Jacobson, Lisa P.

PY - 2012/3/1

Y1 - 2012/3/1

N2 - To estimate the association of antiretroviral therapy initiation with incident acquired immunodeficiency syndrome (AIDS) or death while accounting for time-varying confounding in a cost-efficient manner, the authors combined a case-cohort study design with inverse probability-weighted estimation of a marginal structural Cox proportional hazards model. A total of 950 adults who were positive for human immunodeficiency virus type 1 were followed in 2 US cohort studies between 1995 and 2007. In the full cohort, 211 AIDS cases or deaths occurred during 4,456 person-years. In an illustrative 20% random subcohort of 190 participants, 41 AIDS cases or deaths occurred during 861 person-years. Accounting for measured confounders and determinants of dropout by inverse probability weighting, the full cohort hazard ratio was 0.41 (95% confidence interval: 0.26, 0.65) and the case-cohort hazard ratio was 0.47 (95% confidence interval: 0.26, 0.83). Standard multivariable-adjusted hazard ratios were closer to the null, regardless of study design. The precision lost with the case-cohort design was modest given the cost savings. Results from Monte Carlo simulations demonstrated that the proposed approach yields approximately unbiased estimates of the hazard ratio with appropriate confidence interval coverage. Marginal structural model analysis of case-cohort study designs provides a cost-efficient design coupled with an accurate analytic method for research settings in which there is time-varying confounding.

AB - To estimate the association of antiretroviral therapy initiation with incident acquired immunodeficiency syndrome (AIDS) or death while accounting for time-varying confounding in a cost-efficient manner, the authors combined a case-cohort study design with inverse probability-weighted estimation of a marginal structural Cox proportional hazards model. A total of 950 adults who were positive for human immunodeficiency virus type 1 were followed in 2 US cohort studies between 1995 and 2007. In the full cohort, 211 AIDS cases or deaths occurred during 4,456 person-years. In an illustrative 20% random subcohort of 190 participants, 41 AIDS cases or deaths occurred during 861 person-years. Accounting for measured confounders and determinants of dropout by inverse probability weighting, the full cohort hazard ratio was 0.41 (95% confidence interval: 0.26, 0.65) and the case-cohort hazard ratio was 0.47 (95% confidence interval: 0.26, 0.83). Standard multivariable-adjusted hazard ratios were closer to the null, regardless of study design. The precision lost with the case-cohort design was modest given the cost savings. Results from Monte Carlo simulations demonstrated that the proposed approach yields approximately unbiased estimates of the hazard ratio with appropriate confidence interval coverage. Marginal structural model analysis of case-cohort study designs provides a cost-efficient design coupled with an accurate analytic method for research settings in which there is time-varying confounding.

KW - acquired immunodeficiency syndrome

KW - case-cohort studies

KW - cohort studies

KW - confounding bias

KW - HIV

KW - pharmacoepidemiology

KW - selection bias

UR - http://www.scopus.com/inward/record.url?scp=84857566961&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84857566961&partnerID=8YFLogxK

U2 - 10.1093/aje/kwr346

DO - 10.1093/aje/kwr346

M3 - Article

VL - 175

SP - 381

EP - 390

JO - American Journal of Epidemiology

JF - American Journal of Epidemiology

SN - 0002-9262

IS - 5

ER -