Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Traffic pollution and cardiovascular diseases in Greater Vancouver in association with socioeconomic… Lencar, Cornel Calin 2010

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2010_fall_lencar_cornel.pdf [ 2.07MB ]
Metadata
JSON: 24-1.0071316.json
JSON-LD: 24-1.0071316-ld.json
RDF/XML (Pretty): 24-1.0071316-rdf.xml
RDF/JSON: 24-1.0071316-rdf.json
Turtle: 24-1.0071316-turtle.txt
N-Triples: 24-1.0071316-rdf-ntriples.txt
Original Record: 24-1.0071316-source.json
Full Text
24-1.0071316-fulltext.txt
Citation
24-1.0071316.ris

Full Text

   Traffic pollution and cardiovascular diseases in Greater Vancouver in association with socioeconomic status indicators   by  Cornel Calin Lencar  M.F., The University of British Columbia, 2002    A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE   in  THE FACULTY OF GRADUATE STUDIES  (OCCUPATIONAL AND ENVIRONMENTAL HYGIENE)       THE UNIVERSITY OF BRITISH COLUMBIA  (Vancouver)     August 2010   © Cornel Calin Lencar, 2010  ii Abstract Cardiovascular diseases constitute a major health burden of modern societies. Besides eating habits, smoking or levels of physical activity, which are the most acknowledged risk factors, social determinants of health or air pollution constitute important risks in the development of cardiovascular diseases. Current knowledge about potential interactions between socioeconomic status and the effects of short- and long-term exposure to air pollution on mortality or morbidity due to cardiovascular diseases doesn’t offer substantive answers regarding the effect modification of various socio-economic factors on the risk of developing cardiovascular diseases due to air pollution. These interactions were analyzed using a cohort of 346,536 subjects over 45 years of age from Greater Vancouver Area, British Columbia.  My study found significant evidence that even in areas with low levels of traffic pollution and even for healthy people, there is an increased risk of cardiovascular disease morbidity or mortality associated with exposure to traffic pollution and road proximity, especially when considering socioeconomic variables at medium-scale (neighborhood) levels of aggregation. However, I found consistent results regarding the extent to which socioeconomic indicators modify the effect of traffic pollution on health on the expected trajectory (individuals from more advantaged areas would be less subjected to the effects of traffic pollution compared with individuals living in more disadvantaged areas). At dissemination area levels, in the case of exposure to particulate matter, subjects living in areas with a higher percentage of Chinese population were at a lower risk of CCS health outcomes. Subjects living in areas with a higher proportion of university degrees were also at a lower risk of experiencing CCS in conjunction with black carbon exposure. These results would need to be studied in conjunction with analyses at individual level data to confirm that these trends are real.  Also, the results do not prove that socioeconomic covariates derived for smaller areas strengthen the association or show significance in respect with cardiovascular health outcomes and pollution, since higher levels of risk were found when using covariates at neighborhood levels of aggregation as oppose to those at dissemination area level.   iii Table of  contents Abstract ................................................................................................................................................. ii Table of contents ................................................................................................................................ iii List of tables.......................................................................................................................................... v List of figures......................................................................................................................................vii Abbreviations.....................................................................................................................................viii Introduction.......................................................................................................................................... 1 Literature review................................................................................................................................... 3 Air pollution and CVD................................................................................................................... 3 Acute effects of air pollution on health ................................................................................... 3 Chronic exposure to air pollution and health ......................................................................... 6 Traffic air pollution studies........................................................................................................8 Mechanisms of action...............................................................................................................12 Socioeconomic indicators and cardiovascular diseases............................................................12 The effect of socioeconomic status on the relationship between atmospheric pollution and health..................................................................................................................15 Summary of literature review.......................................................................................................17 Objectives............................................................................................................................................18 Hypotheses..........................................................................................................................................18 Methods...............................................................................................................................................19 Study population............................................................................................................................19 Residential history..........................................................................................................................20 Health outcomes............................................................................................................................22 Individual and small-scale socioeconomic covariates ..............................................................23 Medium scale socioeconomic covariates....................................................................................25 Air pollution exposure assessment..............................................................................................27 Land use regression ..................................................................................................................27 Proximity to roads.....................................................................................................................28 Statistical analysis ...........................................................................................................................28 Results..................................................................................................................................................30 Validation of the administrative database with the Canadian Community Health Survey information...........................................................................................................30 CCHS data summary ................................................................................................................30 Smoking status in the CCHS sub-cohort...............................................................................32 Analysis of the relationship between smoking status and health outcomes.....................32 Correlations between individual CCHS variables and census SES and pollution related variables .........................................................................................................................34 Cohort summary statistics ............................................................................................................35 Exposure related summaries ........................................................................................................37 Traffic exposure ........................................................................................................................37 Road proximity..........................................................................................................................38 Cox analysis ....................................................................................................................................40 Stratified Cox analyses for the low and high levels of various SES covariates.....................42 Traffic related health outcomes ..............................................................................................43 Road proximity related health outcomes...............................................................................50 Discussion ...........................................................................................................................................57  iv Traffic pollution related results ...................................................................................................57 Road proximity related results .....................................................................................................61 SES related analyses ......................................................................................................................63 Racial/Cultural indicators ........................................................................................................65 Income and wealth variables ...................................................................................................67 Education ...................................................................................................................................69 Labor and occupation...............................................................................................................69 Transportation...........................................................................................................................71 Strengths and weaknesses of the study.......................................................................................71 Conclusions.........................................................................................................................................72 References ...........................................................................................................................................75 Appendix – Traffic pollution, road proximity and ACS and CHF outcomes...........................93 Appendix – Traffic pollution, road proximity and ACS and CHF outcomes in relation with low/high levels of SES......................................................................................................................95 Appendix I ........................................................................................................................................107 Appendix II: Dissemination area level covariates in conjunction with traffic related pollutants ...........................................................................................................................................123 ACS health outcomes..................................................................................................................123 CCS health outcomes..................................................................................................................126 CHF health outcomes.................................................................................................................129 Appendix III: Neighborhood area level covariates in conjunction with traffic related pollutants ...........................................................................................................................................133 ACS health outcomes..................................................................................................................133 CCS health outcomes..................................................................................................................136 CHF health outcomes.................................................................................................................139 Appendix IV: Dissemination area level covariates in conjunction with road proximity.......143 ACS health outcomes..................................................................................................................143 CCS health outcomes..................................................................................................................144 CHF health outcomes.................................................................................................................145 Appendix V: Neighborhood area level covariates in conjunction with road proximity........147 ACS health outcomes..................................................................................................................147 CCS health outcomes..................................................................................................................148 CHF health outcomes.................................................................................................................149   v List of  tables Table 1. Short term analyses of pollutants causing adverse cardiovascular health effects ........ 4 Table 2. Hospital admissions for cardiovascular disease................................................................ 5 Table 3. Neighborhood based studies of SES and cardiovascular health outcomes ...............13 Table 4. SES variable to be used in statistical analyses.................................................................26 Table 5. Comparison between the sampling weights of subjects from GVRDplus area and subjects outside GVRDplus......................................................................................................30 Table 6. CCHS health outcomes vs. Administrative health data ................................................32 Table 7. Summary statistics regarding the number of pack years ...............................................32 Table 8. Estimates of odds ratio of having a self reported/administratively derived health outcome for subjects that ever/never smoked..................................................................33 Table 9. Trend analysis of having a self reported health outcome for subjects with different levels of smoking ...............................................................................................................33 Table 10. Trend analysis of having an administratively derived health outcome for subjects with different levels of smoking .......................................................................................34 Table 11. Age, sex and health outcome summaries ......................................................................35 Table 12. Summary statistics for the DA-SES variables form Stats Canada 2001 Census used in the Cox model.........................................................................................................36 Table 13. Summary statistics for the Neighbourhood SES variables from the B.C. Atlas of Child Development in the Cox Model.............................................................................37 Table 14. Traffic pollutants: summary statistics ............................................................................37 Table 15. Correlations between traffic pollutants .........................................................................38 Table 16. Percentage of subjects in the CVD cohort living in the proximity to roads............38 Table 17. Traffic exposure and relative risk for CCS health outcomes .....................................38 Table 18. Road proximity* and relative risk for CCS health outcomes.....................................39 Table 19. Crude hazard ratios for traffic pollutants in relation with CCS outcomes...............40 Table 20. Hazard ratios for traffic pollutants adjusted for DA and neighborhood levels SES covariates in relation with CCS outcomes...................................................................41 Table 21. Crude and adjusted hazard ratios for road proximity adjustment done using SES covariates at different levels of aggregation……………………………………………..…42 Table 22. Comparison between traffic pollution HR for low and high levels of DA-SES variables, when considering CCS health outcomes ......................................................44 Table 23. Comparison between traffic pollution HR for low and high levels of Neighborhood-SES variables, when considering CCS health outcomes...................................48 Table 24. Comparison between HR of different road proximity categories for low and high levels of DA-SES variables, when considering health outcomes ...............................51 Table 25. Comparison between HR of different road proximity categories for low and high levels of Neighborhood-SES variables, when considering health outcomes............54 Table 26. Traffic exposure and relative risk for CCS health outcomes .....................................93 Table 27. Road proximity* and relative risk for CVD health outcomes....................................93 Table 28. Crude hazard ratios for traffic pollutants......................................................................94 Table 29. Hazard ratios for traffic pollutants adjusted for DA and neighborhood levels SES covariates.....................................................................................................................................94 Table 30. Crude and adjusted hazard ratios for road proximity adjustment done using SES covariates at different levels of aggregation...........................................................................94  vi Table 31. Comparison between traffic pollution HR for low and high levels of DA-SES variables, when considering ACS health outcomes.......................................................................95 Table 32. Comparison between traffic pollution HR for low and high levels of Neighborhood-SES variables, when considering ACS health outcomes...................................97 Table 33. Comparison between traffic pollution HR for low and high levels of DA-SES variables, when considering CHF health outcomes......................................................................99 Table 34. Comparison between traffic pollution HR for low and high levels of Neighborhood-SES variables, when considering CHF health outcomes................................101 Table 35. Comparison between HR of different road proximity categories for low and high levels of DA-SES variables, when considering ACS and CHF health outcomes..........103 Table 36. Comparison between HR of different road proximity categories for low and high levels of Neighborhood-SES variables, when considering ACS and CHF health outcomes ...............................................................................................................................105 Table 37. Distribution of health outcomes by different covariate categories (DA-SES) for black carbon traffic pollution...................................................................................................107 Table 38. Distribution of health outcomes by different covariate categories (DA-SES) for NO traffic pollution ..................................................................................................................108 Table 39. Distribution of health outcomes by different covariate categories (DA-SES) for NO2 traffic pollution.................................................................................................................109 Table 40. Distribution of health outcomes by different covariate categories (DA-SES) for PM2.5 traffic pollution................................................................................................................110 Table 41. Distribution of health outcomes by different covariate categories (DA-SES) for road proximity............................................................................................................................111 Table 42. Distribution of health outcomes by different covariate categories (Neighborhood-SES) for black carbon traffic pollution............................................................112 Table 43. Distribution of health