Near/far matching: A study design approach to instrumental variables

Mike Baiocchi, Dylan S. Small, Lin Yang, Daniel E. Polsky, Peter W. Groeneveld

Research output: Contribution to journalArticle

Abstract

Classic instrumental variable techniques involve the use of structural equation modeling or other forms of parameterized modeling. In this paper we use a nonparametric, matching-based instrumental variable methodology that is based on a study design approach. Similar to propensity score matching, though unlike classic instrumental variable approaches, near/far matching is capable of estimating causal effects when the outcome is not continuous. Unlike propensity score matching, though similar to instrumental variable techniques, near/far matching is also capable of estimating causal effects even when unmeasured covariates produce selection bias. We illustrate near/far matching by using Medicare data to compare the effectiveness of carotid arterial stents with cerebral protection versus carotid endarterectomy for the treatment of carotid stenosis.

Original languageEnglish (US)
Pages (from-to)237-253
Number of pages17
JournalHealth Services and Outcomes Research Methodology
Volume12
Issue number4
DOIs
StatePublished - Dec 1 2012
Externally publishedYes

Fingerprint

Propensity Score
Carotid Endarterectomy
Selection Bias
Carotid Stenosis
Medicare
Stents

Keywords

  • Binary outcomes
  • Comparative effectiveness
  • Instrumental variables
  • Matching
  • Medicare data
  • Study design

ASJC Scopus subject areas

  • Health Policy
  • Public Health, Environmental and Occupational Health

Cite this

Near/far matching : A study design approach to instrumental variables. / Baiocchi, Mike; Small, Dylan S.; Yang, Lin; Polsky, Daniel E.; Groeneveld, Peter W.

In: Health Services and Outcomes Research Methodology, Vol. 12, No. 4, 01.12.2012, p. 237-253.

Research output: Contribution to journalArticle

Baiocchi, Mike ; Small, Dylan S. ; Yang, Lin ; Polsky, Daniel E. ; Groeneveld, Peter W. / Near/far matching : A study design approach to instrumental variables. In: Health Services and Outcomes Research Methodology. 2012 ; Vol. 12, No. 4. pp. 237-253.
@article{083da5dd346f4efba44f9293835879f2,
title = "Near/far matching: A study design approach to instrumental variables",
abstract = "Classic instrumental variable techniques involve the use of structural equation modeling or other forms of parameterized modeling. In this paper we use a nonparametric, matching-based instrumental variable methodology that is based on a study design approach. Similar to propensity score matching, though unlike classic instrumental variable approaches, near/far matching is capable of estimating causal effects when the outcome is not continuous. Unlike propensity score matching, though similar to instrumental variable techniques, near/far matching is also capable of estimating causal effects even when unmeasured covariates produce selection bias. We illustrate near/far matching by using Medicare data to compare the effectiveness of carotid arterial stents with cerebral protection versus carotid endarterectomy for the treatment of carotid stenosis.",
keywords = "Binary outcomes, Comparative effectiveness, Instrumental variables, Matching, Medicare data, Study design",
author = "Mike Baiocchi and Small, {Dylan S.} and Lin Yang and Polsky, {Daniel E.} and Groeneveld, {Peter W.}",
year = "2012",
month = "12",
day = "1",
doi = "10.1007/s10742-012-0091-0",
language = "English (US)",
volume = "12",
pages = "237--253",
journal = "Health Services and Outcomes Research Methodology",
issn = "1387-3741",
publisher = "Springer Netherlands",
number = "4",

}

TY - JOUR

T1 - Near/far matching

T2 - A study design approach to instrumental variables

AU - Baiocchi, Mike

AU - Small, Dylan S.

AU - Yang, Lin

AU - Polsky, Daniel E.

AU - Groeneveld, Peter W.

PY - 2012/12/1

Y1 - 2012/12/1

N2 - Classic instrumental variable techniques involve the use of structural equation modeling or other forms of parameterized modeling. In this paper we use a nonparametric, matching-based instrumental variable methodology that is based on a study design approach. Similar to propensity score matching, though unlike classic instrumental variable approaches, near/far matching is capable of estimating causal effects when the outcome is not continuous. Unlike propensity score matching, though similar to instrumental variable techniques, near/far matching is also capable of estimating causal effects even when unmeasured covariates produce selection bias. We illustrate near/far matching by using Medicare data to compare the effectiveness of carotid arterial stents with cerebral protection versus carotid endarterectomy for the treatment of carotid stenosis.

AB - Classic instrumental variable techniques involve the use of structural equation modeling or other forms of parameterized modeling. In this paper we use a nonparametric, matching-based instrumental variable methodology that is based on a study design approach. Similar to propensity score matching, though unlike classic instrumental variable approaches, near/far matching is capable of estimating causal effects when the outcome is not continuous. Unlike propensity score matching, though similar to instrumental variable techniques, near/far matching is also capable of estimating causal effects even when unmeasured covariates produce selection bias. We illustrate near/far matching by using Medicare data to compare the effectiveness of carotid arterial stents with cerebral protection versus carotid endarterectomy for the treatment of carotid stenosis.

KW - Binary outcomes

KW - Comparative effectiveness

KW - Instrumental variables

KW - Matching

KW - Medicare data

KW - Study design

UR - http://www.scopus.com/inward/record.url?scp=84871669938&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84871669938&partnerID=8YFLogxK

U2 - 10.1007/s10742-012-0091-0

DO - 10.1007/s10742-012-0091-0

M3 - Article

AN - SCOPUS:84871669938

VL - 12

SP - 237

EP - 253

JO - Health Services and Outcomes Research Methodology

JF - Health Services and Outcomes Research Methodology

SN - 1387-3741

IS - 4

ER -