Secondary Logo

Journal Logo

Original Article

Long-Term Exposure to Urban Air Pollution and Myocardial Infarction

Rosenlund, Mats*†; Berglind, Niklas*†; Pershagen, Göran*†; Hallqvist, Johan‡§; Jonson, Tage; Bellander, Tom*†

Author Information
doi: 10.1097/01.ede.0000219722.25569.0f
  • Free


The effects of short-term exposure to outdoor air pollution are well documented in numerous epidemiologic studies during the last decades.1–8 Only a handful of studies have investigated health effects of long-term ambient air pollution exposure,8–13 and the health risks from long-term exposure are less well characterized, although the public health implications may be more substantial.14–16 For example, a reduction of air pollution in Dublin due to a coal ban in 1990 was shown to reduce the cardiovascular mortality by 10%.17 The epidemiologic evidence indicates that health effects can occur at air pollution levels currently experienced in large cities, resulting in reevaluation of air pollution standards in many countries.18–20

Proposed mechanisms for an association between short-term exposure to air pollution and cardiovascular effects mainly include effects on blood coagulation and cardiac function.7,21 There are reports of effects on blood fibrinogen levels associated with urban air pollution.22–24 Air pollution has also been associated with heart rate variability, pulse rate, and cardiac arrhythmia.25–32 Furthermore, an association between air pollution and the activation of automatic defibrillators, implanted in patients to prevent life-threatening cardiac arrhythmia, has also been investigated.29,33,34 There is still a lack of conclusive evidence as to which mechanisms may be important with regard to long-term air pollution exposure and myocardial infarction (MI).

Long-term exposure to air pollution has been associated with cardiopulmonary mortality in cohort studies.8–12 The air pollution assessment has been based on average air pollution levels from urban background monitors in large metropolitan areas, ignoring individual differences due to within-city variation,8–10 or calculated according to large grids (kilometers) or as the distance to major roads.11,12 Usually only the baseline address has been available, and changes in residential pollution exposure due to subsequent relocation during the follow up have therefore not been considered. Although some studies assessed potentially sensitive subgroups for health effects from air pollution, there is a lack of conclusive data on possible effect modification by other factors. Some of these issues were addressed in reanalyses of the U.S. Six Cities and American Cancer Society cohort studies,35 but many methodologic aspects still need further elaboration. Because cohort studies have investigated mortality, little is known of potential associations of long-term air pollution exposure with incidence or nonfatal cardiovascular disease. Recent attention has focused on health effects from traffic-generated air pollution,11,12 and it is important to obtain risk estimates separately for different air pollution sources. The aim of our study was to investigate the association between long-term exposure to source-specific urban air pollution and MI morbidity and mortality using a methodology of historical emission data, dispersion modeling, and geographic information system (GIS) coding to assess individual air pollution exposure during 30 years.36,37


Study Subjects

The study originates from a large population-based case–control study of MI in Stockholm (Stockholm Heart Epidemiology Program [SHEEP]) described in detail elsewhere.38 The study includes all first events of MI among Swedish citizens in the ages 45 to 70 years residing in Stockholm county during 1992 to 1993 (1992 to 1994 for women) and population controls from the corresponding study base. Cases of first-time MI were identified from the coronary and intensive care units at emergency hospitals in Stockholm county, the Hospital Discharge Register for the county, or death certificates from the National Cause of Death Register at Statistics Sweden using standard diagnostic criteria.38 Controls with no history of previous MI were randomly selected from the study base after stratification on age, sex, and hospital catchment area. In total, the study included 2246 cases and 3206 controls. A postal questionnaire was answered by 4067 subjects with a response rate among cases of 72% for women and 81% for men, whereas the corresponding figures among controls were 70% and 75%. The subjects responded to the same extent in different age groups and catchment areas.

