A Register-based Case-control Study of Air Pollution and Myocardial Infarction

Abstract

WM2-O-02 Introduction: Long-term exposure to air pollution has been associated with an increased risk for cardiopulmonary mortality in cohort studies. The aim of this study is to investigate the association between myocardial infarction (MI) and long-term residential exposure to source-specific air pollution in a large register-based case-control study. Methods: The study population includes all first events of MI in people 15 to 79 years of age during 1985 to 1996 from a regional MI-register (n = 45,000). Population controls were randomly selected from the study base matched on age and sex (n = 507,000). Individual data on occupation and social class for cases and controls were obtained using historical national censuses of the years 1970, 1975, 1980, 1985, and 1990, and geographic coordinates of home addresses for the years 1980, 1985, and 1990. Annual air pollution exposure was assessed according to the geographic location of the subject's home addresses for the years 1980, 1985, and 1990 using dispersion models based on historical emission databases according to statistics of changes in traffic and other historical land-use variations. These databases describe emissions of nitrogen dioxide (NO2), carbon monoxide (CO), and particulate matter (PM10) from traffic, and sulfur dioxide (SO2) from heating in resolutions of 500 × 500 m in rural areas and 25 × 25 m in the inner-city area. Because of lack of reliable historic data, PM10 was assessed only for the year 2000, thus assuming constant levels during the whole study period. Air pollution exposure was assessed for 19,793 MI cases and 192,225 controls with complete address information within the study area during all censuses. Results: Five-year average residential traffic-related NO2 exposure from the 5th to the 95th percentile before study inclusion was associated with an odds ratio adjusted for age, sex, and socioeconomic status of 1.08 (95% CI, 1.03–1.14) for MI and 1.19 (95% CI, 1.02–1.38) for fatal MI. The other pollutant indicators including SO2 from heating were less associated with MI. Average exposure windows further back in time did not indicate any independent association. Somewhat stronger associations were suggested among those who did not change their home address between the population censuses. Discussion and Conclusions: The limitations of this study include lack of smoking data and misclassification of exposure due to unknown historical addresses. The results of this study add some support to the hypothesis that long-term traffic-related air pollution exposure is associated with an increased risk of cardiovascular disease.