outcomes by different covariate categories (Neighbourhood-SES) for NO traffic pollution .........................................................................113 Table 44. Distribution of health outcomes by different covariate categories (Neighborhood-SES) for NO2 traffic pollution ..........................................................................114 Table 45. Distribution of health outcomes by different covariate categories (Neighborhood-SES) for PM2.5 traffic pollution..........................................................................115 Table 46. Distribution of health outcomes by different covariate categories (Neighborhood-SES) for different road proximity categories ..................................................116 Table 47. Distribution of subjects by the two levels of socioeconomic indicators at DA-level of aggregation and the four quartiles of traffic pollutants ........................................117 Table 48. Distribution of subjects by the two levels of socioeconomic indicators at neighborhood level of aggregation and the four quartiles of traffic pollutants ......................119 Table 49. Distribution of subjects by the two levels of socioeconomic indicators at DA-level of aggregation and the five road proximity categories ..............................................121 Table 50. Distribution of subjects by the two levels of socioeconomic indicators at neighborhood-level of aggregation and the five road proximity categories ............................122   vii List of  figures Figure 1. Mechanisms by which particulate and gaseous air pollutants may exert adverse effects on the cardiovascular system.................................................................................14 Figure 2. Time periods and data availability ...................................................................................23 Figure 3. Dissemination Areas, Neighbourhoods and forward Sortation Areas in Vancouver.......................................................................................................................................27 Figure 4. Fictitious example of dose-response relationship in low and high SES populations .................................................................................................................................29 Figure 5. Adjusted hazard ratios for traffic pollutants and CCS in conjunction with neighborhood level SES....................................................................................................................41  viii Abbreviations  µg/m3  micrograms per cubic meter of air µm  micrometer 95% CI 95 percent confidence interval ACS   acute coronary syndrome AMI  acute myocardial infarction BC   British Columbia CAC   coronary artery calcification CCHS   Canadian Community Health Survey CCS   chronic coronary syndrome CHF   congestive heart failure CI  confidence intervals CO   carbon monoxide COPD  chronic obstructive pulmonary diseases CVD  cardiovascular diseases D  diabetes mellitus DA  (census) dissemination area EA  (census) enumeration area FSA Forward Sortation Area (first three digits of the six digit Canadian postal codes) GAB Georgia Air Basin GIS  geographic information system HD  hypertensive diseases  ix ICD-9   International Classification of Diseases, 9th Edition ICD-10  International Classification of Diseases, 10th Edition IDW   inverse distance weighted IHD  ischemic heart disease LUR   land use regression NO  nitric oxide NO2  nitrogen dioxide O3  ozone OC  organic components OR  odds ratio PM  particulate matter PM10  particulate matter smaller than 10 micrometers in aerodynamic diameter PM2.5  particulate matter smaller than 2.5 micrometers in aerodynamic diameter ppb  parts per billion RR  relative risk SE  standard error SES  socioeconomic status SO2  sulfur dioxide UFPs   ultra fine particles US EPA United States Environmental Protection Agency  1 Introduction Cardiovascular diseases comprise a group of diseases in which both genetic and environmental factors play a causal role. Given this combination of factors, it comes as no surprise that the American Heart Association’s (AHA) “Guidelines for Primary Prevention of Cardiovascular Disease and Stroke”1 and the AHA/American College of Cardiology’s (ACC) “Guidelines for Preventing Heart Attack and Death in Patients with Atherosclerotic Cardiovascular Disease”2 emphasize the need for multi-factorial interventions in preventing CVD. They recommend intensive measures to reduce an individual’s risk of cardiovascular disease, focusing on diet, drugs, exercise, weight management, complete smoking cessation, and the avoidance of second hand smoke.  The multi-factorial view on the causes of CVD has been emphasized even more by the research done in the last couple of decades regarding the potential deleterious effects of ambient air pollution on health and its relationship to heart disease and stroke. Of special interest are several environmental air pollutants that include carbon monoxide, oxides of nitrogen, sulphur dioxide, ozone, lead, and particulate matter (“thoracic particles” [PM10] <10 μm in aerodynamic diameter, “fine particles” [PM2.5] <2.5 μm, and “coarse particles” [PM10 to PM2.5]). Among these pollutants, particulate matter have received the most attention, a recent review by the American Heart Association stating that “Exposure to PM <2.5 μm in diameter (PM2.5) over a few hours to weeks can trigger cardiovascular disease–related mortality and nonfatal events; longer-term exposure (e.g., a few years) increases the risk for cardiovascular mortality to an even greater extent than exposures over a few days and reduces life expectancy within more highly exposed segments of the population by several months to a few years; reductions in PM levels are associated with decreases in cardiovascular mortality within a time frame as short as a few years; and many credible pathological mechanisms have been elucidated that lend biological plausibility to these findings. It is the opinion of the writing group that the overall evidence is consistent with a causal relationship between PM2.5 exposure and cardiovascular morbidity and mortality. This body of evidence has grown and been strengthened substantially since the first American Heart Association scientific statement was published. Finally, PM2.5 exposure is deemed a modifiable factor that contributes to cardiovascular morbidity and mortality.”2a  US EPA Air Quality Criteria for Particulate Matter has confirmed the presence of an apparent linear dose-response relationship between PM and adverse events3. This dose- response curve, derived from data gathered from across North America, has no discernible threshold below which PM concentrations pose no health risk to the general population3. Approximately 40,000 deaths per year in Austria, France, and Switzerland combined have been attributed to PM4. Estimates based on time-series studies suggest that there are circa 5,000 excess deaths per year in Canada5 and 6,000 cardiovascular events in the United Kingdom6 that can be attributed to poor air quality. A study in London, England found that approximately 1 in 50 myocardial infarctions may be triggered by outdoor air pollution.7 On a global scale, the World Health Organization has estimated that PM exposure is responsible for 800,000 deaths and the loss of 7.9 million disability-adjusted life-years annually, among which 89%, or over 712,000 deaths attributable to cardiopulmonary diseases.8   2 Besides the interest in environmental air pollution as a risk factor in the development of CVD, research was also directed at studying the impact of various social determinants of health as risk factors for CVD development. In doing so, most of epidemiological studies have used census data as proxies for individual level information on the socioeconomic status of the study subjects. Although, from a mechanistic perspective, it is difficult to relate the characteristics of whole neighbourhoods with an individual’s propensity to develop cardiovascular diseases, there are convergent studies that relate CVD with stress,9 as well as studies that relate low socioeconomic status with stress and health.10 Moreover, from an epidemiological perspective, there is an emphasis on the impact of neighbourhoods on individuals’ health.11  While the causal link between socioeconomic status and mortality and morbidity is not fully explained, the existence of an inverse gradient between SES and mortality and morbidity has been consistently observed. This inverse relationship is observed whether SES is measured using education, income, or occupational status, and it does not appear to be an artifact of the more physically ill individuals drifting down the SES hierarchy.106 The SES-health gradient extends to a wide array of health problems, including heart disease, cancer, stroke, diabetes, hypertension, infant mortality, arthritis, back ailments, mental illness, kidney diseases, and many others,107 and may predict future developments after illness is present.108,109 For more detailed information on specific studies on SES and health, there are several excellent reviews available.110-122 Not only have various health problems been found to have an inverse correlation with SES, but the prevalence of numerous risk factors associated with these diseases also tends to be inversely associated with SES. A recent review by Laurent et al237 that investigated the potential interactions between socioeconomic status and the short- and long-term effects of air pollution on mortality found that studies using socioeconomic characteristics measured at coarser geographic resolutions (city- or county- wide) found no effect modification, but those using finer geographic resolutions found mixed results, and five of six studies using individually measured socioeconomic characteristics found that pollution affected disadvantaged subjects more.  There are relatively few studies that have examined the contribution of environmental exposure, such as air pollution, to observed socioeconomic health inequalities. The two mechanisms through which air pollution might create or enhance socioeconomic differences that are identified for various diseases are: (1) populations with low SES may be subjected to more frequent and/or intense exposures to air pollution (environmental inequality); (2) populations with low SES may have an increased susceptibility to air pollution as compared to populations with high SES (biological inequality).  Few studies to date have tried to bring together and analyze the relationship between air pollution and CVD by considering not only an individual’s socioeconomic status, but also the characteristics of the subject’s environment.12,13 Of course, one problem can be seen from the start. This is the ecologic fallacy, in which the ambient pollution levels, or, depending on the case, neighbourhood characteristics, are assigned to an individual.14,15 By using high-resolution air pollution exposure models and subjects’ actual addresses, part of the first problem can be mitigated. However, not many of the variables available at a neighbourhood level are available at an individual level, and the more problematic variables are those inherently individual, like smoking.  3 Literature review Air pollution and CVD The majority of epidemiological studies to date have found an increase in the relative risk of CVD due to air pollution. However, for an individual, the increase in relative risk for CVD due to air pollution is small (see page 15 for risk estimates) compared to the impact of other well-established cardiovascular risk factors. Although this risk appears minor, given the fact that people must breathe to live, conservative risk estimates yield a substantial increase in mortality within the population when air pollution is factored in.  The epidemiology of ambient air pollution is well documented; the association between high levels of air pollutants and human illness has been known for more than half a century. There are several hundred published epidemiological studies that link air pollution with human illnesses and a number of extensive reviews on this topic are also available.6,16-27 During the past 15 years, the magnitude of evidence and the number of studies linking air pollution to cardiovascular diseases has grown substantially.6,16-27  Air pollutants are comprised of an extensive variety of substances. However, the most investigated ones are particulate matter, carbon monoxide, sulphur dioxide, nitrogen oxide and nitrogen dioxide, and ozone. Although many pollutants may cause disease individually or in combination (e.g., O3, SO2, and NO2), 28 over the past decade, PM has become a major focus of research. The particles believed to be most deleterious to health are those with an aerodynamic diameter of less than 10 μm (PM10), but most recently the focus has been on the fine and ultra fine component (PM2.5 and PM0.1), which can penetrate deep into the gas exchange region of human lungs. Acute effects of air pollution on health There is a wealth of time series studies and case cross-over studies that link short-term exposure to various pollutants (PM-PM10, PM2.5, ultrafine particles; ozone, SO2, NO2, CO) with adverse cardiovascular health effects (Table 1).  Some of these studies controlled for certain cyclical variables that could possibly influence mortality, like temperature, presence of influenza outbreaks, etc. while the analyses were sometimes stratified by age. The age factor (usually a cut-off value is used for subjects over and under 65 years) does not play a consistent role in the sense that, for some pollutants, significant adverse effects are found for subjects over 65 years, while for other pollutants, significant effects are found for subjects under 65 years.47,48 Temperature was not found to affect the estimates for pollutants60, while smoking status was a major modifier confounder in those studies that used this information. Co-morbidities such as hypertension, diabetes, COPD, and pre-existing cardiac diseases were found to increase the risk of an adverse cardiovascular event by up to 2-3 times.59 A Vancouver, Canada based study 12 that employed socioeconomic indicators found that, for NO2, CO, and SO2, the estimated percent change in daily cardiovascular mortality was more pronounced among those in the low and middle categories of SES. Similar conclusions can be drawn from data on hospital admissions for cardiovascular disease (Table 2).66-78  4 Table 1. Short term analyses of pollutants causing adverse cardiovascular health effects Study Geographic area Cohort size Type of study Pollutants Results Wichmann 198929 West Germany 1985 Population based study -12 million (high vs. low exposed areas) Natural study SO2 TSP 8% ↑ mortality during smog 6% ↑ CVS mortality (15% q admissions) Katsouyanni 199031 Athens 1975–1982 25,138 deaths 3 x 199days Case/control 1:2 SO2, Black smoke Higher respiratory and CVD mortality on polluted days Schwartz 199030 London 1958–1972 Total, Respiratory and cardiovascular deaths in London btw. 1958-1972 time series, autoregressive analysis Black smoke, SO2 Significant predictors of all cause daily mortality Kinney 199132 Los Angeles County 1970–1979 Total, respiratory and cardiovascular deaths time series, autoregressive analysis Particles, Ox CO, SO2 NO2 Strongly associated with daily CVS mortality Schwartz 199233 Philadelphia 1973–1980 Total, respiratory and cardiovascular deaths time series, logistic analysis SO2 TSP 5% ↑ mortality/100 μg ↑ 7% ↑ mortality/100 μg q,10% ↑ CVS mortality Schwartz 199434 Philadelphia 1973–1980 Deaths in 5% high (7,915) and 5% low (7,337)pollution days time series, logistic analysis TSP  RR death = 1.08 on high v low pollution day, ↑ heart disease and stroke deaths Schwartz 199435 Meta-analysis 13 Studies Total deaths  TSP  RR death = 1.06 for 100 μg ↑ Anderson 199636 London 1987–1992 Respiratory and cardiovascular deaths time series – Poisson analysis Black smoke SO2 Ozone 2.5% ↑ daily mortality/7–19 μg/m3 ↑ SO2 also significant 3.6% ↑ CVS mortality/7–36 PB ↑ Ponka 199837 Helsinki 1987–1993 Cardiovascular deaths time series PM10 Ozone NO2 4.1% ↑ CVS mortality/10 μg/m3 ↑ 9.9% ↑ CVS mortality/20 μg/m3 ↑ Additive effect with PM10 and ozone Zmirou 199838 10 European cities Respiratory and cardiovascular deaths time series Black smoke SO2 RR CVS mortality 1.02/50 μg/m3 ↑ RR CVS mortality 1.04/50 μg/m3 ↑ Ostro 199939  Bangkok 1992–1995 Total, respiratory and cardiovascular deaths time series PM10  2% ↑ CVS mortality/10 μg/m3 ↑ Rossi 199940  Milan 1980–1989 Respiratory and cardiovascular deaths time series TSP  7% ↑ heart failure deaths/100 μg/m3 ↑, 10% ↑ myocardial infarction mortality/100 mg/m3 ↑ Checkoway 200042 King County, WA 1988-1994 362 SCA cases case-crossover analysis PM10 RR SCA 0.893/ IQR ↑ PM10 lag1 Peters 200043 Massachusetts 1995-1997 100 subjects with defibrillators – prospective cohort logistic analysis NO2 OR 1.8 (1.1 – 2.9) increased defibrillator interventions lag2 / 26 ppb ↑ Samet 200041 20 US cities 1987–1994 Respiratory and cardiovascular deaths time series PM10 SO2 ,CO, Ozone, NO2 ↑ Rate of CVS/respiratory mortality, 0.68% for each ↑ PM10 of 10 μg/m3 Weak associations Braga 200149 10 US Cities 1986 – 1993 Deaths due to COPD, CVD, MI in 10 U.S. cities between 1986-1993; Time series - GAM PM10 1% ↑ total cardiovascular disease deaths/ 10 μg/m3 ↑ in 7 days mean; 0.7% ↑myocardial infarction deaths/ 10 μg/m3 ↑ in 2 days mean; Goldberg 200147 Montreal 1984- 1993 All deaths btw 1984-1993 Time series - GAM ozone 2.5% ↑ total cardiovascular deaths/IQR ↑ Goldberg 200148 Montreal 1986- 1993 All deaths btw 1984-1993 Time series - GAM PM2.5, COH, Sutton sulfate sulfate mass Consistent association with coronary artery disease and cardiovascular disease deaths for subjects > 65 years Katsouyanni 200145 29 European cities (APHEA 2) 43 million population; all deaths Time series - GAM PM10  ↑ Rate of CVS/respiratory mortality 0.6% for each ↑ PM10 of 10 μg/m3, effect size greater in elderly, with high NO2, or in cold climates Kwon 200144 Seoul 1994–1998 1,807 deaths  case-crossover and GAM (time series) PM10,SO2, CO, Ozone NO2 RR mortality 1.014/IQR ↑ PM10 RR mortality 1.020/IQR ↑ CO Effect 2.5–4.1% higher in CCF Peters 200146 Boston 1995-1996 772 patients with MI case-crossover analysis PM2.5 OR 1.48 MI 25 μg/m3 ↑ 2 h prior MI onset OR 1.69 MI 20 μg/m3 ↑ 24 h prior MI onset Hong 200250 Seoul 1991-1997 Stroke mortality in Seoul between 1991-1997; Time series - GAM TSP SO2 NO2 CO Ozone RR ischemic stroke mortality 1.03/IQR ↑ same day RR ischemic stroke mortality 1.04/IQR ↑ same day RR ischemic stroke mortality 1.04/IQR ↑ 1-day lag RR ischemic stroke mortality 1.06/IQR ↑ 1-day lag RR ischemic stroke mortality 1.06/IQR ↑ 3-day lag D’Ippoliti 200351 Rome 1995-1997 6531 with AMI case-crossover TSP OR 1.028 AMI//10 μg/m3 ↑ 0 to 2 day lag Dominici 200353 88 metropolitan areas in US 1987-1994 NMMPS database: 88 cities between 1987- 1994 2 stage Bayes hierarchical log-linear regression of daily mortality PM10 0.31% ↑ cardiovascular & respiratory mortality/ 10 μg/m3 ↑ Sullivan 200352 Washington State, US 1985-1994 1206 with cardiac arrest case-crossover PM2.5 OR 0.94 out-of-hospital cardiac arrests/13.8 μg/m3 ↑ 0 to 2 days lag APHEA, Air Pollution and Health: a European approach; NMMPS, National Morbidity Mortality Air Pollution Study; CCF, congestive cardiac failure; CVS, cardiovascular system; IQR, inter-quartile range; RR; relative risk; TSP, total suspended particles; COH, coefficient of haze; SCA, sudden cardiac arrest; MI, myocardial infarction; GAM, general additive models; ↑ (increase)   5 Table 1. Short term analyses of pollutants causing adverse cardiovascular health effects Study Geographic area Cohort size Type of study Pollutants Results Villeneuve 200312 Vancouver 1986-1999 550,000 cohort time series PM10-2.5 5.9% ↑ cardiovascular mortality/diff btw 10-90th percentile Bateson 200454 Cook County, Illinois, US 1988-1991 65,180 subjects with heart or lung disease case-crossover PM10 0.74% ↑ mortality for subjects with cardiovascular & respiratory co-morbidities/ 10 μg/m3 ↑ Bell 200457 US  39 studies + NMMAPS Meta-analysis Ozone 0.83% ↑ total mortality / 10 ppb ↑ Daniels 200456 US  20 largest cities in US (NMMAPS) different time series analyses PM10  No threshold for cardiovascular-respiratory mortality for a 10 μg/m3 increase (↑) in PM10 Schwartz 200458 14 US cities 1986-1993 Non-accidental deaths in 14 US cities case-crossover analysis PM10 0.36% ↑ deaths from internal causes/ 10 μg/m3 ↑ Forastiere 200559 Rome 1998-2000 5144 non hospital cardiac deaths; case-crossover analysis Ultrafine particles 7.6% ↑ out-of-hospital coronary deaths/IQR ↑ with effect modification for subjects with hypertension and COPD