Exposure Information

The questionnaire covered a large set of potential risk factors for MI, including physical and psychosocial work environment, social factors, lifestyle factors, and dietary intake. A supplementary telephone interview was performed to reduce nonresponse and missing data. A health examination was also carried out (except for fatal subjects) to collect data on various biologic parameters related to cardiovascular disease. The biologic variables for analysis were primarily based on data from the health examination, but questionnaire information was used for some variables (eg, body mass index [BMI]) for subjects not participating in the clinical testing. The exposure data and definitions of the covariates have been described in other publications.38–44

Air Pollution Assessment

The air pollution exposure was assessed using a methodology previously developed for a study on air pollution and lung cancer in Stockholm36 and is described in detail elsewhere.37 Briefly, all addresses inhabited during more than 2 years since 1960 were transformed into geographic coordinates using standard GIS computer software45 in combination with a regional geographic address database.46 Address data was given in the questionnaire and data gaps were completed using register information from parish offices and tax authorities. No more than 5 years of missing address information was allowed leading to exclusion of 357 cases and 445 controls, mainly due to residency outside the county and information gaps in early years. The computer geocoding process successfully assessed the geographic location for 87% of totally 10,662 addresses either automatically (63%) or interactively (37%) by an operator adjusting street spelling or street number within the software. The remaining 13% were geocoded manually using a digital map to locate the exact position (4%), the correct street (6%), or the correct village, parish, or similar small area (3%).

Emission databases within Stockholm county, describing the air pollution from different sources during each decade since 1960, provided information for dispersion calculations to obtain estimates of annual mean levels of traffic-generated nitrogen oxides (NOx) and nitrogen dioxide (NO2), carbon monoxide (CO), and particulate matter less than 10 μm (PM10) and 2.5 μm (PM2.5) in aerodynamic diameter, respectively, as well as sulfur dioxide (SO2) from heating. The annual mean levels of traffic-generated PM10 and PM2.5 were calculated using an emission database from the year 2000, thus assuming constant levels during the study period. Years of missing air pollution data due to unknown residency were replaced by the mean among the controls for that year.47 The geographic distribution of air pollution was assessed in 3 layers of different resolution applied to regional/countryside area (500 × 500 m), urban area (100 × 100 m), and inner-city area (25 × 25 m). Calibrations of the models were performed to minimize deviation when compared with available measured levels of total concentrations for the corresponding period. In addition, because the air pollution levels in the city may vary considerably depending on local traffic conditions and distance to the ground of the dwelling, a street canyon contribution was used for addresses in the most polluted street segments in the city center (n = 267). The street canyon contribution was based on measured concentrations above roof and at street level and was estimated respectively for wide (≥30 m) and narrow canyons (<30 m). These were street sections with buildings on both sides and defined as air pollution hot spots, ie, where the PM10 levels exceeded the European Union 24-hour limit value of 50 μg/m3 (90th percentile) during 2002. Modeled annual NO2 levels using all sources in the 1995 edition of the emission database correlated well (r = 0.96) with measured levels in 16 locations throughout the county that had not been used to calibrate the models.48 The air pollution data for each year from the dispersion models were linked to the individual address coordinate for each subject for the corresponding year. Thus, annual individual residential source-specific levels of NOx, NO2, CO, SO2, PM10, and PM2.5 from 1960 until study inclusion (1992 to1994) were calculated for 1397 cases of MI and 1870 controls. In total, 4% of all residential exposure years were missing and replaced by the annual mean among the controls. Most subjects with missing data in any year had 1 year absent, and 110 of the 3267 subjects had at most 4 of 30 years missing. In essence, the study design makes it possible to separate traffic-related NO2, CO, and particulate matter emissions from those generated by other sources. Similarly, SO2 was used as a marker for emissions from all heating sources (mainly local oil-fueled residential heating). Because of the high correlation in the sample between the particulate matter measures (r = 0.998) and between NOx and NO2 (r = 0.930) due to similar dispersion patterns, only the results for NO2 and PM10 are presented.

