Adjusting for unmeasured spatial confounding with distance adjusted propensity score matching

Georgia Papadogeorgou, Christine Choirat, Corwin Zigler

Research output: Contribution to journalArticle

Abstract

Propensity score matching is a common tool for adjusting for observed confounding in observational studies, but is known to have limitations in the presence of unmeasured confounding. In many settings, researchers are confronted with spatially-indexed data where the relative locations of the observational units may serve as a useful proxy for unmeasured confounding that varies according to a spatial pattern. We develop a new method, termed distance adjusted propensity score matching (DAPSm) that incorporates information on units' spatial proximity into a propensity score matching procedure. We show that DAPSm can adjust for both observed and some forms of unobserved confounding and evaluate its performance relative to several other reasonable alternatives for incorporating spatial information into propensity score adjustment. The method is motivated by and applied to a comparative effectiveness investigation of power plant emission reduction technologies designed to reduce population exposure to ambient ozone pollution. Ultimately, DAPSm provides a framework for augmenting a "standard" propensity score analysis with information on spatial proximity and provides a transparent and principled way to assess the relative trade-offs of prioritizing observed confounding adjustment versus spatial proximity adjustment.

Original languageEnglish (US)
Pages (from-to)256-272
Number of pages17
JournalBiostatistics
Volume20
Issue number2
DOIs
StatePublished - Apr 1 2019

Fingerprint

Propensity Score
Confounding
Proximity
Adjustment
Observational Study
Unit
Ozone
Spatial Information
Spatial Pattern
Power Plant
Propensity score matching
Pollution
Trade-offs
Vary
Spatial proximity
Evaluate
Alternatives

Keywords

  • Propensity score matching
  • Spatial confounding
  • Unobserved confounding

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

Adjusting for unmeasured spatial confounding with distance adjusted propensity score matching. / Papadogeorgou, Georgia; Choirat, Christine; Zigler, Corwin.

In: Biostatistics, Vol. 20, No. 2, 01.04.2019, p. 256-272.

Research output: Contribution to journalArticle

Papadogeorgou, Georgia ; Choirat, Christine ; Zigler, Corwin. / Adjusting for unmeasured spatial confounding with distance adjusted propensity score matching. In: Biostatistics. 2019 ; Vol. 20, No. 2. pp. 256-272.
@article{e7fafe15a4e04d649372b3dba1a22675,
title = "Adjusting for unmeasured spatial confounding with distance adjusted propensity score matching",
abstract = "Propensity score matching is a common tool for adjusting for observed confounding in observational studies, but is known to have limitations in the presence of unmeasured confounding. In many settings, researchers are confronted with spatially-indexed data where the relative locations of the observational units may serve as a useful proxy for unmeasured confounding that varies according to a spatial pattern. We develop a new method, termed distance adjusted propensity score matching (DAPSm) that incorporates information on units' spatial proximity into a propensity score matching procedure. We show that DAPSm can adjust for both observed and some forms of unobserved confounding and evaluate its performance relative to several other reasonable alternatives for incorporating spatial information into propensity score adjustment. The method is motivated by and applied to a comparative effectiveness investigation of power plant emission reduction technologies designed to reduce population exposure to ambient ozone pollution. Ultimately, DAPSm provides a framework for augmenting a {"}standard{"} propensity score analysis with information on spatial proximity and provides a transparent and principled way to assess the relative trade-offs of prioritizing observed confounding adjustment versus spatial proximity adjustment.",
keywords = "Propensity score matching, Spatial confounding, Unobserved confounding",
author = "Georgia Papadogeorgou and Christine Choirat and Corwin Zigler",
year = "2019",
month = "4",
day = "1",
doi = "10.1093/biostatistics/kxx074",
language = "English (US)",
volume = "20",
pages = "256--272",
journal = "Biostatistics",
issn = "1465-4644",
publisher = "Oxford University Press",
number = "2",

}

TY - JOUR

T1 - Adjusting for unmeasured spatial confounding with distance adjusted propensity score matching

AU - Papadogeorgou, Georgia

AU - Choirat, Christine

AU - Zigler, Corwin

PY - 2019/4/1

Y1 - 2019/4/1

N2 - Propensity score matching is a common tool for adjusting for observed confounding in observational studies, but is known to have limitations in the presence of unmeasured confounding. In many settings, researchers are confronted with spatially-indexed data where the relative locations of the observational units may serve as a useful proxy for unmeasured confounding that varies according to a spatial pattern. We develop a new method, termed distance adjusted propensity score matching (DAPSm) that incorporates information on units' spatial proximity into a propensity score matching procedure. We show that DAPSm can adjust for both observed and some forms of unobserved confounding and evaluate its performance relative to several other reasonable alternatives for incorporating spatial information into propensity score adjustment. The method is motivated by and applied to a comparative effectiveness investigation of power plant emission reduction technologies designed to reduce population exposure to ambient ozone pollution. Ultimately, DAPSm provides a framework for augmenting a "standard" propensity score analysis with information on spatial proximity and provides a transparent and principled way to assess the relative trade-offs of prioritizing observed confounding adjustment versus spatial proximity adjustment.

AB - Propensity score matching is a common tool for adjusting for observed confounding in observational studies, but is known to have limitations in the presence of unmeasured confounding. In many settings, researchers are confronted with spatially-indexed data where the relative locations of the observational units may serve as a useful proxy for unmeasured confounding that varies according to a spatial pattern. We develop a new method, termed distance adjusted propensity score matching (DAPSm) that incorporates information on units' spatial proximity into a propensity score matching procedure. We show that DAPSm can adjust for both observed and some forms of unobserved confounding and evaluate its performance relative to several other reasonable alternatives for incorporating spatial information into propensity score adjustment. The method is motivated by and applied to a comparative effectiveness investigation of power plant emission reduction technologies designed to reduce population exposure to ambient ozone pollution. Ultimately, DAPSm provides a framework for augmenting a "standard" propensity score analysis with information on spatial proximity and provides a transparent and principled way to assess the relative trade-offs of prioritizing observed confounding adjustment versus spatial proximity adjustment.

KW - Propensity score matching

KW - Spatial confounding

KW - Unobserved confounding

UR - http://www.scopus.com/inward/record.url?scp=85062685966&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85062685966&partnerID=8YFLogxK

U2 - 10.1093/biostatistics/kxx074

DO - 10.1093/biostatistics/kxx074

M3 - Article

C2 - 29365040

AN - SCOPUS:85062685966

VL - 20

SP - 256

EP - 272

JO - Biostatistics

JF - Biostatistics

SN - 1465-4644

IS - 2

ER -