Ruidavets 200561 Toulouse 1997-1999 Cohort of ~ 1.1 mil case-crossover Ozone RR 1.05 acute myocardial infarction / 5 μg/m3 ↑ Schwartz 200560 14 US cities 1986-1993 Non-accidental deaths in 14 US cities case-crossover analysis Ozone 0.23% ↑ deaths from internal causes/ 10 ppb ↑ Zeka 200562 US 20 Cities 1989-2000 Mortality in 20 US cities between 1989-2000 case-crossover PM10  0.30% ↑ deaths heart diseases/ 10 μg/m3 ↑ lag1 0.37% ↑ deaths heart diseases/ 10 μg/m3 ↑ lag2 Analitis 200665 29 European cities APHEA 2 43 million population; all deaths; time series PM10 Black smoke 0.76% ↑ cardiovascular deaths/ 10 μg/m3 ↑ lag0-1 0.762% ↑ cardiovascular deaths/ 10 μg/m3 ↑ lag0-1 Martins 200664 Sao Paolo 1996-2001 people over 64 years in São Paulo 1996 to 2001; GAM – time series PM10 SO2 3.17% ↑ congestive heart failure/IQR ↑ 0.89% ↑ total cardiovascular disease/IQR ↑ Murakami 200663 Tokyo 1990-1994 14,950 deaths;  retrospective analysis TSP ↑ RR of Myocardial Infarction deaths within hours after reaching high concentration of TSP APHEA, Air Pollution and Health: a European approach; NMMPS, National Morbidity Mortality Air Pollution Study; CCF, congestive cardiac failure; CVS, cardiovascular system; IQR, inter-quartile range; RR; relative risk; TSP, total suspended particles; COH, coefficient of haze; SCA, sudden cardiac arrest; MI, myocardial infarction; GAM, general additive models; ↑ (increase)  Table 2. Hospital admissions for cardiovascular disease Study Geographic area Cohort size Type of study Pollutants Results Schwartz 199566 Michigan 1986–1989 520,000 (over 65 years) time series Poisson regression PM10, CO  ↑ Ischemic heart disease admissions (RR 1.018 IQR ↑ PM10) and with heart failure (RR 1.024 /IQR ↑ PM10 and 1.022/ IQR ↑ CO) Burnett 199567 Ontario 1983–1988 The Toronto-Hamilton corridor time series analyses Particulate sulphates 2.8% ↑ CVS admission/13 μg/m3 ↑ Morris 199568 7 US cities 1986–1989 Hospital admissions >=65 years and older time series analysis; CO  ↑ Heart failure admissions (RR 1.10–1.37/10 ppm ↑) Wordley 199769 Birmingham 1992–1994 Hospital admissions time series on an ecological retrospective cohort PM10  ↑ Risk of respiratory (2.4%) or cerebrovascular (2.1%) admission for 10 μg/m3 ↑ Schwartz 199770 Tucson 1997  Subjects >= 65yeras time series analysis PM10 CO Ozone/SO2 2.75% ↑ CVS admission/IQR ↑ 2.79% ↑ CVS admission/IQR ↑ Little association Burnett 199771 10 Canadian cities 1981–1991 Congestive heart failures in subjects >=65 time series analysis CO  RR heart failure admission 1.065/IQR ↑ Burnett 199772 Toronto 1970– 1994 Mortality in Toronto between 1970-1990 time series analyses Ozone, NO2, SO2 13% ↑ CVS admissions/IQR ↑ gaseous pollutants Poloniecki 199773 London 1987– 1994 373,556 hospital admissions time series analysis Black smoke  NO2 2.5% myocardial infarction admissions attributable, associated with angina admissions Associated with arrhythmia admissions Schwartz 199974 Eight US counties 1988– 1990 Cardiovascular hospital admissions time series analysis PM10 CO 2.48% ↑ CVS admission/IQR ↑ 2.79% ↑ CVS admission/IQR ↑ Wong 200275 Hong Kong (1995-97) London (1992-94) Daily cardiac admissions  time series analyses NO2 PM10 SO2 RR 0.6/0.7 cardiac admissions / 10 μg/m3 ↑ 0-1 day lag RR 0.5/0.9 cardiac admissions / 10 μg/m3 ↑ 0-1 day lag RR 0.1/1.7 cardiac admissions / 10 μg/m3 ↑ 0-1 day lag Jordi 200376 7 European areas (APHEA-2) Hospital admissions for CVD time series analyses SO2 PM10 0.7% ↑ CVS admission/10 μg/m3 ↑ subjects < 65 years 1.3% ↑ CVS admission/10 μg/m3 ↑ subjects > 65 years Zanobetti 200577 21 US Cities 1985-1999 > 300,000 MI  case-crossover analysis PM10  0.65% ↑ MI admission/10 μg/m3 ↑ with effect modification for subjects with pneumonia or COPD Symons 200678 Baltimore 2002 125 subjects (135 cases)  case-crossover analyses PM2.5 OR 1.09 of congestive heart failure admissions/IQR ↑ APHEA, Air Pollution and Health: a European approach; CCF, congestive cardiac failure; CVS, cardiovascular system; IQR, inter-quartile range; RR; relative risk; TSP, total suspended particles; COH, coefficient of haze; SCA, sudden cardiac arrest; MI, myocardial infarction; CVD, cardiovascular diseases; MI, myocardial infarction; ↑ (increase)  6 It can be seen from both tables 1 and 2 that the magnitude of CVD outcomes varies greatly, from less than 1% to 13% between studies. However, meta-analyses of the time series data suggest that an increase in mean 24-hour fine particulate pollution of 10 μg/m3 increases the relative risk for daily cardiovascular mortality by approximately 0.4% to 1.0%.2c The excess number of deaths due to air pollution reported by various studies3-8 was confirmed by several large scale interventions or natural experiments, as described in more detail below.  A study regarding Dublin’s ban on coal sales in 1990 showed that a reduction in black smoke concentration by 35.6 μg/m3 was associated with a 10.3% decrease in annual cardiovascular mortality.79 In a similar intervention study, a 50% reduction in sulphur dioxide concentrations following legal restrictions on fuel oil sulphur in Hong Kong was immediately followed by a 2.4% reduction in cardiovascular deaths.80 Chronic exposure to air pollution and health The first large, prospective cohort study that demonstrated the adverse health impact of long- term air pollution exposure was the Harvard Six Cities study82 This study showed that chronic exposure to air pollutants is independently related to cardiovascular mortality. The adjusted overall mortality rate ratio for the most-polluted city versus the least-polluted city was 1.26 (95% CI 1.08 to 1.47). Adjustment for a variety of individual-level risk factors that included tobacco smoking, gender, body mass index, educational attainment, occupational exposures, hypertension, and diabetes did not significantly alter the relationship. Cardiovascular deaths accounted for the largest single category of the increased mortality. Among air pollutants, elevations of PM2.5 and sulphates showed the strongest associations with disease. A follow-up study on the Harvard Six Cities initial project has found that improved overall mortality was associated with decreased mean PM2.5 (10 μg/m 3) between the initial study and the follow-up study (RR, 0.73).  Similar observations were reported by the first analysis of air pollution in relation to mortality in the ACS Cancer Prevention II study population.83 A follow-up of the original ACS cohort by Pope et al.,84,85 based on additional subject mortality and ambient pollutant data, has provided the largest study of the long-term health effects of air pollution. In a cohort of approximately 500,000 adults residing in all 50 U.S. states, chronic exposure to multiple air pollutants was linked to mortality statistics for a 16-year period. The ACS follow-up study increased the degree of control for confounding variables such as diet. The primary results showed that each 10 μg/m3 increase in annual PM2.5 mean concentration, based on a number of different averaging periods, was associated with increases in all-cause, cardiopulmonary, and lung cancer mortality of 4%, 6%, and 8%, respectively. The relationship between PM2.5 and adverse health effects was linear and without a discernible lower “safe” threshold. This corresponds with findings in other studies.3,56 Mortality was strongly associated with PM2.5, sulphate particles, and SO2. There also appeared to be an association between cardiopulmonary mortality and summertime O3, when based on mean summer O3 levels from 1982 to 1998. Educational level was a modifier of the risks estimated for PM-associated mortality. However, the increased risks were restricted only to those subjects with no more than a high school education. This suggests that some other unaccounted-for factors, such as intra-urban geographic location or socioeconomic status, may be important determinants of health risk.   7 A recent Norwegian study86 that followed a cohort of 16,209 men 40–49 years of age living in Oslo, Norway from 1974 to 1998 found that the adjusted risk ratio for dying was 1.08 [95% confidence interval (CI), 1.06–1.11] for a 10-μg/m3 increase in average exposure to nitrogen oxides (NOX) after controlling for a number of potential confounders. Corresponding adjusted risk ratios for dying from a respiratory disease other than lung cancer were 1.16 (95% CI, 1.06– 1.26); from lung cancer, 1.11 (95% CI, 1.03–1.19); from ischemic heart diseases, 1.08 (95% CI, 1.03–1.12); and from cerebrovascular diseases, 1.04 (95% CI, 0.94–1.15). A study87a from Los Angeles for a cohort consisting of 22,905 subjects from the American Cancer Society cohort for the period 1982–2000 (5,856 deaths) looked at the association between mortality due to different causes and ambient PM2.5 and O3. After controlling for 44 individual covariates, for PM2.5, ischemic heart disease mortality was elevated (in the range of 1.24 –1.6, depending on the model used). A recent US study88 on a cohort of 65,893 postmenopausal women without previous cardiovascular disease in 36 U.S. metropolitan areas from 1994 to 1998, with a median follow- up of 6 years, has also found a relation between PM2.5 and cardiovascular adverse events. Hazard ratios were adjusted for age, race or ethnic group, smoking status, educational level, household income, body-mass index, and presence or absence of diabetes, hypertension, or hypercholesterolemia. Each increase of 10 μg/m3 was associated with a 24% increase in the risk of a cardiovascular event (HR, 1.24; 95% CI, 1.09 to 1.41) and a 76% increase in the risk of death from cardiovascular disease (HR, 1.76; 95% CI, 1.25 to 2.47). For cardiovascular events, the between-city effect appeared to be smaller than the within-city effect. The risk of cerebrovascular events was also associated with increased levels of PM2.5 (HR, 1.35; 95% CI, 1.08 to 1.68).  A study focused on women88a between 50 and 59 years of age based in the Ruhr area of Germany has studied the long term relation between ambient air pollution (NO2 and PM10) and cardiopulmonary mortality. The cohort comprised 4,874 women followed in the 1980s and 1990s to monitor health status and migration. One-year and five-year average exposure levels were found to be associated with cardiopulmonary mortality. The adjusted relative risk for the one-year average NO2 exposure was 1.57 (1.23–2.00), while for the five-year NO2 average, the relative risk was 1.74 (1.29–2.33). The adjusted relative risk for the one-year average PM10 exposure was 1.34 (1.06–1.71), while for the five-year PM10 average, the relative risk was 1.59 (1.23–2.04). The adjustment was done considering smoking status, body mass index, education level of the subject and her partner, as well as existent co-morbidities.  The above studies focusing on women, which found an increase risk of cardiovascular events associated with ambient exposure to particulates or nitrogen oxides, are complemented by a study88c conducted in California that followed a cohort of 3,239 subjects for 22 years, between 1977 and 1989, to investigate the effect of long-term ambient particulate matter on the risk of fatal coronary heart disease. Monthly concentrations of ambient air pollutants (PM10, PM2.5, ozone, sulphur dioxide, nitrogen dioxide) were used in the analyses as exposure variables. All participants had information on environmental tobacco smoke and other personal sources of air pollution, and all subjects with prevalent CHD, stroke, or diabetes at baseline (1976) were excluded. The analyses were also controlled for a number of potential confounders, including lifestyle. In females, the relative risk for fatal CHD with each 10 μg/m3 increase in PM2.5 was  8 1.42 (95% CI, 1.06–1.90) in the single pollutant model and 2.00 (95% CI, 1.51–2.64) in the two-pollutant model with O3. Corresponding RRs for a 10 μg/m 3 increase in PM10-2.5 and PM10 were 1.62 and 1.45 respectively in all females, and 1.85 and 1.52 respectively in postmenopausal females. No associations were found in males. Thus, a positive association with fatal CHD was found with all three PM fractions in females but not in males. The risk estimates were strengthened when adjusting for gaseous pollutants, especially O3, and were highest for PM2.5. Traffic air pollution studies A European study4 that looked at the association between PM10 and hospitalizations has found an overall relative risk of 1.013 (CI 1.007-1.019) for cardiovascular hospital admissions for the assessments done in Austria, France and Switzerland. The study found that the traffic-related proportion of the total cases attributable to air pollution corresponded to the traffic-related fraction of PM10, amounting to 43% in Austria, 56% in France, and 53% in Switzerland.  A Dutch study88b, based on an ongoing cohort study on diet and cancer (NLCS-AIR study), with 120,852 subjects followed from 1987 to 1996, investigated the association between the exposure to black smoke, nitrogen dioxide, sulphur dioxide, particulate matter ≤ 2.5 μm (PM2.5), and various variables related to traffic and mortality from all causes and for specific causes, including cardiovascular diseases. Hazard ratios were adjusted for age, sex, smoking status, and several socioeconomic indicators at area level. Although there was an increased risk for all mortality cases, none of the ambient pollutants investigated or traffic-related variables were significantly greater than 1 for cardiovascular mortality.  A Dutch study89 looked at personal and home outdoor NO2 concentrations for 241 children from six different primary schools in the Netherlands. The study found that personal and outdoor NO2 concentrations differed significantly among children attending schools in areas with different degrees of urbanization (the difference among average classroom concentrations in the very urban and non-urban school was 12.2 μg/m3) and among children attending schools in areas close to highways with different traffic densities (an estimated annual difference of 8.2 μg/m3 (SE 1.8) in personal NO2 exposure between the school with the highest traffic density and the school with the lowest traffic density). For the children living near highways, personal and outdoor NO2 concentrations also significantly decreased with increasing distance of the home address to the highway. This study has shown that personal and outdoor NO2 concentrations are influenced significantly by the degree of urbanization of the city district and by the traffic density of and distance to a nearby highway.  The importance of within-city residential variations as a risk factor for mortality due to air pollution was confirmed by Hoek et al.90 It was found that the exposure to traffic-related air pollutants was more highly related to mortality than were citywide background pollution levels. Of different metrics used in the analysis, an indicator variable for living near a major road was most strongly associated with cardiopulmonary mortality (RR 1.95, 95% CI 1.09 to 3.52). This study suggests that an individual’s exposure to the toxic components of air pollution may vary as much within a single city as across different cities.  A different approach to looking at traffic exposure was taken by Peters et al.91 Theirs was a case–crossover study in which cases of myocardial infarction were identified with the use of  9 data from the Health Research Cooperative in the Region of Augsburg Myocardial Infarction Registry in Augsburg, southern Germany, for the period from February 1999 to July 2001. For the 691 subjects, an association was found between exposure to traffic and the onset of a myocardial infarction within one hour afterward (OR 2.92; CI 2.22 to 3.83). The time the subjects spent in cars, on public transportation, or on motorcycles or bicycles was consistently linked with an increase in the risk of myocardial infarction.  A Canadian study conducted by Finkelstein92 investigated the rate advancement periods associated with traffic pollution exposures. The mortality from all natural causes during 1992– 2001 was modeled in relation to lung function; body mass index; a diagnosis of chronic pulmonary disease, chronic ischemic heart disease, or diabetes mellitus; household income; and residence within 50 m of a major urban road or within 100 m of a highway. The study found that subjects living close to a major road had an increased risk of mortality (RR 1.18; CI 1.02- 1.38). The mortality rate advancement period associated with residence near a major road was 2.5 years (CI 0.2-4.8). The rate advancement period attributable to chronic ischemic heart disease was 3.1 years.  In most of the studies concerning the effect of air pollution on various health outcomes, the exposure is usually determined using only community average concentrations. This may lead to measurement error that lowers the estimates of the health burden attributable to poor air quality because, theoretically, classic exposure measurement error induced by central monitors may bias results toward the null. Jerrett et al.13 modeled the association between air pollution and mortality using small-area exposure measures in Los Angeles, California. A sub-cohort of 22,905 subjects extracted from the American Cancer Society cohort for the period 1982–2000 (5,856 deaths) was linked with pollution exposures interpolated from 23 fine particle (PM2.5) and 42 ozone (O3) fixed-site monitors. Proximity to expressways was tested as a measure of traffic pollution. The impact of traffic was assessed by assigning buffers that included zip code- area centroids within either 500 or 1000 meters of a freeway. All-cause mortality had a relative risk of 1.17 (CI 1.05–1.30) for an increase of 10 μg/m3 PM2.5 and a RR of 1.11 (CI 0.99 –1.25) with maximal control for both individual and contextual confounders. The RRs for mortality resulting from ischemic heart disease and lung cancer deaths were elevated, in the range of 1.24 –1.6, depending on the model used. However, in their models, distance to freeways did not have a significant impact, the RR for being under 500 m from a freeway was 0.90 (0.71-1.14) while the RR for being within 1000 m of a freeway was 1.05 (0.89-1.24).   In the study conducted by Gehring et al. (2006),88a already mentioned under chronic effects of air pollution, the authors also investigated the relation between proximity to roads as a proxy for traffic pollution and mortality due to various causes. For cardiopulmonary mortality, the adjusted relative risk of living under 50 m from roads versus living more than 50 m from roads was 1.70 (1.02–2.81).  The NLCS-AIR study88b mentioned earlier also investigated the relation between traffic-related air pollution and mortality due to various causes. Thus, traffic intensity on the nearest road was found to be independently associated with mortality. Relative risk for a 10-μg/m3 increase in black smoke concentrations (10-μg/m3 representing the difference between the 5th and 95th percentile) were 1.04 (0.95–1.13) for cardiovascular mortality. Results were similar for NO2 and PM2.5, but no associations were found for SO2.  10 A study93b conducted in the U.S. in greater Worcester, Massachusetts, consisted of 1,389 patients hospitalized with acute heart failure (HF) in 2000. These patients were followed for survival through December 2005. Daily traffic information for the roads found within 100 m and 300 m buffers of participants’ residences as well as the distance from their residences to major roadways and bus routes were used as proxies for residential exposure to traffic-related air pollution. Mortality risk for the exposure variables was assessed using Cox proportional hazards models adjusted for prognostic factors. The inter-quartile range increase in daily traffic within 100 m of the home was associated with a mortality hazard ratio (HR) of 1.15 (1.05– 1.25), whereas for traffic within 300 m this association was 1.09 (1.01–1.19). The mortality risk decreased with increasing distance to bus routes (HR = 0.88; 95% CI, 0.81–0.96) and was larger for those living within 100 m of a major roadway or 50 m of a bus route (1.30; 1.13– 1.49).  In the Worchester Heart Attack Study involving 5,049 subjects with AMI, Tonne et al.93d used cumulative traffic within 100 m of subjects’ residences and distance from major roadway as proxies for exposure to traffic-related air pollution in order to investigate the association of traffic pollution and occurrence of AMI. They estimated the relationship between exposure to traffic and occurrence of AMI using case-control logistic regression, with adjustment for age, sex, section of the study area, point sources emissions of particulate matter with aerodynamic diameter < 2.5 μm, area socioeconomic characteristics, and percentage of open space. The researchers found that an increase in cumulative traffic near the home was associated with a 4% increase in the odds of AMI per inter-quartile range (95% CI, 2–7%), whereas living near a major roadway was associated with a 5% increase in the odds of AMI per kilometre (95% CI, 3–6%).  