Statistical Analyses

The association between air pollution exposure and MI was analyzed using logistic regression, and the results are presented as odds ratios (ORs) and 95% confidence intervals (95% CIs). We adjusted the logistic regression models for the matching variables of age (5-year age groups), sex, and hospital catchment area (10 categories). Important covariates included in the multivariate model were smoking (never, former, and current smokers of 1–10 g/d, 11–20 g/d, or more than 20 g/d), physical inactivity (dichotomous),44 diabetes (dichotomous), and socioeconomic status (3 levels).38 We also evaluated other models, including hypertension,38 BMI, job strain,39 diet, passive smoking,40 alcohol consumption,49 coffee intake,42 and occupational exposure to motor exhaust and other combustion products,43 but these covariates indicated less impact on the risk estimates for air pollution. Air pollution variables were used as continuous variables (results presented for the 5th to 95th percentile difference according to the distribution among controls). Analysis of cases as nonfatal, fatal in the hospital within 28 days, and out-of-hospital death within 28 days was performed by multinomial logistic regression using all controls as reference. To explore potential effect modification, analyses were performed with interaction terms between the pollutants and other variables. All statistical analyses were performed with Stata 8.2 (Stata Corp., College Station, TX). The study was approved by the ethical committee at Karolinska Institutet.


The study population and the distribution of exposure among cases and controls are displayed in Table 1. A higher proportion of cases than controls were current smokers, were physically inactive, had diabetes, were blue collar workers, were overweight, had hypertension, and were exposed to job strain or combustion products at work, whereas the proportion of never smokers, exsmokers, and white collar workers was higher among controls. The average consumption of coffee and alcohol was slightly higher among the cases. The estimated level of long-term residential air pollution exposure was slightly higher among cases than controls. The 30-year population average level of NO2 was strongly correlated with PM10 (r = 0.93) and somewhat less correlated with CO (r = 0.74) and SO2 (r = 0.73). CO was least correlated with SO2 (r = 0.49) and PM10 (r = 0.66).

Selected Characteristics of Myocardial Infarction Cases and Population Controls in Stockholm 1992 to 1994

There was no association between long-term air pollution exposure and overall MI incidence. Classifying the cases according to whether they survived their infarction or died within 28 days suggested increased risks of fatal MI associated with the 30-year average air pollution exposure, especially for NO2 and PM10 from motor vehicles (Table 2). The highest estimates tended to appear among those who died outside the hospital. Analyses of potential effect modification suggested some differences in risk according to sex, smoking habits, education, and diabetes (Fig. 1). The association appeared strongest among women, never smokers, those with more than a high school education, and nondiabetics.

Association of 30-Yr Average Exposure to Air Pollution From Traffic (NO2, CO, and PM10) and Heating (SO2) With Myocardial Infarction
Adjusted odds ratios (diamonds) and 95% confidence intervals (vertical lines) for fatal myocardial infarction associated with a difference from the 5th to the 95th percentile of the 30-year average exposure to traffic-generated nitrogen dioxide (NO2), corresponding approximately to 30 μg/m3, stratified by sex, smoking habits, educational level (highest level attained [HS = high school]), and diabetes.

Several variables exerted positive confounding in analyses of MI risks related to air pollution exposure (eg, smoking), and risk estimates for NO2 were reduced after adjustment (Fig. 2). Additional adjustment for heating by including SO2 in the model tended to increase the estimates of the traffic-generated pollutants, whereas the SO2 estimate decreased (data not shown). The OR for fatal MI associated with a difference from the 5th to the 95th percentile of the 30-year average NO2 exposure increased to 1.64 (95% CI = 0.90–3.01) when adjusted for SO2. The corresponding SO2 estimate changed to 0.87 (0.45–1.69). After adjustment for heating-related SO2, the estimates for fatal MI was 1.21 (0.95–1.54) for CO and 1.40 (0.86–2.26) for PM10.

Odds ratios (diamonds) and 95% confidence intervals (horizontal lines) for fatal myocardial infarction associated with 30-year average residential traffic-generated NO2 exposure. Values are calculated for a difference in estimated exposure from the 5th to the 95thpercentile according to the distribution among controls. Analyses were made on subjects with complete data on all variables in the model, ie, n = 3210 for the final model (A), and n = 2797 for the full model (B). All models included the matching variables. The asterisk (*) indicates comparison model, ie, to which each model with a new variable under A and B should be compared. Plus sign (+) indicates that the variable was added to the comparison model above, ie, each variable was included in the model one at a time and then removed.

