Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Short-term and Long-term Exposure to Air pollution and Stroke risk

Oudin, Anna LU (2009) In Lund University Faculty of Medicine Doctoral Dissertation Series 2009:120.
Abstract
Background: Stroke is a major cause of mortality and morbidity across the world. Air pollution has become an acknowledged risk factor for stroke, especially for acute effects. There is less support for effects by chronic exposure due to the methodological challenges inherent in such studies. Such challenges can regard missing individual level data or residual confounding and can partly be overcome by employing two-phase methods. The general idea with such methods is to incorporate study subjects with partly missing data into the analysis. The main aims in this work were to investigate if air pollution increase stroke risk in Scania (Sweden), where pollution levels are rather low, and to investigate the performance of different two-phase... (More)
Background: Stroke is a major cause of mortality and morbidity across the world. Air pollution has become an acknowledged risk factor for stroke, especially for acute effects. There is less support for effects by chronic exposure due to the methodological challenges inherent in such studies. Such challenges can regard missing individual level data or residual confounding and can partly be overcome by employing two-phase methods. The general idea with such methods is to incorporate study subjects with partly missing data into the analysis. The main aims in this work were to investigate if air pollution increase stroke risk in Scania (Sweden), where pollution levels are rather low, and to investigate the performance of different two-phase methods.

Methods: Nitrogen oxides (NOx) concentrations were modelled with high resolution. Data from the National Stroke Register (Riks-stroke) were obtained on hospital admissions for ischemic and hemorrhagic stroke during the years 2001 to 2006. The association between annual mean levels of NOx (as a marker for long-term exposure to air pollution) and ischemic stroke was studied in a two-phase case-control design. The short-term association between daily levels of inhalable particulate air pollution (PM10), ozone, NOx and temperature and the number of strokes occurring each day was studied with time series methods. The performance of several different two-phase methods was compared in a simulated case-control study.

Results: No association between annual mean NOx and ischemic stroke was observed. Regarding short-term effects, high levels of PM10 increased the risk of ischemic stroke the next day, whereas high levels of temperature decreased the risk the next day. For hemorrhagic strokes, a similar, but less precise pattern was observed. Two-phase methods seemed to benefit from incorporating group-level data on exposure. The method which was most suitable for incorporating such data could be generalised to polytomous exposure categories and to include confounding factors.

Conclusions: No evidence for long-term exposure to air pollution to increase ischemic stroke risk in Scania was observed. Short-terms effects by PM10 were documented, contributing to evidence for risk increases even in low-level areas. The role of temperature should be investigated further. The use of two-phase methods was illustrated in both a methodologial study and an empirical study. (Less)
Please use this url to cite or link to this publication:
author
supervisor
opponent
  • Docent Forsberg, Bertil, Yrkes-och Miljömedicin, Umeå University
organization
publishing date
type
Thesis
publication status
published
subject
keywords
Long-term effects, Two-phase methods, Hemorrhagic stroke, Ischemic stroke, Short-term effects, Hospital admissions, Air Pollution
in
Lund University Faculty of Medicine Doctoral Dissertation Series
volume
2009:120
pages
162 pages
publisher
Occupational and Environmental Medicine
defense location
Blocket, F3
defense date
2009-12-17 09:15:00
ISSN
1652-8220
ISBN
978-91-86443-09-2
language
English
LU publication?
yes
id
319644d0-7f88-4c5e-9ab4-c098b80cbff2 (old id 1510693)
date added to LUP
2016-04-01 13:01:20
date last changed
2019-05-22 02:33:09
@phdthesis{319644d0-7f88-4c5e-9ab4-c098b80cbff2,
  abstract     = {{Background: Stroke is a major cause of mortality and morbidity across the world. Air pollution has become an acknowledged risk factor for stroke, especially for acute effects. There is less support for effects by chronic exposure due to the methodological challenges inherent in such studies. Such challenges can regard missing individual level data or residual confounding and can partly be overcome by employing two-phase methods. The general idea with such methods is to incorporate study subjects with partly missing data into the analysis. The main aims in this work were to investigate if air pollution increase stroke risk in Scania (Sweden), where pollution levels are rather low, and to investigate the performance of different two-phase methods. <br/><br>
Methods: Nitrogen oxides (NOx) concentrations were modelled with high resolution. Data from the National Stroke Register (Riks-stroke) were obtained on hospital admissions for ischemic and hemorrhagic stroke during the years 2001 to 2006. The association between annual mean levels of NOx (as a marker for long-term exposure to air pollution) and ischemic stroke was studied in a two-phase case-control design. The short-term association between daily levels of inhalable particulate air pollution (PM10), ozone, NOx and temperature and the number of strokes occurring each day was studied with time series methods. The performance of several different two-phase methods was compared in a simulated case-control study.<br/><br>
Results: No association between annual mean NOx and ischemic stroke was observed. Regarding short-term effects, high levels of PM10 increased the risk of ischemic stroke the next day, whereas high levels of temperature decreased the risk the next day. For hemorrhagic strokes, a similar, but less precise pattern was observed. Two-phase methods seemed to benefit from incorporating group-level data on exposure. The method which was most suitable for incorporating such data could be generalised to polytomous exposure categories and to include confounding factors.<br/><br>
Conclusions: No evidence for long-term exposure to air pollution to increase ischemic stroke risk in Scania was observed. Short-terms effects by PM10 were documented, contributing to evidence for risk increases even in low-level areas. The role of temperature should be investigated further. The use of two-phase methods was illustrated in both a methodologial study and an empirical study.}},
  author       = {{Oudin, Anna}},
  isbn         = {{978-91-86443-09-2}},
  issn         = {{1652-8220}},
  keywords     = {{Long-term effects; Two-phase methods; Hemorrhagic stroke; Ischemic stroke; Short-term effects; Hospital admissions; Air Pollution}},
  language     = {{eng}},
  publisher    = {{Occupational and Environmental Medicine}},
  school       = {{Lund University}},
  series       = {{Lund University Faculty of Medicine Doctoral Dissertation Series}},
  title        = {{Short-term and Long-term Exposure to Air pollution and Stroke risk}},
  url          = {{https://lup.lub.lu.se/search/files/3110377/1510695.pdf}},
  volume       = {{2009:120}},
  year         = {{2009}},
}