In a follow-up of the Worchester Heart Attack Study based in Worchester, Massachusetts, Tonne et al.93c employed a case-control analysis of subjects diagnosed with AMI between 1995 and 2003 and controls taken from the same area. Traffic pollution represented by NO2 and PM2.5 was modelled using a semi-parametric latent variable regression model with samples collected at 36 locations in the area. The authors found that the inter-quartile range increase in modelled traffic particles was associated with a 10% (4% to 16%) increase in the odds of AMI. When accounting for spatial dependence at the census tract, but not block group, it was found that scale substantially attenuated this association. Although the results provide some support for an association between long-term exposure to traffic particles and risk of AMI, they were sensitive to the scale selected for the analysis of spatial dependence. The latent variable model captured variation in exposure, although on a relatively large spatial scale.  In the same study that looked at chronic ambient exposure to PM2.5 and ozone, Jerret et al. 87a investigated the association between mortality due to ischemic heart diseases and proximity to traffic. Thus, subjects living within 500 m from the freeway had a relative risk of dying of IHD of 0.90 (0.71–1.14), while subjects living within 1000 m from freeways had a relative risk of 0.92 (0.77–1.08).  A large study conducted by Rosenlund et al.87b in Rome, comprising all residents of Rome aged 35–84 years during the period 1998–2000, assessed the association between residential NO2 exposure due to traffic pollution (derived by a land-use regression model) and coronary events. The study focused on the 6,513 survivors of cardiac events that were followed for 4.0 –7.5 years for readmission or mortality, starting 28 days from the date of the first event. Relative  11 risks per 10 μg/m3 of NO2 exposure, adjusted for age, sex, and socioeconomic status, were calculated by Poisson regression and Cox regression. The relative risk for incidence in coronary events per 10 μg/m3 of NO2 was 1.03 (1.00 –1.07). Stronger associations were found for fatal cases (1.07; 1.02–1.12) and out-of-hospital deaths (1.08; 1.02–1.13). Using NO2 exposure at the time of the first event, there was no association between air pollution exposure and either subsequent hospital readmission or mortality among survivors of the first coronary event.  Besides considering cardiovascular mortality or morbidity in relation with ambient and traffic pollution, some researchers have focused on cardiovascular disease progression or indicators. Thus, a 2005 study87 from Los Angeles found that for a cross-sectional exposure contrast of 10 μg/m3 PM2.5, carotid intima-media thickness (CIMT) increased by 5.9% (95% CI, 1–11%). Adjustment for age reduced the coefficients, but further adjustment for covariates indicated robust estimates in the range of 3.9–4.3%. Among older subjects (≥ 60 years of age), women, never smokers, and those reporting lipid-lowering treatment at baseline, the associations of PM2.5 and CIMT were larger, with the strongest associations in women ≥ 60 years of age (15.7%, 5.7–26.6%).  In a more recent study in Los Angeles area87c, data from five double-blind randomized trials that assessed effects of various treatments on the change in common CIMT was reviewed. Spatial models and land-use data were used to estimate the home outdoor mean concentration of particulate matter up to 2.5 μm in diameter (PM2.5), and to classify residence by proximity to traffic-related pollution (within 100 m of highways). PM2.5 and traffic proximity were positively associated with CIMT progression. Adjusted coefficients were found to be larger than crude associations, not sensitive to modelling specifications, and statistically significant for highway proximity while of borderline significance for PM2.5 (p = 0.08). Annual CIMT progression among those living within 100 m of a highway was accelerated (5.5 μm/yr [95%CI: 0.13–10.79; p = 0.04]) or more than twice the population mean progression. For PM2.5, coefficients were positive as well, reaching statistical significance in the socially disadvantaged; in subjects reporting lipid lowering treatment at baseline; among participants receiving on-trial treatments; and among the pool of four out of the five trials.  A German study (Heinz Nixdorf Recall Study) from 200593 tried to assess the long-term personal traffic exposure and background air pollution by comparing 3,399 residents living within 150 m of major roads with those living further away. The principal outcome variable was clinically manifest coronary heart disease (CHD). The crude odds ratio (OR) for the prevalence of CHD with high traffic exposure was 1.62 (1.12–2.34) and rose to 1.85 (1.21– 2.84) after adjusting for cardiovascular risk factors and background air pollution. Subgroup analysis showed stronger effects for men [OR 2.33 (1.44–3.78)], participants younger than 60 years [OR 2.67, (1.24–5.74)], and never-smokers [OR 2.72 (1.40–5.29)]. A larger cohort (4,494 subjects),93a belonging to the Heinz Nixdorf Recall Study, was used to investigate the association between the level of coronary artery calcification and the distances between residences and major roads. Compared with participants living more than 200 m away from a major road, participants living within 50, 51 to 100, and 101 to 200 m had odds ratios of 1.63 (1.14 to 2.33), 1.34 (1.00 to 1.79), and 1.08 (0.85 to 1.39), respectively, for a high CAC (CAC above the age- and gender-specific 75th percentile). The study also found that a reduction in the distance between the residence and a major road by half was associated with a 7.0% (0.1 to 14.4) higher CAC.  12 Mechanisms of action Epidemiological studies over the last 20 years have shown with few exceptions42,52 that there is a relationship between air pollution (especially particulate matter) and cardiovascular diseases. The acute and chronic effects of ambient and traffic pollution were replicated by studies at various locales, and several large scale interventions that tried to control pollution were followed by a substantive decrease in cardiovascular outcomes. In parallel with these epidemiological studies, other researchers tried to look at the possible pathological mechanisms that link air pollution and various air pollution components with pathological changes in the body that lead to or precipitate heart failure or myocardial infarction. Animal and human studies that target various precursors and biomarkers of cardiovascular exacerbations were studied in relation to air pollution and several biological mechanisms of action were proposed.  The ample reviews mentioned previously6,16-27 not only describe the accumulated epidemiological evidence linking air pollutants with health effects (especially with cardiovascular outcomes), but they review the accumulated evidence for potential biological mechanisms that link pollutants to adverse cardiovascular events. From the several hypotheses that have been proposed, evidence is accumulating in support of two possibly interlinked mechanisms by which low concentrations of particles in inspired air may have adverse cardiovascular effects. In the first pathway proposed, the inhalation and the passage of fine particles through the alveolar epithelium may provoke an inflammatory response in the lungs with the consequent release into the circulation of prothrombotic and inflammatory cytokines, impairing vascular function and accelerating atherosclerosis.94-97 A systemic acute phase response of this nature would put people with coronary atheroma at increased risk of plaque rupture and thrombosis. This pulmonary oxidative stress/inflammation induced by inhaled pollutants represents fewer acute and chronic indirect effects.  The second, interlinked pathway suggests that direct effects may occur via agents that readily cross the pulmonary epithelium into the circulation, such as gases, and possibly ultra fine particles98-100 along with soluble constituents of PM2.5 (e.g., transition metals). In addition, the activation of pulmonary neural reflexes secondary to PM interactions with lung receptors may play a role. Exposure to PM may have an adverse effect on cardiac autonomic control,101-104 leading to an increased risk of arrhythmia in susceptible patients. These direct effects of air pollution represent a plausible explanation for the occurrence of rapid (within a few hours) cardiovascular responses, such as myocardial infarctions. A general scheme illustrating potential mechanisms of the effects of PM on the cardiovascular system is shown in Figure 1. Socioeconomic indicators and cardiovascular diseases The number of studies that investigate the link between air pollution and health is dwarfed by the research and effort gone into assessing the relationship between socioeconomic factors and cardiovascular health outcomes health. The simplest explanation for this fact lies probably in the availability of data, with information at an individual level of these parameters being more readily available than pollution information at an individual level. Even aggregate neighbourhood socioeconomic indicators based usually on census areas or zip/postal codes have a higher resolution than most of the ambient models used in assigning air exposure to individuals.   13 Table 3. Neighborhood based studies of SES and cardiovascular health outcomes Study Geographic area Cohort size Type of study Outcome Covariates Results  Krieger123 1992 N. California 1980 & 1985 14,420 subjects Retrospective study HBP Job type (area) Education (area) Race (area) 1.0 (0.9,1.2) 1.3 (1.2,1.5) 1.8 (1.6,2.0) Wing183 1992 US 1962 - 1978 VS mortality data for white women Retrospective study CVD Job type (area) Education (area) Income (area) All significant, visual analysis Diez-Roux184 1997 4 US communities 12,601 Retrospective study CVD prevalence (morbidity) % adults w/o high school (area) Median income (area) Median house value (area) % adults in occ. categ II-VI (area) 1.88 (1.00 – 3.52) 1.61 (1.11 – 2.87) 2.17 (1.20 – 3.94) 2.82 (1.29 – 6.16) Sundquist186 1999 Sweden 1988 - 1989 9,240 Retrospective study BMI Physical activity smoking Care Need Index (CNI) (area level aggregate index) 1.18 (1.02 – 1.36) 1.61 (1.34 – 1.93) 1.69 (1.42 – 2.01) Diez-Roux175 2000 44 US States 1990 70,534 Hierarchical analysis HBP Sedentarism Smoking Robin Hood Index 1.61 (1.17 – 2.21) 2.06 (1.27 – 3.35) 0.86 (0.59- 1.26) Diez-Roux190 2001 4 US communities 1987 - 1999 13,009 Prospective study CHD Neighbourhood aggregate factor, income, education, etc. 1.6 (1.1 – 2.2) Villeneuve12 2003 Vancouver 1986 - 1999 550,000 Time series study CVD mortality QAIPPE Not significant Sundquist194 2004 Sweden 1986 - 1993 25,319 Prospective study; Cox mod. Incident CHD Neighbourhood income Neighbourhood education 1.23 (1.00 – 1.52) 1.25 (1.02 – 1.54) Mujahid197 2005 4 US communities 1987 - 1999 13,167 Population based study BMI Census based SES aggregate factor + individual SES (-) association with income, education, neighb. SES for women; (-) association with income, education, neighb. SES for white men; (+) association with income, education, neighb. SES for white men; McGrath205 2006 2 schools in Pittsburgh; 212 adolescents Cross-sectional study; multilevel analysis BP (SBP/DBP), HR, negative mood Individual income, education Neighbourhood income, education, race profile  Individual income and education, and neighborhood race predicted heart rate Mobley202 2006 US 2001 - 2002 2,692 women Retrospective study BMI 10-year CHD risk Racial segregation land use crime rates neighbourhood income CHD ↓ BMI ↓ 2.6 kg/m2; CHD ↓ 20% BMI ↑; CHD ↑ BMI ↓; CHD ↓ Stjarne208 2006 Stockholm County 1992 - 1994 2,246 incidents population- based case– control study MI incidence rate ratio Median income Income distribution Women:1.94(1.28-2.96);Men:1.51(1.15-1.89) Women:0.77(0.50-1.19));Men:1.15(0.87-1.52) Chaix213 2007 Scania region, Sweden 1987 - 2002 ~ 1,000,000 Longitudinal study AMI incidence Previous heart diseases Alone vs. cohabiting Educational attainment Occupation 20-ye averaged income Neighborhood SES position Residential stability 1.54 (1.15 – 2.03) 1.34 (1.20 – 1.49) 1.43 (1.24 – 1.66) 1.14 (1.01 – 1.30) 1.65 (1.38 – 1.97) 1.67 (1.39 – 2.03) 1.19 (1.00 – 1.41) Lisabeth212 2007 Corpus Cristi, US 2000 - 2003 1,247 Cohort study; Poisson analyses Ischemic stroke Neighborhood SES score 90% vs. 10% RR: 1.06 (0.81–1.39) Ross210 2007 Canada 2000 - 2001 131,535 Cross-sectional study BMI recent immigrants, density, sprawl, education, median household income Significant for % immigrants (for men), education (men and women), and sprawl (men) MI: myocardial infarction; CHD: coronary heart disease; HBP: high blood pressure; VS vital statistics; CVD: coronary vascular disease; BMI body mass index; BP blood pressure; SBP systolic blood pressure; DBP diastolic blood pressure; SES socio economic status; AMI acute myocardial infarction  The relation between socioeconomic status and health is a problem that has been extensively studied.105 Research studies have shown a consistent inverse relationship between SES and morbidity and mortality rates. Morbidity and mortality rates generally decrease when moving up the SES ladder. This inverse relationship is observed whether SES is measured using education, income, or occupational status, and does not appear to be an artefact of the more physically ill individuals drifting down the SES hierarchy.106 The SES-health gradient extends to a wide array of health problems, including heart disease, cancer, stroke, diabetes, hypertension, infant mortality, arthritis, back ailments, mental illness, kidney diseases, and many others,107 and may predict prognosis after illness is present.108,109 For more detailed information on  14 specific studies on SES and health, there are several excellent reviews available.110-122 A condensed overview of some studies that used neighbourhood characteristics in relation with various health outcomes is presented in Table 3.   Figure 1. Mechanisms by which particulate and gaseous air pollutants may exert adverse effects on the cardiovascular system (from Routledge et al.6).   15 The effect of socioeconomic status on the relationship between atmospheric pollution and health The previous sub-sections reviewed the research literature that investigated the relation between ambient air and traffic pollution and cardiovascular diseases as well as the relation between socioeconomic status and cardiovascular diseases. However, there are relatively few studies that have examined the contribution of various environmental exposures, such as air pollution, to socioeconomic health inequalities.223 Some authors hypothesise that air pollution contributes to creating or accentuating the socioeconomic disparities seen in various diseases (including cancer224, asthma,225 and cardiovascular diseases226) and thus in premature death rates.227  As mentioned in the introductory section, two hypotheses have been suggested to explain the interplay between air pollution, SES, and health outcomes. The environmental inequality hypothesis proposes that populations with low SES may be more frequently or more intensely exposed to air pollution than those with high SES228, 229 and in this case air pollution acts as a confounder in the causal relation between SES and health outcomes. The evidence accumulated so far to reinforce this hypothesis linking the distribution of exposure to air pollution in populations with different SES is ‘‘mixed and inconclusive’’230 according to Bowen. Other studies231–234 support this observation. These mixed results might be explained by the great methodological diversity of these studies and the variety of their settings.232  The biological inequality hypothesis, which is also pursued by the present research, postulates that those populations with low SES may be more susceptible to air pollution than those with high SES.228 This susceptibility is caused by risk factors that are more prevalent in less advantaged populations and that can act as effect modifiers of the relationship between pollution and mortality. These risk factors include poor health status (for example, diabetes, obesity, and chronic obstructive pulmonary disease),228 addictions (including smoking),235 and multiple pollutant exposures (passive smoking, occupational exposure); these are likely to act in addition to or in synergy with urban pollution.235  In Laurent et al.,237 a review of the relevant research literature that looks at the contribution of air pollutants to socioeconomic health inequalities, the authors concentrate their research only on the published articles that deal with the second potential mechanism, arguing that the existing research that looks at the environmental inequality hypothesis has mixed and inconclusive results. Laurent’s review237 found, for both short-term and long-term studies of the effect of air pollution on mortality that those studies using socioeconomic characteristics measured at coarser geographic resolution showed no effect modification, those studies using finer geographic resolutions showed mixed results, and those studies using individually measured socioeconomic characteristics showed that pollution affected disadvantaged subjects more.  The conclusion of Laurent’s237 review is that populations of different SES levels need to be tested for a higher range of pollutant concentrations and that further research should consider the largest possible number of SES indicators (both individual and contextual at different geographic resolutions) in order to identify those that are most discriminating in terms of the relative risks of mortality or morbidity associated with pollution.   16 The same review found fifteen articles (time series, case-crossover, and cohort) that examined short-term effects of air pollution on mortality. Because of the variety of socioeconomic indicators studied, a formal comparison was difficult. However, the reviewers concluded that studies using socioeconomic characteristics measured at coarser geographic resolutions (city- or county-wide) found no effect modification, but those using finer geographic resolutions found mixed results. Five of six studies using individually-measured socioeconomic characteristics found that pollution affected disadvantaged subjects more. These findings from the short-term studies were complemented by the six studies (which employed cohorts of subjects) of long-term effects that the reviewers identified as suitable for inclusion in the review. The same problem, substantial methodological differences, plagued the interpretation of the long term studies, and the same general observation re-emerged. The reviewers concluded that current evidence does not provide sufficient and definitive evidence that socioeconomic characteristics modify the effects of air pollution on mortality. However, the results so far, by their tendency to show greater effects among the more deprived, emphasise the need for further investigation of this topic.  A recent study238 conducted by Dragano investigated whether the association between traffic exposure and sub-clinical cardiovascular disease is modified by socioeconomic characteristics of individuals and neighbourhoods. The cohort used in the study was mentioned previously93,93a as being used to investigate the relationship between traffic pollution and cardiovascular health outcomes. However, in this particular study, the cohort (2,264 women and 2,037 men aged 45–75 years) was used to investigate the associations between high traffic and coronary artery calcification within strata of SES to determine effect modification. The researchers found that high traffic and low SES were both associated with higher amounts of calcification (>75th age-specific percentile). Although a higher number of participants with low SES were found to live close to major roads, the stratified analyses did not indicate higher susceptibility in low SES groups. However, the study found that participants with low SES and exposure to high traffic had the highest levels of CAC. When considering individual level of education, better-educated men with low traffic exposure had a prevalence of high calcification of 23.9%, but it was 37.7% in lower-educated men with high traffic exposure (women: 22.0% vs. 28.1%). For neighbourhood levels of unemployment, it was found that men living in neighbourhoods with low unemployment rates and within 50 m of roads had a prevalence of high calcification of 25%, while men living in neighbourhoods with high unemployment rates and within 50 m from roads had a prevalence of high calcification of 50% (29.2 % vs. 42.1% for women). The authors conclude that high traffic exposure was associated with coronary calcification in all social groups, but because low SES individuals had higher calcification in general and were also more exposed to traffic, the existing inequalities could be further shaped by traffic exposure.  