Analysis in different time windows did not indicate substantially different associations than the 30-year average exposure. However, because of low mobility of the population across the study period, the power to detect any such temporal patterns was low; for example, the correlation between the 30-year average NO2 exposure level and the levels during each of the 3 decades was between 0.85 and 0.95.

For subjects who had ever lived at an address that was assigned a street canyon contribution to the air pollution level, the adjusted OR for MI was 1.23 (0.85–1.78) and 1.40 (0.78–2.52) for fatal MI, suggesting that a dichotomous classification of residency close to hot spot streets might capture much of the long-term risk associated with traffic-generated air pollution. The adjusted OR for fatal MI associated with ever residency at such hot spot streets with additional adjustment for SO2 was 1.58 (0.86–2.90).


This study did not indicate an association between long-term air pollution exposure and overall incidence or nonfatal MI. However, the results suggest an increased risk of fatal MI associated with 30 years of residential exposure to air pollution, especially for out-of-hospital death. Our results are in accordance with previous cohort studies reporting an association between long-term air pollution exposure and cardiovascular mortality.9,11,12,50,51 We used traffic-generated NO2, CO, and PM10 as surrogates for the pollution mixture from road traffic, and we used SO2 as an indicator of air pollution from residential heating. The importance of the traffic pollutant increased in multipollutant models that incorporated both a traffic-related component and the marker for heating, suggesting an association primarily between long-term traffic-generated air pollution exposure and fatal MI.

Because the aim of this study was to investigate the regional source-specific contribution of air pollution in relation to MI, we did not assess the total levels of these pollutants. Focusing on the total levels by adding the long-range transported fraction and contributions from other regional sources would not notably change the spatial contrast in the study. According to measurements from monitors throughout the region, the annual average regional background level of NO2 during 2000 was 3 μg/m3, the urban background was 20 μg/m3, and the inner-city street level was 45 μg/m3.52 The CO levels were 200 μg/m3, 300 μg/m3, and 950 μg/m3, respectively; for PM10, they were 12 μg/m3, 17 μg/m3, 40 μg/m3; and for SO2, they were 0.5 μg/m3, 2 μg/m3, and 2 μg/m3.

We assessed exposure using residential address information from different sources, detailed historical emission data, dispersion modeling, and GIS techniques. This method was developed for the purposes of quantifying annual long-term air pollution exposure in a previously reported case–control study on lung cancer36,37; it has been improved to now allow for a resolution up to 25 m grids. The exposure was assessed without knowledge of case–control status, making differential misclassification of exposure unlikely. Nondifferential misclassification of residential exposure, however, is likely to occur and may have contributed to the overall null effects. Such bias could result from incorrect address information, errors in geographic location of addresses, or inexact dispersion calculations, especially for calculated levels far back in time, because of less reliable emission data. However, the address information was collected from several partly overlapping sources, the exact geographic location could not be found for only 9% of the addresses using both automatic and manual procedures, and the dispersion calculations have been calibrated and improved thoroughly over the last decade. We considered only residential data, thus ignoring outdoor air pollution at other locations, including the workplace. The resulting imprecision in the exposure assessment might further have attenuated any associations. Data quality may be of special concern regarding proxy reporting of historical addresses for fatal cases and for years long before infarction, which would lead to underestimation of the effect, although parish office and tax authority registers were used to minimize such bias.

The participation rate was 75% overall and not systematically different across geographic areas, speaking against exposure-related selection bias. In addition, the missing exposure information due to years with incomplete address data might have reduced exposure contrast, because replacement of missing data was based on the annual specific mean among the controls47; however, this issue applies to only 4% of all exposure years. To reduce this potential problem, we excluded subjects with more than 5 years of unknown address history.

