Spatial misalignment in time series studies of air pollution and health data

Roger Peng, Michelle L. Bell

Research output: Contribution to journalArticle

Abstract

Time series studies of environmental exposures often involve comparing daily changes in a toxicant measured at a point in space with daily changes in an aggregate measure of health. Spatial misalignment of the exposure and response variables can bias the estimation of health risk, and the magnitude of this bias depends on the spatial variation of the exposure of interest. In air pollution epidemiology, there is an increasing focus on estimating the health effects of the chemical components of particulate matter (PM). One issue that is raised by this new focus is the spatial misalignment error introduced by the lack of spatial homogeneity in many of the PM components. Current approaches to estimating short-term health risks via time series modeling do not take into account the spatial properties of the chemical components and therefore could result in biased estimation of those risks. We present a spatial-temporal statistical model for quantifying spatial misalignment error and show how adjusted health risk estimates can be obtained using a regression calibration approach and a 2-stage Bayesian model. We apply our methods to a database containing information on hospital admissions, air pollution, and weather for 20 large urban counties in the United States.

Original languageEnglish (US)
Pages (from-to)720-740
Number of pages21
JournalBiostatistics
Volume11
Issue number4
DOIs
StatePublished - Oct 2010

Fingerprint

Air Pollution
Misalignment
Health
Time series
Particulate Matter
Environmental Exposure
Weather
Statistical Models
Biased Estimation
Regression Calibration
Toxicants
Time Series Modelling
Calibration
Epidemiology
Bayesian Model
Air pollution
Databases
Homogeneity
Statistical Model
Health risk

Keywords

  • Acute health effects
  • Cardiovascular disease
  • Chemical speciation
  • Measurement error
  • Particulate matter
  • Spatial modeling

ASJC Scopus subject areas

  • Medicine(all)
  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

Spatial misalignment in time series studies of air pollution and health data. / Peng, Roger; Bell, Michelle L.

In: Biostatistics, Vol. 11, No. 4, 10.2010, p. 720-740.

Research output: Contribution to journalArticle

@article{4c63298d7d6f4b50957ef9e8114538e1,
title = "Spatial misalignment in time series studies of air pollution and health data",
abstract = "Time series studies of environmental exposures often involve comparing daily changes in a toxicant measured at a point in space with daily changes in an aggregate measure of health. Spatial misalignment of the exposure and response variables can bias the estimation of health risk, and the magnitude of this bias depends on the spatial variation of the exposure of interest. In air pollution epidemiology, there is an increasing focus on estimating the health effects of the chemical components of particulate matter (PM). One issue that is raised by this new focus is the spatial misalignment error introduced by the lack of spatial homogeneity in many of the PM components. Current approaches to estimating short-term health risks via time series modeling do not take into account the spatial properties of the chemical components and therefore could result in biased estimation of those risks. We present a spatial-temporal statistical model for quantifying spatial misalignment error and show how adjusted health risk estimates can be obtained using a regression calibration approach and a 2-stage Bayesian model. We apply our methods to a database containing information on hospital admissions, air pollution, and weather for 20 large urban counties in the United States.",
keywords = "Acute health effects, Cardiovascular disease, Chemical speciation, Measurement error, Particulate matter, Spatial modeling",
author = "Roger Peng and Bell, {Michelle L.}",
year = "2010",
month = "10",
doi = "10.1093/biostatistics/kxq017",
language = "English (US)",
volume = "11",
pages = "720--740",
journal = "Biostatistics",
issn = "1465-4644",
publisher = "Oxford University Press",
number = "4",

}

TY - JOUR

T1 - Spatial misalignment in time series studies of air pollution and health data

AU - Peng, Roger

AU - Bell, Michelle L.

PY - 2010/10

Y1 - 2010/10

N2 - Time series studies of environmental exposures often involve comparing daily changes in a toxicant measured at a point in space with daily changes in an aggregate measure of health. Spatial misalignment of the exposure and response variables can bias the estimation of health risk, and the magnitude of this bias depends on the spatial variation of the exposure of interest. In air pollution epidemiology, there is an increasing focus on estimating the health effects of the chemical components of particulate matter (PM). One issue that is raised by this new focus is the spatial misalignment error introduced by the lack of spatial homogeneity in many of the PM components. Current approaches to estimating short-term health risks via time series modeling do not take into account the spatial properties of the chemical components and therefore could result in biased estimation of those risks. We present a spatial-temporal statistical model for quantifying spatial misalignment error and show how adjusted health risk estimates can be obtained using a regression calibration approach and a 2-stage Bayesian model. We apply our methods to a database containing information on hospital admissions, air pollution, and weather for 20 large urban counties in the United States.

AB - Time series studies of environmental exposures often involve comparing daily changes in a toxicant measured at a point in space with daily changes in an aggregate measure of health. Spatial misalignment of the exposure and response variables can bias the estimation of health risk, and the magnitude of this bias depends on the spatial variation of the exposure of interest. In air pollution epidemiology, there is an increasing focus on estimating the health effects of the chemical components of particulate matter (PM). One issue that is raised by this new focus is the spatial misalignment error introduced by the lack of spatial homogeneity in many of the PM components. Current approaches to estimating short-term health risks via time series modeling do not take into account the spatial properties of the chemical components and therefore could result in biased estimation of those risks. We present a spatial-temporal statistical model for quantifying spatial misalignment error and show how adjusted health risk estimates can be obtained using a regression calibration approach and a 2-stage Bayesian model. We apply our methods to a database containing information on hospital admissions, air pollution, and weather for 20 large urban counties in the United States.

KW - Acute health effects

KW - Cardiovascular disease

KW - Chemical speciation

KW - Measurement error

KW - Particulate matter

KW - Spatial modeling

UR - http://www.scopus.com/inward/record.url?scp=77956210910&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77956210910&partnerID=8YFLogxK

U2 - 10.1093/biostatistics/kxq017

DO - 10.1093/biostatistics/kxq017

M3 - Article

C2 - 20392805

AN - SCOPUS:77956210910

VL - 11

SP - 720

EP - 740

JO - Biostatistics

JF - Biostatistics

SN - 1465-4644

IS - 4

ER -