A very recent study by Ren et al.238b looked at the socioeconomic modifiers of short-term effects of ozone on mortality in eastern Massachusetts. In this study, the authors used a case- crossover design to examine whether impacts of ozone on mortality were modified by socioeconomic status coded at the tract level or characteristics at an individual level in eastern Massachusetts, US for the period between May 1995 and September 2002. The authors looked at 157,197 non-accidental deaths among those aged 35 years or older and used moving averages of maximal 8-hour concentrations of ozone monitored at 8 stationary stations as personal exposure. They found that a 10 ppb increase in the four-day moving average of maximal 8-hour ozone was associated with non-significant changes in cardiovascular diseases,  17 heart diseases, acute myocardial infarction, and stroke respectively (0.44% (95% CI: -1.45%, 2.37%), -0.83% (95% CI: -2.94%, 1.32%), -1.09% (95% CI: -4.27%, 2.19%) and 6.5% (95% CI:1.74%, 11.49%)), and concluded that there was no evidence that the associations were significantly modified by socioeconomic status or individual characteristics, although small differences of estimates across subpopulations were demonstrated. Summary of literature review Epidemiological evidence has consistently shown that various air pollutants, especially particulate matter, are aggravating risk factors in the triggering, progression, and full manifestation of various cardiovascular events2a. This was demonstrated when considering exposures at large scales and small scales as well as proximity to roads. Meta analyses of the time series data suggest that an increase in fine particulate pollution of 10 μg/m3 is associated with an increase in total mortality of 1.8% and cardiovascular mortality of about 1.4%.6 Hospitalization analyses indicate similar results. The excess number of deaths due to air pollution reported by various studies3-8 was confirmed by several large scale interventions. There is a confirmed presence of an apparent linear dose-response relationship between PM and adverse health events, relationship that has no discernible threshold below which PM concentrations pose no health risk to the general population3.The adverse cardiovascular outcomes in the general population are seen at levels at or below existing air quality standards. Pathophysiological studies on animals and humans have shown that several biological mechanisms can explain the interaction between the human body and particulate matter and/or various gases that lead to a cardiac event.94-104  Epidemiological and sociological studies have shown the correlation between lower socioeconomic status and health, considering individual and neighbourhood characteristics. Income, wealth, education, and personal support, are all important indicators of health. A lack of them is associated with higher stress levels that have been shown to be a mechanism for the onset of coronary heart disease.9, 182 Access to health care,180,200 better food,200 recreational facilities,201 social support, better housing conditions,140,156 and public transportation,160-195 with an absence of violence, and an environment away for sources of pollution,170 are all important factors in preserving a healthy life. Neighbourhoods that are lacking in such amenities are more inductive to unhealthy life styles and stress, and ultimately to poorer health for their inhabitants.132,145-147,148,150,158,159  It is apparent that the number of factors involved in the final health outcome of an individual can be quite large. For instance, Jerrett et al.13 used an excess of 44 individual potential confounders identified in earlier ACS studies84 of air pollution health effects to which he added some neighbourhood identifiers. These variables include lifestyle, dietary, demographic, occupational, and educational factors that may confound the association between air pollution and mortality.  While there is compelling evidence of the effects of air pollution on health and of the impact of SES on health, there is less evidence and understanding of the mechanisms and the magnitude through which SES modifies the effect of air pollution on health, and this research tries to address this particular knowledge gap.  18 Objectives It is evident from the array of studies conducted to date that the environment plays a significant role in the onset of CVD. Air pollution and socioeconomic factors (among others) contribute to the relative risk of CVD outcomes. The objective of this project was to analyze, in a combined framework, the risk of CVD relative to traffic air pollution, taking into account the socioeconomic status of subjects’ neighbourhoods. The study will use two levels of aggregation of SES variables. One level of aggregation is at a large scale, using Statistics Canada Dissemination Areas, while the other level of aggregation is represented by actual geographical neighbourhoods as defined by their residents. These two levels will be used in separate analysis to check the assumption that aggregate socioeconomic variables at a finer grained resolution are capable of producing more significant results than those based on coarser level of aggregation. This study will consider chronic exposure to air pollution and will assess several SES indicators at the two levels of aggregation as potential effect modifiers for the risk of cardiovascular morbidity due to long-term exposure to traffic air pollution in a large population cohort. To summarize, the two main questions that this study tried to address were: (1) is increased air pollution (road proximity) associated with increased CVD outcomes? and (2) is there effect modification by socioeconomic status? This study is relevant first by exploring the impact of lower levels of traffic pollution than usual and road proximity in a large cohort with excellent residential history. Also, this study will provide maybe for the first time a consistent assessment on the way the joint effects of socio-economic status and pollution are impacting cardiovascular health outcomes. Hypotheses Two main hypotheses will be investigated: 1. that increased traffic pollution and closer proximity to main roads are associated with an increased incidence of CVD outcomes. 2. that although increased air pollution levels affect people indiscriminately, people living in neighbourhoods with higher socio-economic status will be less affected by air pollution than those living in neighbourhoods with lower socio-economic status.        19 Methods Study population The study population consisted of all residents of greater Vancouver metropolitan region who were 45 to 84 years of age as of January 1st, 1999, had lived in the area for the 5 years prior to 1999 (1994-1998), were alive as of December 31st, 1998, and did not have a diagnosis of cardiovascular diseases or diseases considered as a risk factor for developing cardiovascular diseases prior to January 1, 1999. The study cohort1 was assembled by extracting data from a series of linked administrative datasets obtained from the British Columbia Ministry of Health Services and British Columbia Vital Statistics Agency.  Several criteria were applied in order to define the cohort subset eligible for the analysis and to establish the end of follow-up for each of the subjects. All subjects had to be registered continuously, with registration gaps of a maximum of 6 months (183 days) being permitted between April 1st, 1994 and December 31st, 1998 and onward until Dec 31st, 2002 or the end of the registration in the provincial universal medical plan, whichever was earliest (April 1st was chosen instead of January 1st because the registration dates are related to the financial year rather than the calendar year, and also because the three-month gap at the beginning would satisfy the condition that there should not be registration gaps greater than 183 days). The follow up period for the study’s subjects is from January 1st, 1999 to December 31st, 2002. For subjects that died in the follow-up period (between January 1st, 1999 and December 31st, 2002), the registration end date was set to be at the end of the month of death, if the actual registration end date was later than the last calendar date of the month of death.  All subjects had to continuously reside in the study area (Greater Vancouver metropolitan area) between January 1st, 1994 and December 31st, 1998. The end of residential history for subjects with continuous residence in the area between January 1st, 1994 and December 31st, 1998 was considered to be the date when they left the area after December 31st, 1998, or Dec 31st, 2002, whichever date occurred first. Thus, the end date of the follow-up for a subject for a particular outcome of interest was considered to be the earliest date between December 31st, 2002, the date of death (if the subject died in the follow-up period), the end of registration in the provincial universal medical plan, the end of residence in the study area, or the date when the outcome of interest occurred (if there was an outcome of interest).  From the original 876,473 subjects living in Georgia Air Basin, only 534,856 were found to have lived continuously in the Greater Vancouver metropolitan area from January 1, 1994 to the end of follow-up period (stretching from January 1, 1999 to December 31st, 2002). Out of this number, 6,471 subjects were excluded from the study because there were gaps in the registration history that were greater than 183 days. Also, 3 additional subjects were found to be included as a consequence of erroneous linkage between the several administrative  1 The original cohort was extracted from a larger area, namely the Georgia Air Basin, and consisted of 876,473 subjects identified as satisfying the age eligibility criteria. However, because the traffic exposure assessment was available only for the greater Vancouver metropolitan region, the number of subjects retained was substantially smaller.  20 databases used. Thus, a total of 528,382 subjects with complete medical and residential history were available for analyses in the Greater Vancouver metropolitan area.  By considering the eligibility criteria (the non-presence of any cardiovascular health outcomes and health outcomes considered as risk factors for cardiovascular diseases), only 356,893 subjects were further retained from those with a full residential history in the Greater Vancouver metropolitan area. The cardiovascular health outcomes and the health outcomes considered as risk factors for cardiovascular diseases prior to the start of the follow-up period were:   acute coronary syndrome (ACS), defined as either acute myocardial infarction, unstable angina or other acute forms of ischemic heart disease (ICD9 = 410 or 411 or ICD-10 = I20.0, I21, I22 or I24.9);  chronic coronary syndrome (CCS), defined as stable angina pectoris, other chronic forms of ischemic heart disease, or atherosclerotic cardiovascular disease (ICD9 = 413, 414, 429.2 or ICD-10 I20.1, I20.8, I20.9, I25.0, I25.1, I25.9, or I51.6);  congestive heart failure (CHF) (ICD9= 428 or ICD-10 = I50);  hypertensive disease (HTN) (ICD9 = 401-405 or ICD-10 = I10-I15);  chronic obstructive pulmonary diseases (COPD) (ICD9=466, 490-492, 496 or ICD-10 = J41-J44);  diabetes mellitus (DM) (ICD9= 250 or ICD10= E10-E14).  In order to determine if a person had one of these pre-disposing conditions, one hospital diagnostic (principal or primary) or two (in case of HD, three) out-patient diagnostics per year were required as a case definition.  Subsequent linkages with the socioeconomic indicators from the Canadian Census further diminished the number of subjects. A total of 346,536 subjects with full census, demographic, and residential data were retained for analyses. Additional subjects were excluded from pollutant specific analyses if traffic pollution measurements were not available: between 304 and 18,246 subjects did not have exposure data for various traffic- related pollutants (NO: 304, NO2: 326, PM2.5: 18,246, and black carbon - B.C.: 1,519). Residential history Three sets of data were used to reconstitute the residential history2 of each individual at postal code level. These files are:   BC Ministry of Health Services Registration (BC MoHS) & Premium Billing (R&PB) files  BC's health services utilization files:  Medical Services Plan (MSP) Payment Information  Discharge Abstract Database - DAD (Hospital Separations)  2 Because personal information (i.e. full six digit postal code address) was not directly available, the residential history was reconstituted under the privacy screen by an analyst at CHSPR.   21 These data files provided the set of postal code ‘observations’ for the cohort, including associated dates. Only records with postal codes were retained. Although the three data sources are not necessarily independent, due to BC MoHS Client Registry input into MSP and Hospital records, the MSP and Hospital files may reflect updated postal codes more quickly than the R&PB file, as the latter relies only on R&PB file. The creation of residential history for each subject in the cohort required a substantial amount of processing due to the inherent messiness of the data with issues like multiple postal codes recorded on the same service date, invalid postal codes, non-residential postal codes, etc.  The problem of the substantial amount of potentially spurious or uninformative data was resolved by retaining only postal codes that had a minimum of two observations (encounters with the medical system) at least one month apart. By applying this rule, many non- residential postal codes were removed from the data. Many inconsistencies in residential history were resolved by eliminating spurious and non-residential postal codes. The remaining inconsistencies were resolved by removing postal codes ending before 1994 and by comparing attributes of postal codes in terms of spanned overlap (e.g., if postal code B was observed only a couple of times, but postal code A spanned the length of the data, then A was set as the address throughout; if multiple reasonable postal codes overlapped, then they were all accepted and transition dates were set to remove overlaps). Also, hospital address postal codes were retained when they were consistent with a nursing home address (and if appeared last in the data).  By applying all these strategies, 46% of subjects retained a single postal code after the deletion of inconsistent postal codes and non-residential addresses. An additional 27% of subjects had more than one postal code, without overlaps. About 7% of subjects had a complete overlap of a postal code with relatively more observations over another postal code with relatively fewer observations that left a consistent residential history after being deleted. About 12% of subjects had a partial overlap that was resolved when transition dates were set. Another 2% of subjects had a shortfall of less than a year in coverage at the beginning and end of follow-up due to the way the dates were originally set. By extending the first and last postal code of a subject by up to one year to match the dates of subject registration, this problem was resolved. Thus 6% of the subjects had no usable residential information and they were not considered for analyses (2% of subjects were unresolved due to urban PO Box addresses and 4% of subjects had relocated postal codes or other non- residential postal codes, or too few observations).  In order to test the validity of the residential history derived from administrative databases, the Canadian Community Health Survey (v. 2001) (CCHS) was used as a benchmark. The first three digits of the postal codes (indicating the Forward Sortation Area – FSA) from the CCHS dataset were compared with the administratively-derived FSAs of the subjects, and the FSAs were recorded in the same time interval in which the CCHS survey was conducted.        22 Health outcomes Health data are available from the British Columbia Linked Health Database (BCLHD) for research purposes, through an approved process governed by a data access agreement3 between the researchers and the data stewards. Medical services and hospitalization data were provided and governed by the Ministry of Health, Government of British Columbia, and vital statistics data by the British Columbia Vital Statistics Agency. The research database was constructed by merging vital statistics death records (for cohort enumeration according to residential postal codes) with outpatient medical services billing records and inpatient hospital discharge records, for identification of cases for the period of 1999–2002 (and all co-morbidities between 1991 and 2002). Socioeconomic indicators for education, income, and other attributes were available from Statistics Canada census data. The research database was provided to the research team with all personal identifiers removed and replaced by anonymous study identifiers. The identifiers were unique to each individual and enabled identification of the same individuals across data sources. The study protocol was approved by the Behavioral Research Ethics Board of The University of British Columbia.  The health outcomes of interest were separated into three broad categories:  acute coronary syndrome (ACS), defined as either acute myocardial infarction, unstable angina or other acute forms of ischemic heart disease (ICD9 = 410 or 411 or ICD-10 = I20.0, I21, I22 or I24.9);  chronic coronary syndrome (CCS), defined as stable angina pectoris, other chronic forms of ischemic heart disease, or atherosclerotic cardiovascular disease (ICD9 = 413, 414, 429.2 or ICD-10 I20.1, I20.8, I20.9, I25.0, I25.1, I25.9, or I51.6);  congestive heart failure (CHF) (ICD9= 428 or ICD-10 = I50)  Because the focus of this study was to investigate the health effects of chronic exposure to traffic, the most relevant cardiovascular health outcome for long term traffic pollution exposure is represented by the diseases grouped under the chronic coronary syndrome. Thus the results chapter will focus only on the CCS outcomes while the analyses pertaining with ACS and CHF will be presented in the appendix.   The follow-up period was between January 1st, 1999 and December 31st, 2002. A subject in the cohort was considered to have one of these health outcomes if there was a hospital admission with a principal diagnosis (from the Hospitalization Discharge File) or a death (from BC Vital Statistics deaths file) due to one of these health outcomes.  The diagram in Figure 2 depicts the overlay between the time frame for which co-morbidity data was available and considered for analysis, the chronic exposure data, the residential history data, census data, and the follow-up period for which the health outcomes of interest (CCS, ACS and CHF) were assessed.     3 Chamberlayne R, Green B, Barer ML, Hertzman C, Lawrence WJ, Sheps SB. Creating a population-based linked health database: a new resource for health services research. Can J Public Health. 1998;89(4):270– 273  23  Figure 2. Time periods and data availability Individual and small-scale socioeconomic covariates Only sex, age, postal codes, and health outcomes were available at the individual level from the existing administrative data sources. Statistics Canada 2001 Census data were used to assign socioeconomic information to each subject based on their residence at the Census Dissemination Area resolution level. Dissemination areas are the smallest geographic areas for which Canadian Census data are aggregated, are randomly derived, and they correspond to one or more neighboring blocks with target populations of 400 to 700 persons.215  Residential postal codes were allocated to their corresponding DA using DMTIs CanMap multiple enhanced Postal Code, 2005 files. Also, using the compiled residential history, each subject in the cohort was merged with the SES variables dataset, with the SES values corresponding to the most extended residential location in terms of time spent there between 1999 and 2002 for each individual.  