The exposure assessment in our study has several advantages compared with previously reported cohort studies of long-term exposure to air pollution. In some earlier studies, the exposure was assessed for groups of individuals residing in the same city or close to the same pollution monitor and thus more crude.9,13,50,51 Other previous studies used less accurate modeling approaches such as the distance to major roads11 or much larger exposure grids.12

Other strengths of our study include a high reliability of case identification and good quality of diagnosis using several sources of information to ascertain the MI cases together with application of common diagnostic criteria and a comparatively high autopsy rate.38 Thus, potential misclassification of disease or bias due to nonresponse was minimized. It is unlikely that the identification or diagnosis of MI would be different in areas with higher air pollution levels; it is also unlikely that people in such areas would have differential rates of study participation or that they would systematically respond differently on questions regarding their lifestyle and habits. In addition, potential confounding from a large set of exposures and cardiovascular risk factors was considered.

Comparison of estimates with other mortality studies is difficult, because we have modeled historical source-specific levels at the subject's home address rather than using measurements of total ambient concentrations at a few sites at study baseline, or other modeling approaches using distance to major streets or total concentrations. Nevertheless, the OR for fatal MI associated with an increase in traffic-generated NOx of 10 μg/m3 in our study (1.06 [95% CI = 0.99–1.13]) might be compared with a risk ratio for ischemic heart disease death associated with a 10-μg/m3 increase in exposure to NOx in the cohort of Norwegian men (1.08 [95% CI = 1.03–1.12]), where monitored concentrations and emission data were used to calculate dispersion in kilometer grids.12 The Dutch cohort study showed a relative risk of 1.81 (95% CI = 0.98–3.34) for cardiopulmonary mortality associated with a change in NO2 concentration from the 5th to the 95th percentile from both urban background and local contribution, which was rounded to 30 μg/m3 and thus of equal magnitude as in our study, whereas the result was 1.54 (0.81–2.92) when considering only the background concentration.11 Our estimate for fatal MI associated with the same increase in traffic-generated NO2 was an OR of 1.51 (0.96–2.37). Our results are also in accordance with 2 U.S. cohort studies. The Six Cities study reported a relative risk for the most polluted compared with the least polluted city of 1.37 (95% CI = 1.11–1.68) for cardiopulmonary mortality, corresponding to a change in NO2 of approximately 30 μg/m3 (6 to 22 ppb).9 The relative risk of cardiopulmonary mortality comparing the least polluted with the most polluted city in the American Cancer Society study was 1.31 (95% CI = 1.17–1.46), equivalent to an increase of approximately 24 μg/m3 ofPM2.5.10

Possible biologic mechanisms for an association between long-term air pollution exposure and cardiovascular disease are primarily systemic inflammation (as indicated by increased levels of inflammatory markers in the blood),22–24 progression of atherosclerosis,53 and altered cardiac function by changes in heart rate and blood pressure.25–32 Our results suggest the strongest association among those who died outside of the hospital, which implies that sudden death might be of special importance in relation to long-term air pollution exposure. Persons who die of MI before reaching a hospital may have had an underlying heart disease to which chronic air pollution exposure might have contributed.7,21 However, it is not possible to assess etiologic mechanisms in detail in our study.

Fine-particle emissions from traffic have been proposed as the agent responsible for the increased cardiovascular risk from ambient air pollution.2 Although we modeled particulate matter for all addresses in our study, the long-term assessment of exposure to particulate matter assumed constant levels during 1960 to 2000 due to lack of historical measurements and past emission data for particles. This makes our long-term exposure estimates for particles somewhat less valid. However, the high correlations among the main traffic indicators NO2, CO, and PM10 and the strong similarities in the estimated effects suggest that the gaseous pollutants probably are good proxies for particulate exposure. In general, the risk estimates for CO seemed to be more precise than those for NO2, although these were indicators of the same source, ie, traffic. Because CO is a more direct primary pollutant from traffic, it has a steeper geographic gradient, which was reflected by a greater contribution for street canyons and thus larger contrast in exposure. Nevertheless, the source-specific pollutants are spatially correlated, and the results for the single traffic pollutants should be regarded as different estimates of the association between traffic emissions and MI.