Hundreds of socioeconomic variables were available from Stats Canada 2001 Census data. However, based on the literature review performed prior to the analyses, about 25 variables were initially chosen to act as socioeconomic covariates at DA level, number that ulterior, in the end, was reduced to ten variables. These variables are: the average individual and family income, the percentage of people with a university degree, the percentage of people from China or of Chinese descent, the proportion of people that used bikes, public transit, or walking to commute to work, the coefficient of income variation, the percentage of home ownership, the percentage of people working in management jobs, the proportion of people with low income, and the rate of employment for the. These ten socioeconomic variables used at DA level can be broadly classified in five categories of variables. These categories are:  Racial/Cultural The racial/cultural category is represented by the percentage of Chinese minority variable. The percentage of immigrants210 and time since moving to the neighborhood213 were Jan 1991 Jan 1994 Jan 1998 Jan 1999 Jan 2001 Dec 2002 Medical pre-conditions Health outcomes Census data Exposure assessment data Residential history data Follow-up period  24 variables found to have an impact on health outcomes. A Canadian study210 showed a relationship between racial characteristics and obesity in females and although the individual subjects in the study do not have a racial profile available (except for the native status), the percentage of population of Chinese origin or descent is available at DA/EA. A recently released Canadian study272 investigating cardiovascular risk profiles among people living in Ontario found that the risk profiles (smoking, hypertension, obesity, and diabetes mellitus) among blacks, whites, Chinese, and South East Asians varied considerably. There are many studies available that link obesity with socioeconomic status and with cardiovascular diseases.197,199,202 Most of the studies in the U.S. use a racial indicator in their analyses because income in U.S. is well correlated with race.129,137, 139,155,202,205 Although this might not be the case in Canada, the introduction of a variable describing the racial composition of individual DAs can be useful from other standpoints, such as the one mentioned before, or from broader standpoints, such as culture.119, 151-153  The variable representing the percentage of visible minorities of Chinese descent in the DA was considered to be used as a variable representing Chinese cultural values and broader East-Asian cultural values and norms. The decision to use this variable was made after investigating the correlation between the variable representing the percentage of total visible minorities, the variable representing the percentage of Chinese visible minorities, the variable representing the percentage of Asian visible minorities (Chinese, Japanese, Korean, Filipino), and the variable representing the percentage of visible minorities of South Asian (Indian) descent.  Income and wealth This category of socioeconomic variables includes the average personal income variable, the average family income variable, the percentage of people with low income, the income variation variable, and the percentage of occupied dwellings that are owner-occupied variable. The coefficient of variation for income gives an idea of the range of values around the mean income for each DA.188,208 Dwelling value and ownership gives an indication of the wealth of a person147 which is considered to be a supplementary indicator that might be linked with stress, with higher accumulated wealth being potentially associated with lower levels of stress.  Education This group consists in only one variable, the percentage of total population with any university degree.  Labour This group includes the employment rate variable and percentage of people working in management variable. The type of work performed,110,117,119,134,140,156,173 was another variable of significance in relation to health.  Transportation means This category consists of the percentage of people biking, walking, or using public transit in their daily commute to work variable. The variable representing the percentage of people that used biking, public transit, and walking to commute to work was used because it was found that commuting and the way it is done might impact cardiovascular health outcomes (Peters et al. 2004)91. Also it was assumed that people walking or biking to work are exposed  25 to more traffic pollution than car users because they are just beside traffic without the protection of a car’s interior. Also, the stress in traffic,195 was another variable of significance in relation to health.  All these SES variables were included in the analyses as categorical variables, after being partitioned in quintiles. There were a total of 6,572 DAs in the Greater Vancouver metropolitan area, but for some of them not all of the variables of interest were available, thus subjects living in these DAs were discarded from the analyses. Medium scale socioeconomic covariates Statistics Canada Dissemination Areas are statistical units designed for collecting and presenting census information, and do not necessarily have any relation with neighborhoods as people actually perceive them. To avoid this drawback of census-defined areas and also to employ socioeconomic variables at a different level of aggregation, I employed the neighborhoods defined by the B.C. Atlas of Child Development.216 In the Atlas, school districts were used as a blueprint for more refined neighborhoods. Communities and volunteers participating in the Early Development Initiative were involved in determining neighborhood boundaries that more accurately reflect the lived experience of a diverse range of people that reside in the area. Local representatives were invited to draw lines on maps of their area to signal the presence of perceived divides in their community. While some opted to maintain the Census or another existing boundary system, others opted for totally different configurations.  For the creation of the Atlas, the study team worked with Statistics Canada to amalgamate SES indicators at the neighborhood level. Statistics Canada used the 2001 census information collected at the block face level and aggregated it according to the boundaries defined in the Atlas. Although the reported SES variables created for the Atlas do not entirely match those available at the DA level, most of them can still be used. There were ten variables at the neighborhood level that were selected to be used in the analyses.  The average individual and family income, the percentage of home ownership, the incidence of low income, and the percentage of people with a university degree variable from the B.C Atlas of Child Development matched entirely the variables at the DA level. The rate of employment at the DA level is related with the rate of unemployment at the neighborhood level. The variable representing the percentage of visible Chinese minorities from DA has a clear relationship with the variable representing the percentage of total population whose home language is neither English nor French and with the variable representing the percentage of total population without knowledge of English or French from the Atlas. The stress variable from the Atlas, which represents the percentage of families spending 30% or more of income on shelter costs, was also considered relevant. All the variables retained for the analyses were transformed in categorical variables by being partitioned in quintiles.  Using the B.C. Atlas of Child Development and the GIS boundary layers employed by the Human Early Learning Partnership (HELP) Institute to produce the B.C. Atlas of Child Development, there were a total of 321 neighborhoods identified in GAB, of which 5 neighborhoods were not surveyed and did not have aggregate census variables calculated. Table 4 presents a synopsis of the variables selected at dissemination area and  26 neighbourhood levels, while Figure 3 presents an outlay of dissemination areas and neighbourhoods in Vancouver, which is included in the Lower Mainland.  Table 4. SES variable to be used in statistical analyses Category of variables SES variable at DA level SES variable at Neighbourhood level Expected behaviour Foreign Home Language: % of total population whose home language is neither English nor French Cultural/Racial Percent of visible minorities from China Linguistic Isolation: % of total population without knowledge of English or French Areas with high levels of minorities will have lower CVD HRs Education Percent of people over 15 with university education University Education: % of total population (>=20 years of age) with any university degree Areas with high levels of university education will experience lower CVD HRs Average 2000 total income $ in population over 15 years Average Employment Income: average annual employment income in dollars Areas with high levels of personal income will experience lower CVD HRs Average 2000 family income $ Median Family Income: median annual family income in dollars Areas with high levels of family income will experience lower CVD HRs Coefficient of variation of income in population over 15 years Income from Government Transfers: % of aggregate neighbourhood income from any government transfer Areas with high levels of CV will experience lower CVD HRs; Areas with high levels of gov transfers will experience higher CVD HRs Incidence of low income in 2000 % Persons Below LICO: % of persons in households below the low-income cut-off (LICO) Areas with high levels of low income will experience higher CVD HRs Homeownership Rate: % of occupied dwellings that are owner-occupied Areas with high levels of home ownership will experience lower CVD HRs Income and wealth Percent of owned dwellings  Housing Stress Index: % of families spending 30% or more of income on shelter costs Areas with high levels of housing stress index will experience higher CVD HRs Percent of people employed in population over 25 years Areas with high levels of employment will experience lower CVD HRs Labour Percent of people in labour working in management Unemployment Rate: seasonally adjusted unemployment rate among persons aged 25 years and over Areas with high levels of management workers will experience lower CVD HRs Transportation Percent of working people that uses transit, bikes or  walks to work  Areas with high levels of usage of transit, biking, walking will experience high CVD HRs   27  Figure 3. Dissemination Areas, Neighbourhoods and forward Sortation Areas in Vancouver Air pollution exposure assessment Land use regression The land use regression models used in this study were previously developed for the study region by Henderson217 and colleagues in 2007 to provide improved local spatial resolution. Pollution data was considered a time-dependent covariate in the Cox proportional hazards model employed in the analyses. For each month in the follow-up period (January 1st, 1999 – December 31st, 2002) the average exposure of the previous year was calculated using the residential history and the land use regression model. Thus, each subject had up to 48 months of exposure information.  The land use regression model was built by assessing the association between variables describing land use and traffic information and the NO and NO2 concentration measured at 116 sites in the study area over two 14-day periods. The mean concentrations during these two periods closely approximated annual averages from regulatory monitoring network data, and were highly correlated with these averages. The PM2.5 model was developed using data from a subset of 25 locations during a 2-month sampling period.  28 For a subset of 36 sites, particle absorbance (Black Carbon) was measured using a Particle Soot Absorption Photometer (Radiance Research, Seattle WA) in a mobile monitoring platform. These measurements were then adjusted for temporal variation based upon repeated measurements at a centrally-located site. For NO, the model had an R2 of 0.62 and included the number of major roads within 100-m and 1,000-m radius circular buffers around the measurement sites, the number of secondary roads within a 100-m buffer, the population density within a 2,500-m radius, and elevation. For NO2, the model (R 2 = 0.56) included the same variables as well as the amount of commercial land use within 750 m. For PM2.5 the model (R 2 = 0.52) included the amount of commercial and industrial land use within 300 m, the amount of residential land use within 750 m, and elevation. For black carbon, the model (R 2 = 0.56) included the number of secondary roads within a 100-m buffer, the distance to the nearest highway, and the amount of industrial land use within 750 m (Brauer et al. 2008).216a  The model output consisted of yearly exposure averages based on the measurements done in 2003 for a 10 m2 grid for the study area. Ambient monitoring data obtained from the local monitoring network was used to identify the long-term trends in NO, NO2, and particulate pollution, and the coefficients obtained from these trend analyses (performed using the Times Series Forecasting System from SAS v 9.1.2) were used to adjust the land use regression yearly averages in order to obtain the average exposures of the previous year for each month in the follow-up period. Proximity to roads Road proximity for home postal codes of all cohort members was calculated by the author as a proxy for traffic exposure. Road classifications (DMTI ArcView street file dataset for British Columbia, Canmap Streetfiles, v2006.3, 2006) were used to determine whether a home postal code was within 50 m of an expressway or primary highway (R-I), between 50 m and 150 m of an expressway or primary highway (R-II), within 50 m of a secondary highway or major road/arterial road (R-III), between 50 m to 150 m of a secondary highway or major road/arterial road (R-IV)or within 150 m of a secondary highway or major road, or within 50 m of an expressway or primary highway (R-V). The R-I and R-II road categories and R-III and R-IV road categories are mutually exclusive but this not preclude a subject living within 50 m of an expressway or highway to also live within 50 m or within 150 m of a primary road or a major road. This is why the sum of subject living in the proximity of R-I, R-II, and R-III will be less than the number of subjects found for R-V category. Statistical analysis The initial analyses performed consisted in validating the administrative databases and especially the health outcomes derived from the administrative databases with data derived from the 2001 Canadian Community Health Survey (CCHS). The information from the CCHS was also use to investigate the effect of smoking on the CVD outcomes as well as the correlation between the individual level variables and census SES and pollution variables.  There are two main sets of statistical analyses that were performed for this study. For the first set of analyses, Cox proportional hazard analysis (using SAS v 9.1.2) was used to investigate the effect of air pollution on cardiovascular health outcomes adjusting for age, gender, and SES at the DA and neighbourhood levels. Pollution exposure (yearly average of  29 traffic pollution exposure prior to the event) and road proximity variables were treated as time-dependent variables.  The second set of statistical analyses followed the methodology suggested by Laurent et al.237 for investigating the modifying effect of SES on the relationship between traffic pollution and cardiovascular health outcomes. Figure 4 illustrates a fictitious example suggested by Laurent:237 the slope of a dose–response curve corresponding to a population with low SES might be stronger than that of a population with high SES for some concentration ranges (between x1 and x2), and lower for a range of higher concentrations (between x3 and x4). The slopes of these curves may be considered equivalent to a hazard ratio or relative risk. This shows the importance of taking into account the range of pollutant concentrations tested for which SES might be an effect modifier.  Figure 4. Fictitious example of dose-response relationship in low and high SES populations (from Laurent et al. 2007)237  In order to follow this methodological approach, all SES variables retained for analyses, regardless of their level of aggregation, were ranked and classified by quintiles. Stratified Cox proportional hazard analyses were run (using SAS v 9.1.2) for the lowest and highest quintiles for each of the SES variables and for all four traffic pollutants available. Pollution levels were ranked and classified by quartiles and were used in the analyses as time- dependent variables.  30 Results Validation of the administrative database with the Canadian Community Health Survey information CCHS data summary A pilot validation study containing subjects from the Border Air Quality Study that participated in the 2001 Canadian Community Health Survey was performed. A dataset with a total of 2,824 subjects was obtained form CHSPR. One subject with duplicate Study ID was eliminated and another subject was also eliminated because it had two different ages. Thus, the final number of subjects from CCHS found in the BAQ Study is 2,821. Out of these, 1,470 subjects were matched to the study cohort.  The proportion of CCHS subjects with full data in the study area, out of the total CCHS survey data with full information (2,752) was only 53% which indicates that there might be some underlying problems that led to a greater reduction in the CCHS/GVRDplus sub- cohort compared with the GAB/ GVRDplus population.  One thing to consider is that the CCHS surveys people between 12 and 74 years old while the BAQ Study looked only at people over 45 years of age as of January 1st 2009. Another cause of the difference might reside in the different sampling intensities that Statistics Canada uses for different regions. For this reasons we compared the sampling weights of the subjects from CCHS in the GVRDplus area with the sampling weights of the subjects that are outside this area. The results are presented in Table 5.  Table 5. Comparison between the sampling weights of subjects from GVRDplus area and subjects outside GVRDplus Area N Mean Std Dev Minimum Maximum GVRD plus 1,470 237.55 171.89 25.27 1946.85 Outside GVRD plus 1,279 175.26 115.83 15.09 938.25  It can be seen from Table 1 that there are major differences between the two sub-cohorts in terms of the sampling intensity used by Statistics Canada. Assuming that subjects with smaller weights demanded a higher sampling intensity, it appears that CCHS survey over sampled in areas with smaller population. The CCHS survey data was also used to check the validity of the administratively derived demographical, medical, and residential medical history data. Residential history check The first three digits of the postal code (indicating the Forward Sortation Area) from the CCHS dataset were compared with the FSAs of the subjects recorded on the same time interval the CCHS survey was conducted. Thus from the 1470 subjects from GVRDplus in CCHS, one has left the GVRDplus area prior to the date of the survey so it does not appear in the residential history during that time frame. However, out of the remaining 1469 subjects from the greater Vancouver metropolitan area in CCHS, only 144 (9.80%) showed differences between their declared residence in the CCHS and the residence obtained from  31 the residential history file compiled using the medical data files. When no time restriction (postal code address from the 2001 CCHS survey had to match the postal code of residence derived from the administrative databases for the same time interval the CCHS was conducted) was applied to the postal codes in the residential history file, an even smaller discrepancy with CCHS addresses was noted. Demographic check Gender There was a misclassification of gender when comparing the BAQ study information with the CCHS information. Thus, 7 males (according with CCHS) were reported as females by the BAQ Study data. These subjects were not included in any further analyses.  Age There were differences in age among the 1469 subjects from CCHS in GVRDplus when comparing the BAQ Study derived age with the CCHS derived age. A total of 109 subjects differed in their age when comparing CCHS year and month of birth with registry year and month of birth. However, there were only 65 subjects that differed in their year of birth. Out of these 65, 32 subjects had a difference of 1 year, 4 subjects had a difference of 2 years, and for 4 subjects there seems to be juxtaposition or misreading in the year.  Thus, only 25 subjects were discarded. After removing the 7 subjects that did not have similar genders in the two databases, only 1,442 subjects were left for future analyses. A cross-check was done on the 25 subjects considered as being problematic in respect with the year of birth, to see if the CCHS address matches the BAQ study address from residential history. Only 4 subjects out of 25 had different residential histories Health history check The subjects in the CCHS survey were asked several questions regarding their health status in respect with several diseases. A general variable defines the health status of the subjects in respect with heart diseases while other variables are more specific in respect with what type of heart disease one has. The health status of CCHS subjects in respect with hypertension (has high blood pressure), COPD (has emphysema, COPD or chronic bronchitis), and diabetes (has diabetes) is represented by individual Yes/No variables.  The variables indicating the presence/absence of heart diseases, hypertension, diabetes and COPD were compared with the MSP and hospitalization health outcomes obtained from the respective files. A relatively low level of matching could be noted across all health outcomes studied. Table 6 shows the proportion of subjects in the CCHS (1,442 subjects) that had declared that they have one of the three health outcomes of interest and were found from the MSP and hospital discharge files that they actually have those health outcomes. One caveat regarding the health outcomes from administrative data is that only data from 1991 to 2003 was available for creating a health history for any individual, which might explain the low level of concordance between the CCHS answers and the administrative health data.  The summaries from Table 6 indicate that there is a relatively high level of concordance between the health status reported in the CCHS and the administrative databases used to extract and define the health status for the subjects in the cohort. Thus, there was a 93,3%  32 match regarding ACS status, 89% match regarding CCS status, and 95.4% regarding CHF status.  Table 6. CCHS health outcomes vs. Administrative health data Administrative health data All Heart Diseases ACS CCS CHF CCHS Health Status No Yes No Yes No Yes No Yes No 1160 (80.44%) 109 (7.56%) 1321 (91.61%) 54 (3.74%) 1229 (85.23%) 145 (10.06%) 1361 (94.38%) 46 (3.19%) Yes 48 (3.33%) 125 (8.67%) 28 (1.94%) 39 (2.70%) 15 (1.04%) 53 (3.68%) 21 (1.46%) 14 (0.97%)  Smoking status in the CCHS sub-cohort CCHS provided relevant information regarding the smoking status of the surveyed subjects making possible to calculate the number of pack-years. In order to do that, the variable SMKADSTY (Type of smoker – derived variable) was used as a start point. This variable classifies people in 7 non overlapping categories: (1) daily smoker, (2) occasional smoker (former daily smoker), (3) always an occasional smoker, (4) former daily smoker, (5) former occasional smoker, (6) never smoked, (7) not stated. For each of these categories (except for people that did not state their status), a different algorithm was used to determine the number of pack-years. Also, to simplify the analysis and to create classes of smokers with sufficient number of subjects, a new categorical variable was created with the purpose of analyzing the correlation between road proximity, a categorical variable, and smoking. The new categorical variable (SMOKING) has three classes: (1) current smokers (current daily smokers, always occasional smokers and occasional smokers – former daily smokers), (2) former smokers (former daily smokers and former occasional smokers), and (3) subjects that never smoked. In Table 7 are presented the overall summary statistics for the number of pack years and also the summary statistics grouped by SMOKING class.  Table 7. Summary statistics regarding the number of pack years Category N N Miss 5th Ptcl. Lower quartile Mean Median Upper quartile 95th Ptcl. Std Dev Missing 1 1 0 0 0 0 0 0 . Current smokers 226 0 1.75 17.50 30.98 27.87 44.10 66.00 20.74 Former smokers 760 0 1.82 1.82 27.28 18.87 40.25 85.95 30.54 Non smokers 454 0 0 0 0 0 0 0 0 Overall 1442 1 0.00 0.00 19.24 6.50 31.00 66.00 27.05 Analysis of the relationship between smoking status and health outcomes Because for the main analyses there wasn’t available information at individual level on the smoking status that could be used to better assess the effects of traffic pollution on health, an analysis was conducted using the CCHS data to see the impact of smoking on the CVD outcomes on the population of this survey.  Proc FREQ was used to analyze the relation between the declared health status for several health outcomes (ACS, CCS, and CHF) and smoking status (ever smoked/never smoked).  33 Table 8 presents the odds ratios of being sick (self reported and administratively derived health outcomes), for the subjects that ever smoke as opposed to subjects that never smoked.  Another analysis using Proc FREQ was done to determine weather or not there is a trend of having some of the health outcomes of interest present when comparing non-smokers with occasional smokers and daily smokers. Tables 9 and 10 depict for every one of the health outcomes of interest (self reported and derived from administrative data) the trend in the proportion of adverse effects due to increase in the smoking levels, Somers’ D C|R statistics’ 95% CI (Somers’ D C|R statistic measures the association treating the column variable (Health outcome) as the response and the row variable (Smoking frequency) as a predictor.  A strong positive association exists when the   asymptotic 95% confidence limits do not contain zero.), and the Cochran-Armitage test (The small left-sided p-values for the Cochran-Armitage test indicate that the probability of the Column 1 level (Health outcome='No') decreases as smoking frequency increases or, equivalently, that the probability of the Column 2 level (Health outcome='Yes') increases as smoking frequency increases. The two-sided p-value tests against either an increasing or decreasing alternative. This is an appropriate hypothesis when one wants to determine whether the tested treatment has progressive effects on the probability of adverse effects but the direction is unknown.  Although the tests were not significant for any of the three health outcomes, a more consistent trend, at least for non-smokers and occasional smokers emerged for the administratively derived health outcomes. The inconsistency in the trend for current smokers might be due to the much smaller number of current smokers.  Table 8. Estimates of odds ratio of having a self reported/administratively derived health outcome for subjects that ever/never smoked Odds ratio & 95% CI Health outcome Self reported health outcome Administratively derived health outcome Acute Coronary Syndrome 1.34 (0.77 – 2.36) 1.83 (1.09 – 3.07) Chronic Coronary Syndrome 0.88 (0.52 – 1.47) 1.15 (0.83 – 1.60) Congestive Heart Failure 0.77 (0.39 – 1.55) 1.37 (0.75 – 2.48)   Table 9. Trend analysis of having a self reported health outcome for subjects with different levels of smoking Health outcome Proportion of adverse effects for non- smokers, occasional smokers and daily smokers Somers’ D C|R statistics’ 95% CI Cochran-Armitage test (one sided/two sided) ACS 3.74%, 5.92%, 1.77% -0.0183 – 0.0129 0.3509 / 0.6392 CCS 5.07%, 4.87%, 3.10% -0.0275 – 0.0089 0.1802 / 0.3498 CHF 2.86%, 2.76%, 0.44% -0.0232 – 0.0012 0.0624 / 0.1240    34 Table 10. Trend analysis of having an administratively derived health outcome for subjects with different levels of smoking Health outcome Proportion of adverse effects for non- smokers, occasional smokers and daily smokers Somers’ D C|R statistics’ 95% CI Cochran-Armitage test (one sided/two sided) ACS 4.19%, 8.82%, 2.65% -0.0113  -   0.0237 0.4306 / 0.8094 CCS 12.56%, 15.92%, 8.41% -0.0374   -  0.0195 0.2351 / 0.4563 CHF 3.30%, 5.13%, 2.21% -0.0150  -   0.0156 0.46890 / 0.9211 Correlations between individual CCHS variables and census SES and pollution related variables Pearson correlations were calculated between the continuous CCHS and Census SES variables while Spearman correlations were calculated between the categorical CCHS and traffic quartiles of exposure as well as with the road proximity data (0/1). Spearman correlations were also calculated between the above mentioned categorical CCHS variables and the quintiles of Census SES variables. Correlations with traffic pollution and road proximity Traffic pollution Although there were several statistically significant correlations between traffic pollution data and CCHS variables, the highest correlation coefficient in absolute value was only  0.1454, while the majority of coefficient of correlations were in the 10-2 order of magnitude. This indicates that there are no correlations between traffic pollution and individual CCHS variables. The greater correlations were found between individual and family income variables and traffic generated pollutants, and these were found to be inverse correlations.  Road proximity Although there were several statistically significant correlations between road proximity data and CCHS variables, the majority of correlation coefficients were in the 10-2 order of magnitude with some coefficients being in the 10-3 order of magnitude. This indicates that there are no correlations between traffic pollution and individual CCHS variables. Correlations with dissemination area level SES While the expectation for the pollution and CCHS variables was that there will be little or no correlation, the expectation was that the CCHS individual data and Census SES data were correlated, not only between similar variables (i.e. CCHS personal income/Census SES personal income), but also between smoking, drinking, BMI, physical activity index and some of the Census SES variables.  This expectation was not confirmed by the correlation analyses performed with the continuous and categorical data. The highest correlation was found to be 0.3097, between CCHS family income and Census SES family income. There was little or no correlation found between smoking, physical activity, drinking and eating habits and any of the Census SES variables.  While these results preclude the use of some of the Census SES variables as proxies for lifestyle indicators, it is important to remark that the correlation analysis between the  35 individual CCHS variables indicated that there is no correlation between smoking, drinking, eating and activity habits and income or education for instance. Additional analyses Exposure data and smoking & drinking Although the correlation analyses have shown that there is little or no correlation between traffic pollutants and road proximity and the individual level variables from CCHS dataset (i.e. smoking status, income, alcohol consumption), additional analyses were performed to ascertain this lack of relation. Two additional analyses were thus carried: one consisted in performing analyses of variance and comparing the exposure means to various pollutants between various smoking categories; the other method consisted in producing box plots for exposure by categories of smoking.  All ANOVA analyses (done using PROC GLM in SAS 9.1 - to account for potential unbalanced data) were not significant (including tests for mean differences). A similar ANOVA analysis was performed for drinking category and almost similar results were obtained for most of the pollutants with the exception of ambient PM10, where differences were detected between people that never drank and regular drinkers in terms of ambient PM10 exposures. Also differences were found between former drinkers and regular drinkers in terms of traffic generated black carbon and PM2.5 exposures. Cohort summary statistics There were 346,536 residents of the greater Vancouver metropolitan area in the cohort. The sex and age characteristics of the cohort and the number of cases in each stratum are presented in Table 11.  Table 11. Age, sex and health outcome summaries SEX Health outcome rates (per 1000) Age by 10 years classes Female freq. (%) Male freq. (%) Total freq. (%) ACS CCS CHF 75 and over 18,092 (5.2) 11,385 (3.3) 29,477 (8.5) 29.3 21.6 13.5 65 - 74 30,757 (8.9) 25,192 (7.3) 55,949 (16.1) 18.1 19.9 4.3 55 - 64 49,859 (14.4) 46,024 (13.3) 95,883 (27.7) 9.9 12.7 1.1 45 -54 86,560 (25.0) 78,667 (22.7) 165,227 (47.7) 4.7 5.5 0.3 Total 185,268 (53.5) 161,268 (46.5) 346,536 100.0) 10.4 11.2 2.3  In the greater Vancouver metropolitan area baseline cohort during the follow-up period, the total number of hospitalizations or deaths for each of the three health outcomes of interest was: ACS – 3,588; CCS – 3,878; CHF – 794. From the total number of health outcomes of interest, the deaths due to one of the three health outcomes of interest were as follows: ACS – 594; CCS – 475; CHF – 80.   36 Tables 12 and 13 present the cohort summaries for the socioeconomic variables at the DA and neighborhood levels that are used in the two sets of Cox proportional hazards analyses. These summaries give an idea of the distribution of each of the variables describing the cohort. It can be seen from the two tables that, at least for the common variables between the levels, there is attenuation in the magnitude of the statistics from the DA level to the neighborhood level. Also Tables 12 and 13 present the summaries for each of the 20 SES variables (10 at the DA level and 10 at the neighborhood level). In both sets of analyses performed in this study, the SES variables are categorized by quintiles and these categories were actually used in the analyses. Particularly in the second set of analyses, only the lower and higher strata of each variable were used to compare the effects of pollution on the cardiovascular health outcomes. As was the case for the overall numbers pertaining to each SES variable in both sets of analyses, the summary statistics for quintiles for DA-level variables have a smaller minimum and a greater maximum than the quintiles of the variables at the neighborhood level of aggregation. However, the number of subjects corresponding to each quintile for the similar variables at the DA and neighborhood levels of aggregation was very similar.  Table 12. Summary statistics for the DA-SES variables form Stats Canada 2001 Census used in the Cox model SES Indicator (Stats Can Variable) Count Mean Maximum Minimum Std Dev Percentage of Chinese visible minorities (VIS_CHINESE) 346,536 16 89 0 19 Average 2000 total personal income ($) (INCOME) 346,536 33,020 187,691 9,087 13,304 Percent of people with university degree (UNIVERSITY) 346,536 33 99 0 15 Percent of people that use transit, walk or bike for work (TRANSPORTATION) 346,536 9 48 0 7 Coefficient of variation of income in population over 15 years (INCOME_VAR) 346,536 9 47 0 4 Percent of owned dwellings (OWNED_HOMES) 346,536 69 100 0 24 Average 2000 family income ($) (FAM_INCOME) 346,536 74,122 543,603 0 33,716 Employment rate (%) (EMPLOYMENT) 346,536 63 95 3 12 % people in labor working in management (MANAGEMENT) 346,536 12 67 0 7 Incidence of low income in 2000 % (LOW_INCOME) 346,536 18 94 0 13         37 Table 13. Summary statistics for the Neighbourhood SES variables from the B.C. Atlas of Child Development in the Cox Model VARIABLE Count Mean Maximum Minimum Std Dev % of total population whose home language is neither English nor French (OTHLANG) 346,536 13 49 0 10 % of total population without knowledge of English or French (LINGISOL) 346,536 4 26 0 4 % of total population (>=20 years of age) with any university degree (UNIVERSITY) 346,536 22 70 5 11 Seasonally adjusted unemployment rate among persons aged 25 years and over (UNEMPLOYMENT) 346,536 6 21 2 3 Median annual family income ($) (FAM_INCOME) 346,536 60,740 102,951 26,971 14,534 Average annual employment income ($) (INCOME) 346,536 34,792 69,604 19,730 8,569 % of aggregate neighbourhood income from any government transfer (TRANSFERS) 346,536 10 41 4 4 % of persons in households below the low-income cut-off (LICO) (LOW_INCOME) 346,536 19 65 4 9 % of occupied dwellings that are owner-occupied (OWNED_HOMES) 346,536 66 92 9 18 % of families spending 30% or more of income on shelter costs (STRESS) 346,536 29 54 16 6 Exposure related summaries Traffic exposure All correlations between traffic pollutants were positive, with a relatively high correlation between NO and NO2 (r = 0.54). There was a weak correlation between PM2.5 and all the other pollutants, while the correlations between black carbon and NO and NO2 were moderately high. From a longitudinal perspective, in all pollutants there was a weaker correlation between the more distant monthly windows of exposure and the more proximal ones.  Table 14. Traffic pollutants: summary statistics Pollutant Mean Minimum Maximum STD Inter-quartile range Traffic NO (μg/m3) 30.90 0.4 206.6 19.09 23.2 Traffic NO2  (μg/m3) 31.11 0.6 66.8 9.44 10.2 Traffic PM2.5   (μg/m3) 4.05 0.0 12.2 1.81 1.8 Traffic Black Carbon (B.C.) (10-5/m filter absorbance) 1.51 0.0 6.2 1.24 0.8        38  Table 15. Correlations between traffic pollutants Variable NO NO2 PM2.5 B.C NO 1.0000 0.54 NO2 <0.0001 1.0000  0.09 0.40 PM2.5 <0.0001 <0.0001 1.0000 0.30 0.19 0.16 B.C. <0.0001 <0.0001 <0.0001 1.0000 Road proximity Table 16 presents a summary of the total numbers of subjects living in the proximity of each of the five road types. It is necessary to mention that the number of subjects living in the proximity of roads was derived only from the first month of follow-up, namely January 1999, while the health outcomes were derived for the whole follow-up period. In the analysis, road proximity is a time-dependent variable.  Table 16. Percentage of subjects in the CVD cohort living in the proximity to roads Road proximity type In Road proximity Subjects living within 50 m from expressways and primary highways 1.73% Subjects living between 50 and 150 m from expressways and primary highways 4.92% Subjects living within 50 m from secondary highways and major roads 10.62% Subjects living between 50 and 150 m from secondary highways and major roads 21.78% Subjects living within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads 16.93%    Table 17. Traffic exposure and relative risk for CCS health outcomes Traffic exposure RR of health outcome for subjects in the 4th vs. 1st quartiles of traffic pollution exposure NO 0.90 (0.82 – 0.92) NO2 0.86 (0.78 – 0.94) Black Carbon 1.10 (1.00 – 1.20) PM2.5 0.92 (0.84 – 1.01)                    * Note: exposure is determined based on the first month of follow-up, January 1999          39 Table 18. Road proximity* and relative risk for CCS health outcomes Road proximity RR of health outcome for subjects in road proximity vs. subjects not in road proximity Subjects living within 50 m from expressways and primary highways (R-I) 1.21 (0.97 – 1.51) Subjects living between 50 and 150 m from expressways and primary highways (R-II) 1.06 (0.92 – 1.22) Subjects living within 50 m from secondary highways and major roads (R-III) 1.01 (0.92 – 1.12) Subjects living between 50 and 150 m from secondary highways and major roads (R-IV) 1.05 (0.98 – 1.13) Subjects living within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (R-V) 1.05 (0.96 – 1.14)    * Note: exposure is determined based on the first month of follow-up, January 1999  The main analysis employed to assess the effect of pollution on different cardiovascular morbidities and mortality was done using Cox proportional hazards regression, with pollution as a time-dependent covariate (thus varying on a monthly basis over the four years of the follow-up period, between January 1999 and December 2002). However, Table 17 only shows the distribution of the events in the CCS morbidities and mortalities that occurred during the whole follow-up period for the four traffic pollutants, when considering only the first month of follow-up, January 1999. This exposure represents the yearly average pollution exposure that a person in the cohort was subjected to between January 1998 and December 1998.  In Table 17, the crude relative risk on the occurrence of morbidity or mortality of CCS for subjects in the 4th versus 1st quartile of particular traffic pollutants are presented. For NO, NO2, and PM2.5, the relative risk is smaller than one (between 0.86 and 0.92). However, the relative risk of CCS was larger than one in the case of black carbon [1.10 (CI: 1.00 – 1.20].  In Table 18, the crude relative risk was calculated comparing subjects that were in the proximity of a certain road type and those who were not. Relative risks were higher than one for all five road type categories in the case of CCS health outcomes but none was significantly higher than one. However, the crude relative risk of experiencing CCS for the subjects living within 50m of expressways and highways was high (1.21) and only marginally non-significant [CI: 0.97 - 1.51]. All these relative risk values give only an indication of what the calculated risk would be, because we have to also take in account the effect of age, gender and different SES covariates that will be used in the analyses. Tables 30 and 31 in the Appendix show the crude RRs for ACS and CHF health outcomes in relation with traffic pollution and road proximity.  