There was a suggestion of effect modification by sex, smoking, education, and diabetes. Potential effect modification by smoking and educational level has also been reported previously.35,50 However, the only covariate demonstrating statistically significant effect modification in the reanalysis of the 2 previous U.S. cohorts was education, showing a decreasing risk with increasing education,35 contrary to our findings. This inconsistency may be due to differences between the countries in air pollution exposure in socioeconomic groups, because socioeconomically privileged people in our study tend to live in the city center and certain suburbs in close vicinity of the city, whereas another demographic situation may be present in U.S. cities.

We adjusted for important covariates, but residual confounding and potential influence from other unmeasured factors cannot be ruled out. Our results did not appear to be sensitive to different analytic approaches. Furthermore, we found that living close to major hot spot roads that are narrow and have dense traffic might be a good proxy for the risk of MI from long-term traffic-generated urban air pollution. Similar results have been reported by Hoek and coworkers11 showing high-risk estimates for fatal cardiovascular disease among people living close major roads in The Netherlands.

In conclusion, the results from this study did not indicate any association between long-term ambient air pollution exposure and MI incidence, but they provide some support for an association between long-term air pollution exposure and cardiovascular mortality.


1. Le Tertre A, Medina S, Samoli E, et al. Short-term effects of particulate air pollution on cardiovascular diseases in eight European cities. JEpidemiol Community Health. 2002;56:773–779.
2. Samet JM, Dominici F, Curriero FC, et al. Fine particulate air pollution and mortality in 20 US cities, 1987–1994. N Engl J Med. 2000;343:1742–1749.
3. Katsouyanni K, Touloumi G, Samoli E, et al. Confounding and effect modification in the short-term effects of ambient particles on total mortality: results from 29 European cities within the APHEA2 project. Epidemiology. 2001;12:521–531.
4. Samet JM, Zeger SL, Dominici F, et al. The National Morbidity, Mortality, and Air Pollution Study. Part II: Morbidity and mortality from air pollution in the United States. Res Rep Health Eff Inst. 2000;94:5–70 discussion 71–79.
5. Bell ML, Samet JM, Dominici F. Time-series studies of particulate matter. Annu Rev Public Health. 2004;25:247–280.
6. Goldberg MS, Burnett RT, Stieb D. A review of time-series studies used to evaluate the short-term effects of air pollution on human health. Rev Environ Health. 2003;18:269–303.
7. Dockery DW. Epidemiologic evidence of cardiovascular effects of particulate air pollution. Environ Health Perspect. 2001;109(suppl 4):483–486.
8. Abbey DE, Nishino N, McDonnell WF, et al. Long-term inhalable particles and other air pollutants related to mortality in nonsmokers. Am J Respir Crit Care Med. 1999;159:373–382.
9. Dockery DW, Pope CA 3rd, Xu X, et al. An association between air pollution and mortality in six US cities. N Engl J Med. 1993;329:1753–1759.
10. Pope CA 3rd, Thun MJ, Namboodiri MM, et al. Particulate air pollution as a predictor of mortality in a prospective study of US adults. Am J Respir Crit Care Med. 1995;151:669–674.
11. Hoek G, Brunekreef B, Goldbohm S, et al. Association between mortality and indicators of traffic-related air pollution in The Netherlands: a cohort study. Lancet. 2002;360:1203–1209.
12. Nafstad P, Haheim LL, Wisloff T, et al. Urban air pollution and mortality in a cohort of Norwegian men. Environ Health Perspect. 2004;112:610–615.
13. Grazuleviciene R, Maroziene L, Dulskiene V, et al. Exposure to urban nitrogen dioxide pollution and the risk of myocardial infarction. Scand J Work Environ Health. 2004;30:293–298.
14. Brunekreef B, Holgate ST. Air pollution and health. Lancet. 2002;360:1233–1242.
15. Medina S, Plasencia A, Ballester F, et al. Apheis: public health impact of PM10 in 19 European cities. J Epidemiol Community Health. 2004;58:831–836.
16. Kunzli N, Kaiser R, Medina S, et al. Public-health impact of outdoor and traffic-related air pollution: a European assessment. Lancet. 2000;356:795–801.
17. Clancy L, Goodman P, Sinclair H, et al. Effect of air-pollution control on death rates in Dublin, Ireland: an intervention study. Lancet. 2002;360:1210–1214.
18. World Health Organization. Air quality guidelines for Europe. WHO Reg Publ Eur Ser. 2000;V–X:1–273.
19. Health effects of outdoor air pollution. Committee of the Environmental and Occupational Health Assembly of the American Thoracic Society. Am J Respir Crit Care Med. 1996;153:3–50.
20. World Health Organization. Health aspects of air pollution: results from the WHO project ‘Systematic review of health aspects of air pollution in Europe.’ WHO Regional Office for Europe, Copenhagen, June 2004. (document EUR/04/5046026). Available at: Accessed 2 November 2004.
21. Brook RD, Franklin B, Cascio W, et al. Air pollution and cardiovascular disease: a statement for healthcare professionals from the Expert Panel on Population and Prevention Science of the American Heart Association. Circulation. 2004;109:2655–2671.
22. Pekkanen J, Brunner EJ, Anderson HR, et al. Daily concentrations of air pollution and plasma fibrinogen in London. Occup Environ Med. 2000;57:818–822.
23. Holgate ST, Devlin RB, Wilson SJ, et al. Health effects of acute exposure to air pollution. Part II: Healthy subjects exposed to concentrated ambient particles. Res Rep Health Eff Inst. 2003;31–50 discussion 51–67.
24. Schwartz J. Air pollution and blood markers of cardiovascular risk. Environ Health Perspect. 2001;109(suppl 3):405–409.
25. Holguin F, Tellez-Rojo MM, Hernandez M, et al. Air pollution and heart rate variability among the elderly in Mexico City. Epidemiology. 2003;14:521–527.
26. Gong H Jr, Linn WS, Terrell SL, et al. Altered heart-rate variability in asthmatic and healthy volunteers exposed to concentrated ambient coarse particles. Inhal Toxicol. 2004;16:335–343.
27. Chan CC, Chuang KJ, Shiao GM, et al. Personal exposure to submicrometer particles and heart rate variability in human subjects. Environ Health Perspect. 2004;112:1063–1067.
28. Pope CA, Dockery DW, Kanner RE, et al. Oxygen saturation, pulse rate, and particulate air pollution: a daily time-series panel study. Am J Respir Crit Care Med. 1999;159:365–372.
29. Peters A, Liu E, Verrier RL, et al. Air pollution and incidence of cardiac arrhythmia. Epidemiology. 2000;11:11–17.
30. Liao D, Duan Y, Whitsel EA, et al. Association of higher levels of ambient criteria pollutants with impaired cardiac autonomic control: a population-based study. Am J Epidemiol. 2004;159:768–777.
31. Pope CA 3rd, Hansen ML, Long RW, et al. Ambient particulate air pollution, heart rate variability, and blood markers of inflammation in a panel of elderly subjects. Environ Health Perspect. 2004;112:339–345.
32. Devlin RB, Ghio AJ, Kehrl H, et al. Elderly humans exposed to concentrated air pollution particles have decreased heart rate variability. Eur Respir J Suppl. 2003;40:76s–80s.
33. Vedal S, Rich K, Brauer M, et al. Air pollution and cardiac arrhythmias in patients with implantable cardioverter defibrillators. Inhal Toxicol. 2004;16:353–362.
34. Rich KE, Petkau J, Vedal S, et al. A case–crossover analysis of particulate air pollution and cardiac arrhythmia in patients with implantable cardioverter defibrillators. Inhal Toxicol. 2004;16:363–372.
35. Health Effects Institute. 2000. Reanalysis of the Harvard Six Cities Study and the American Cancer Society Study of Particulate Air Pollution and Mortality: A special report of the Institute's Particle Reanalysis Project. Cambridge, MA: Health Effects Institute; 2000. Available at: Accessed 10 January 2005].
36. Nyberg F, Gustavsson P, Jarup L, et al. Urban air pollution and lung cancer in Stockholm. Epidemiology. 2000;11:487–495.
37. Bellander T, Berglind N, Gustavsson P, et al. Using geographic information systems to assess individual historical exposure to air pollution from traffic and house heating in Stockholm. Environ Health Perspect. 2001;109:633–639.
38. Reuterwall C, Hallqvist J, Ahlbom A, et al. Higher relative, but lower absolute risks of myocardial infarction in women than in men: analysis of some major risk factors in the SHEEP study. The SHEEP Study Group. J Intern Med. 1999;246:161–174.
39. Theorell T, Perski A, Akerstedt T, et al. Changes in job strain in relation to changes in physiological state. A longitudinal study. Scand J Work Environ Health. 1988;14:189–196.
40. Rosenlund M, Berglind N, Gustavsson A, et al. Environmental tobacco smoke and myocardial infarction among never-smokers in the Stockholm Heart Epidemiology Program (SHEEP). Epidemiology. 2001;12:558–564.
41. Kolegard Stjarne M, Diderichsen F, Reuterwall C, et al. Socioeconomic context in area of living and risk of myocardial infarction: results from Stockholm Heart Epidemiology Program (SHEEP). J Epidemiol Community Health. 2002;56:29–35.
42. Hammar N, Andersson T, Alfredsson L, et al. Association of boiled and filtered coffee with incidence of first nonfatal myocardial infarction: the SHEEP and the VHEEP study. J Intern Med. 2003;253:653–659.
43. Gustavsson P, Plato N, Hallqvist J, et al. A population-based case-referent study of myocardial infarction and occupational exposure to motor exhaust, other combustion products, organic solvents, lead, and dynamite. Stockholm Heart Epidemiology Program (SHEEP) Study Group. Epidemiology. 2001;12:222–228.
44. Fransson E, De Faire U, Ahlbom A, et al. The risk of acute myocardial infarction: interactions of types of physical activity. Epidemiology. 2004;15:573–582.
45. MapInfo Professional User's Guide. New York: MapInfo Corp; 1995.
46. Tätort 2000 [in Swedish]. Stockholm, Sweden: Kartcentrum, Lantmäteriverket (National Land Survey of Sweden); 1996.
47. Weinberg CR, Moledor ES, Umbach DM, et al. Imputation for exposure histories with gaps, under an excess relative risk model. Epidemiology. 1996;7:490–497.
48. Johansson C, Hadenius A, Johansson P, et al. The Stockholm Study on Health Effects of Air Pollution and their Economic Consequences. Part I: NO2 and Particulate Matter in Stockholm, Concentrations and Population Exposure. Stockholm Environment and Health Protection Agency, Air Quality and Noise Analysis; 1999.
49. Romelsjo A, Branting M, Hallqvist J, et al. Abstention, alcohol use and risk of myocardial infarction in men and women taking account of social support and working conditions: the SHEEP case–control study. Addiction. 2003;98:1453–1462.
50. Pope CA 3rd, Burnett RT, Thun MJ, et al. Lung cancer, cardiopulmonary mortality, and long-term exposure to fine particulate air pollution. JAMA. 2002;287:1132–1141.
51. Abbey DE, Mills PK, Petersen FF, et al. Long-term ambient concentrations of total suspended particulates and oxidants as related to incidence of chronic disease in California Seventh-Day Adventists. Environ Health Perspect. 1991;94:43–50.
52. SLB-analys. Luften i Stockholm. Årsrapport 2003. [The air in Stockholm. Annual report 2003 (in Swedish).] SLB-analys, Miljöförvaltningen i Stockholm: SLB; 2003.
53. Kunzli N, Jerrett M, Mack WJ, et al. Ambient air pollution and atherosclerosis in Los Angeles. Environ Health Perspect. 2005;113:201–206.
No Caption Available.
© 2006 Lippincott Williams & Wilkins, Inc.