Tables 37 to 50 in Appendix I present the counts for censored subjects (those without any morbidity or mortality due to one of the three cardiovascular health outcomes of interest) and those that experienced such an outcome (event) during the follow-up period for each of  40 the health outcomes of interest and the covariates at two levels of aggregation used in the analyses for all traffic pollutants as well as for road proximity analyses. The values for the four traffic pollutants differ slightly because there were differences in the number of total subjects that had available exposure information for a particular pollutant, while for road proximity, all numbers are the same regardless of the road type because all subjects had information regarding their residential address (which was all that was required for calculating the distance to a particular road type). Cox analysis Table 19 provides the results for the crude hazard ratio estimates for CCS health outcomes in relation with traffic exposure while Table 32 in the Appendix provides the results for the crude hazard ratio estimates for the ACS and CHF health outcomes in relation with traffic exposure. There are only few instances of a linear descending or ascending trend for any of the estimates. The hazard ratios start to be higher than one for black carbon and especially for PM2.5. For NO, NO2, and black carbon the results were non-significant and mostly without any particular trend, or with the opposite trend than expected. In the case of PM2.5 the hazard ratios for the CCS outcomes and all quartiles of exposure were greater than one. There is also a noticeable increasing trend in HRs for PM2.5 that goes from non –significant for the second and third quartiles and becomes significant for the forth quartile.  Table 20 provides the results for the adjusted hazard ratio estimates for the three health outcomes of interest in relation with traffic exposure. The two sets of results are for the analyses using the DA level covariates plus age and gender and for the neighborhood level covariates plus age and gender respectively, all in relation with traffic pollution exposure.  For the analyses done using DA level SES covariates (Table 20), all hazard ratio estimates are non significant and there were no striking patterns. The hazard ratio estimates for the analyses performed using neighborhood level SES covariates were almost all greater than the corresponding estimates obtained adjusting with the DA level covariates. This was opposite to what was expected, given some of the results in the literature (Laurent et al237) but somewhat explained by other arguments that refer though to higher levels of aggregation (Wilkinson and Pickett122, see Discussion Chapter) For NO, NO2, and PM2.5 there was an increasing linear trend in the hazard ratio estimates for CCS, trend that is better depicted in Figure 5.   Table 19. Crude hazard ratios for traffic pollutants in relation with CCS outcomes Crude HR and 95% CI Pollutant 1st 2nd 3rd 4th NO 1.00 0.99 (0.91 - 1.08) 0.89 (0.81 - 0.97) 0.94 (0.86 - 1.02) NO2 1.00 0.90 (0.83 - 0.98) 0.85 (0.78 - 0.93) 0.88 (0.80 - 0.96) PM2.5 1.00 1.07 (0.98 - 1.17) 1.07 (0.97 - 1.17) 1.12 (1.02 - 1.22) Black Carbon 1.00 1.02 (0.93 - 1.11) 1.01 (0.93 - 1.11) 0.95 (0.86 - 1.04)     41 Table 20. Hazard ratios for traffic pollutants adjusted for DA and neighborhood levels SES covariates in relation with CCS outcomes DA SES Adjusted1 HR and 95% CI Neighborhood SES Adjusted2 HR and 95% CI Pollutant 2nd 3rd 4th 2nd 3rd 4th NO 1.04 (0.95 - 1.13) 0.97 (0.88 - 1.07) 1.05 (0.95 - 1.16) 1.06 (0.97 - 1.16) 1.04 (0.94 - 1.15) 1.12 (1.01 - 1.24) NO2 0.96 (0.88 - 1.05) 1.01 (0.91 - 1.12) 0.99 (0.89 - 1.10) 1.00 (0.91 - 1.10) 1.11 (1.00 - 1.24) 1.13 (1.01 - 1.25) PM2.5 1.03 (0.95 - 1.13) 1.04 (0.95 - 1.14) 1.05 (0.96 - 1.15) 1.05 (0.96 - 1.16) 1.08 (0.98 - 1.20) 1.10 (1.00 - 1.21) Black Carbon 1.05 (0.95 - 1.15) 1.08 (0.98 - 1.19) 1.03 (0.93 - 1.14) 1.05 (0.96 - 1.16) 1.11 (1.00 - 1.22) 1.05 (0.95 - 1.16) 1 The adjustment was done for sex, age class, and 10 DA level SES covariates; 2 The adjustment was done for sex, age class, and 10 Neighborhood level SES covariates; SES variables grouped in quintiles   Figure 5. Adjusted hazard ratios for traffic pollutants and CCS in conjunction with neighborhood level SES  Table 21 presents the crude and adjusted (with both DA level and neighborhood level SES covariates) hazard ratio estimates for CCS health outcomes due to proximity to different road types. As was suggested by the relative risk estimates in Table 20, all hazard ratio estimates (crude and adjusted) for the CCS outcomes due to proximity to roads are greater than one, with many significantly so. Also, as for the HR estimates from traffic pollution  42 analyses, the estimates derived by employing neighborhood level SES covariates are larger than when employing DA level SES covariates.  The HR estimates for CCS health outcomes show a definite decreasing trend from subjects living in the closer proximity of expressways (R-I: within 50 m from expressways and primary highways) towards subjects living further away from secondary roads (R-IV: between 50 and 150 m from secondary highways and major roads) only when employing neighborhood level SES covariates. In the case of DA level covariates there are no significant HRs and no clear trend from close proximity/high traffic roads to further away/less traffic roads.  The hazard ratios for traffic pollutants adjusted for DA and neighborhood level covariates in relation with ACS and CHF health outcomes are found in table 33 in the Appendix while the crude and adjusted (for DA and neighborhood level covariates) hazard ratios for road proximity in relation with ACS and CHF health outcomes are found in table 34 in the Appendix.  Table 21. Crude and adjusted hazard ratios for road proximity adjustment done using SES covariates at different levels of aggregation  1 The adjustment was done for sex, age class, and 10 DA level SES covariates; 2 The adjustment was done for sex, age class, and 10 Neighborhood level SES covariates; SES variables grouped in quintiles Stratified Cox analyses for the low and high levels of various SES covariates As was described in the methodology section, in order to assess the modifier effect of pollution on the SES covariates in relation to CCS, a series of stratified analyses were conducted. The Cox proportional hazards analyses were conducted for the lowest and highest quartile of subjects of each SES variable available for the study, for both levels of aggregation. Each analysis was thus adjusted for sex, age and the pollutant of interest. No other SES variable was included in the analyses in order to assess independently each SES variable and because of the significant correlation between many of the covariates.  Analyses HR and 95% CI Pollutant Crude HR DA SES Adjusted1 HR Neighborhood SES Adjusted2 HR Within 50 m from expressways and primary highways (R-I) 1.27 (1.02 - 1.57) 1.08 (0.87 - 1.34) 1.46 (1.18 - 1.80) Between 50 and 150 m from expressways and primary highways (R-II) 1.07 (0.93 - 1.23) 1.04 (0.90 - 1.19) 1.12 (0.97 - 1.29) Within 50 m from secondary highways and major roads (R-III) 1.09 (0.99 - 1.20) 1.03 (0.93 - 1.14) 1.16 (1.05 - 1.28) Between 50 and 150 m from secondary highways and major roads (R-IV) 1.03 (0.96 - 1.11) 1.06 (0.98 - 1.15) 1.01 (0.93 - 1.09) Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (R-V) 1.11 (1.03 - 1.21) 1.04 (0.96 - 1.13) 1.20 (1.10 - 1.30)  43 Tables 22 to 25 present the results of these stratified analyses for CCS health outcomes, all traffic pollutants and road proximity categories and both sets of SES covariates, at DA level of aggregation and neighborhood level of aggregation respectively. These tables present the hazard ratio estimates (and their 95% confidence intervals) of experiencing CCS for different traffic exposure levels or road proximities. For these particular analyses, the main focus is not necessarily the magnitude and significance of each individual estimate, but rather the relative difference between the low level SES estimate and high level SES estimate for the same pollution level. A divergent trend with HR estimates without overlapping confidence intervals would maximally indicate the effect modification induce by traffic pollution/road proximity on the SES in conjunction with CCS morbidity and mortality. Traffic related health outcomes DA level covariates Following the literature review performed on the relation between socioeconomic factors and cardiovascular diseases and prior to the start of the stratified analyses for low/high levels of the socioeconomic variables, assumptions were made regarding the behaviour of the HRs for each of the variable investigated. Thus, for the percentage of Chinese minority population, university education, employment, family and personal income, home ownership, and for the percentage of people working in management, the assumption was that the HRs will be higher for the low levels of these variables compared with the high levels. Opposite to this assumption, the assumption for the transportation means variable, low levels of income, and income variation variable was that at higher levels there will be larger HRs compared with the low levels of these variables.  Table 22 presents the comparison between HR estimates for CCS outcomes in conjunction with traffic pollution for the low and high levels of the ten available SES variables at DA level of aggregation. Appendix II also presents in a graphic format the results from Table 22 as well as the results for ACS and CHF health outcomes derived from Tables 31 and 33 in the Appendix. The expectation was that the results will be similar for ACS and CCS, but this was not necessarily the case. Similar results were obtained only for the education variable (for B.C.), for family income (for NO), for transportation (for PM2.5), for low income (for NO and B.C.), for home ownership (for B.C.), income variation (for B.C.), and for the work type variable (for NO). There were few circumstances where the HR estimates were significantly higher than unity, and although not many, there were definitely more than for ACS outcomes. For the analyses done based on the percentage of Chinese population in the DA of residence grouping, almost all HR estimates were smaller than the unit. Only for black carbon and particulate matter traffic pollution exposure the HR estimates are clearly and consistently higher for the group of subjects living in areas with a low percentage of Chinese minorities compared with the subjects living in DAs with a higher percentage of Chinese people.  The analyses showed that for each of these variables, at least for one of the traffic pollutants investigated, but in many cases for three or all four of them, there were differences between the low and high levels of the covariates. However, the expected behaviour of the HRs for the low/high levels of the SES covariates did not always matched actual results.   44 Table 22. Comparison between traffic pollution HR for low and high levels of DA- SES variables, when considering CCS health outcomes Health Outcome CCS Pollutant Quartile DA-level SES Pollutant SES Level 1st Q 2nd Q 3rd Q 4th Q Low 1.00 0.97 (0.83 - 1.14) 0.91 (0.76 - 1.09) 0.90 (0.73 - 1.11) NO High 1.00 0.88 (0.64 - 1.19) 0.78 (0.58 - 1.05) 0.98 (0.74 - 1.29) Low 1.00 0.87 (0.74 - 1.01) 0.98 (0.81 - 1.19) 0.74 (0.59 - 0.92) NO2 High 1.00 0.86 (0.6 - 1.23) 0.92 (0.67 - 1.27) 0.90 (0.64 - 1.25) Low 1.00 1.05 (0.88 - 1.25) 0.96 (0.79 - 1.16) 1.10 (0.92 - 1.32) Black Carbon High 1.00 0.93 (0.73 - 1.18) 0.92 (0.73 - 1.17) 0.99 (0.79 - 1.24) Low 1.00 0.97 (0.81 - 1.16) 1.02 (0.85 - 1.22) 0.94 (0.77 - 1.15) Chinese population PM2.5 High 1.00 0.83 (0.63 - 1.10) 0.77 (0.59 - 1.00) 0.92 (0.71 - 1.18) Low 1.00 0.98 (0.84 - 1.16) 1.03 (0.87 - 1.23) 0.93 (0.76 - 1.14) NO High 1.00 1.11 (0.88 - 1.39) 0.97 (0.77 - 1.22) 1.18 (0.96 - 1.47) Low 1.00 0.85 (0.72 - 1.00) 0.94 (0.78 - 1.14) 0.94 (0.78 - 1.12) NO2 High 1.00 0.85 (0.67 - 1.09) 0.96 (0.77 - 1.20) 1.01 (0.80 - 1.26) Low 1.00 1.10 (0.90 - 1.33) 1.06 (0.87 - 1.29) 1.17 (0.97 - 1.42) Black Carbon High 1.00 0.87 (0.69 - 1.09) 0.97 (0.79 - 1.20) 1.10 (0.90 - 1.35) Low 1.00 0.89 (0.72 - 1.08) 1.02 (0.84 - 1.23) 1.02 (0.83 - 1.24) University PM2.5 High 1.00 1.02 (0.81 - 1.29) 1.20 (0.98 - 1.48) 0.91 (0.74 - 1.13) Low 1.00 0.88 (0.72 - 1.08) 0.83 (0.68 - 1.01) 0.80 (0.66 - 0.98) NO High 1.00 1.13 (0.93 - 1.37) 1.05 (0.86 - 1.28) 0.96 (0.78 - 1.18) Low 1.00 0.82 (0.67 - 1.00) 0.73 (0.6 - 0.88) 0.69 (0.58 - 0.84) NO2 High 1.00 1.03 (0.86 - 1.24) 1.05 (0.86 - 1.29) 0.94 (0.77 - 1.16) Low 1.00 0.96 (0.78 - 1.18) 0.96 (0.78 - 1.17) 1.02 (0.84 - 1.25) Black Carbon High 1.00 1.05 (0.86 - 1.27) 0.98 (0.80 - 1.20) 1.00 (0.81 - 1.23) Low 1.00 0.94 (0.76 - 1.16) 0.92 (0.75 - 1.14) 0.90 (0.74 - 1.11) Employment PM2.5 High 1.00 1.12 (0.92 - 1.37) 1.12 (0.91 - 1.37) 0.99 (0.80 - 1.23) Low 1.00 0.80 (0.65 - 0.98) 0.67 (0.55 - 0.82) 0.72 (0.60 - 0.88) NO High 1.00 1.05 (0.86 - 1.29) 1.15 (0.94 - 1.40) 1.12 (0.90 - 1.40) Low 1.00 0.86 (0.70 - 1.07) 0.64 (0.52 - 0.79) 0.71 (0.59 - 0.85) NO2 High 1.00 0.93 (0.77 - 1.13) 0.95 (0.77 - 1.15) 0.92 (0.72 - 1.18) Low 1.00 0.96 (0.76 - 1.21) 0.94 (0.75 - 1.17) 0.95 (0.77 - 1.18) Black Carbon High 1.00 0.94 (0.78 - 1.14) 0.91 (0.73 - 1.12) 1.03 (0.84 - 1.27) Low 1.00 0.83 (0.65 - 1.04) 0.84 (0.67 - 1.05) 0.80 (0.64 - 1.00) Family income PM2.5 High 1.00 0.96 (0.79 - 1.17) 1.09 (0.90 - 1.33) 0.80 (0.63 - 1.01) Low 1.00 0.82 (0.66 - 1.02) 0.64 (0.52 - 0.80) 0.72 (0.59 - 0.88) NO High 1.00 1.02 (0.83 - 1.26) 1.05 (0.85 - 1.29) 1.24 (1.01 - 1.52) Low 1.00 0.78 (0.62 - 0.98) 0.61 (0.49 - 0.75) 0.65 (0.53 - 0.80) NO2 High 1.00 0.94 (0.77 - 1.14) 0.99 (0.81 - 1.21) 1.08 (0.86 - 1.35) Low 1.00 0.92 (0.73 - 1.17) 0.94 (0.76 - 1.17) 0.94 (0.76 - 1.16) Black Carbon High 1.00 0.98 (0.81 - 1.20) 0.95 (0.77 - 1.18) 1.12 (0.92 - 1.37) Low 1.00 0.88 (0.68 - 1.14) 0.85 (0.66 - 1.10) 0.91 (0.71 - 1.17) Personal income PM2.5 High 1.00 0.94 (0.77 - 1.15) 1.16 (0.95 - 1.41) 0.91 (0.73 - 1.14)   45 Table 22. Comparison between traffic pollution HR for low and high levels of DA- SES variables, when considering CCS health outcomes (cont.) Health Outcome CCS Pollutant Quartile DA-level SES Pollutant SES Level 1st Q 2nd Q 3rd Q 4th Q Low 1.00 0.99 (0.83 - 1.17) 1.06 (0.88 - 1.28) 0.95 (0.78 - 1.16) NO High 1.00 1.30 (0.91 - 1.85) 0.98 (0.69 - 1.38) 1.18 (0.85 - 1.65) Low 1.00 0.88 (0.75 - 1.04) 0.86 (0.69 - 1.06) 1.05 (0.84 - 1.32) NO2 High 1.00 1.18 (0.73 - 1.90) 1.01 (0.65 - 1.59) 1.11 (0.71 - 1.72) Low 1.00 0.96 (0.81 - 1.15) 1.11 (0.92 - 1.36) 1.10 (0.91 - 1.33) Black Carbon High 1.00 1.11 (0.82 - 1.50) 1.11 (0.84 - 1.47) 1.14 (0.87 - 1.51) Low 1.00 1.25 (1.04 - 1.50) 1.15 (0.94 - 1.40) 1.35 (1.10 - 1.65) Transportation PM2.5 High 1.00 1.02 (0.78 - 1.32) 1.02 (0.79 - 1.33) 0.81 (0.63 - 1.05) Low 1.00 1.09 (0.91 - 1.30) 1.14 (0.95 - 1.38) 1.12 (0.91 - 1.38) NO High 1.00 0.74 (0.59 - 0.93) 0.65 (0.52 - 0.81) 0.72 (0.59 - 0.88) Low 1.00 1.00 (0.85 - 1.18) 1.18 (0.97 - 1.44) 0.85 (0.65 - 1.10) NO2 High 1.00 0.72 (0.56 - 0.93) 0.62 (0.49 - 0.78) 0.69 (0.56 - 0.87) Low 1.00 1.19 (1.00 - 1.42) 1.11 (0.90 - 1.36) 1.06 (0.87 - 1.30) Black Carbon High 1.00 0.89 (0.69 - 1.14) 0.95 (0.76 - 1.18) 0.98 (0.79 - 1.22) Low 1.00 1.08 (0.90 - 1.29) 1.11 (0.92 - 1.35) 1.02 (0.81 - 1.29) Low income PM2.5 High 1.00 0.94 (0.73 - 1.21) 1.01 (0.79 - 1.29) 0.92 (0.73 - 1.17) Low 1.00 0.94 (0.73 - 1.21) 0.69 (0.54 - 0.89) 0.80 (0.64 - 1.01) NO High 1.00 0.97 (0.81 - 1.15) 1.02 (0.85 - 1.23) 0.98 (0.80 - 1.20) Low 1.00 0.71 (0.53 - 0.95) 0.65 (0.51 - 0.84) 0.71 (0.56 - 0.90) NO2 High 1.00 0.85 (0.72 - 1.00) 0.99 (0.82 - 1.20) 0.84 (0.66 - 1.07) Low 1.00 0.92 (0.71 - 1.20) 0.84 (0.67 - 1.07) 0.88 (0.70 - 1.11) Black Carbon High 1.00 1.24 (1.05 - 1.48) 1.13 (0.92 - 1.38) 1.31 (1.08 - 1.60) Low 1.00 0.83 (0.65 - 1.07) 1.01 (0.80 - 1.27) 0.82 (0.65 - 1.03) Home ownership PM2.5 High 1.00 1.03 (0.86 - 1.23) 1.13 (0.93 - 1.37) 1.09 (0.89 - 1.35) Low 1.00 1.02 (0.85 - 1.22) 0.93 (0.77 - 1.13) 0.92 (0.77 - 1.10) NO High 1.00 0.90 (0.73 - 1.11) 0.94 (0.76 - 1.16) 1.02 (0.83 - 1.26) Low 1.00 0.93 (0.77 - 1.13) 1.01 (0.83 - 1.22) 0.85 (0.71 - 1.01) NO2 High 1.00 0.78 (0.63 - 0.95) 0.81 (0.66 - 0.99) 0.84 (0.67 - 1.05) Low 1.00 1.23 (1.00 - 1.51) 1.08 (0.88 - 1.32) 1.23 (1.01 - 1.49) Black Carbon High 1.00 0.85 (0.69 - 1.03) 0.92 (0.75 - 1.13) 0.87 (0.71 - 1.07) Low 1.00 1.23 (1.00 - 1.51) 1.25 (1.01 - 1.54) 1.07 (0.88 - 1.32) Income variation PM2.5 High 1.00 0.96 (0.78 - 1.18) 1.04 (0.84 - 1.27) 0.96 (0.77 - 1.19) Low 1.00 0.76 (0.63 - 0.92) 0.67 (0.55 - 0.81) 0.64 (0.52 - 0.78) NO High 1.00 1.09 (0.90 - 1.33) 0.92 (0.75 - 1.13) 1.19 (0.98 - 1.44) Low 1.00 0.73 (0.60 - 0.89) 0.66 (0.54 - 0.80) 0.66 (0.55 - 0.80) NO2 High 1.00 0.86 (0.71 - 1.04) 0.95 (0.78 - 1.16) 0.94 (0.76 - 1.16) Low 1.00 0.98 (0.80 - 1.21) 1.00 (0.81 - 1.22) 0.86 (0.70 - 1.06) Black Carbon High 1.00 1.10 (0.90 - 1.33) 1.16 (0.95 - 1.42) 1.18 (0.97 - 1.43) Low 1.00 0.98 (0.79 - 1.21) 0.93 (0.75 - 1.15) 0.89 (0.71 - 1.11) Management PM2.5 High 1.00 1.33 (1.10 - 1.61) 1.33 (1.09 - 1.62) 1.01 (0.82 - 1.25) Thus, for the variable representing the percentage of Chinese origin people, the expected differences between the low/high levels of the variable materialized only for black carbon  46 and PM2.5 (low level corresponding to higher HR). These results were in spite of the fact that a larger proportion of subjects living in areas with a high percentage of Chinese population were expose to higher levels of pollution (Appendix I, Table 47 – e.g. for NO, 9,280 subjects in areas with a low percentage of Chinese population and high levels of pollution versus 26,781 subjects living in areas with a high percentage of Chinese population and high levels of pollution).  For the variable representing the percentage of university degrees in the area the expected differences between the low/high levels of the variable materialized only for black carbon (low level corresponding to higher HR). Although smaller in magnitude, a similar distribution was found for the subjects living in areas with high levels of university graduates and high pollution levels (Appendix I, Table 47 – e.g. for NO 20,684) versus subjects living in areas with low levels of university graduates and high pollution levels (Appendix I, Table 47 – e.g. for NO 10,405).  The variable representing income variation in the area showed also the expected results for NO2 particulate matter and black carbon, low levels of income variation corresponding to higher HR. In the case of this variable, the distribution of subjects on low/high levels of variable and pollutants was more in accord with the results, a smaller proportion of subjects living in areas with low income variation and high pollution (Appendix I, Table 47 – e.g. for NO 19,718) versus subjects living in areas with high income variation and high pollution (Appendix I, Table 47 – e.g. for NO 16,660). The only other result that confirmed the expectations was for the transportation variable and that only for nitrogen dioxide exposure (low level corresponding to low HRs). In the case of this variable, the distribution of subjects on low/high levels of variable and pollutants was also more in accord with the results, a smaller proportion of subjects living in areas with high usage of public transportation, etc. and high pollution (Appendix I, Table 47 – e.g. for NO2 35,389) versus subjects living in areas with low usage of public transportation, etc. and high pollution (Appendix I, Table 47 – e.g. for NO2 7,798).  Contrary to the assumptions made prior to the stratified analyses, the results for the rest of the variables showed opposite trends than expected. Thus, for the variables representing family and personal income as well as for the variable representing the percent of employment in the area, the results showed that for nitrogen oxides and particulate matter low level of family and personal income as well as low level of employment correspond to lower HR. All the above results were despite the fact that more subjects were from areas with low income (personal, family) or low employment and high levels of pollution as opposed to subjects from areas with high income (personal, family) or high employment and high levels of pollution.  For the variables representing percentage of home ownership in the area and the percentage of people working in management the results showed, contrary to the expectations, that for all four traffic pollutants the HRs corresponding to lower levels of these two variables were smaller than the HRs for the high levels of these two SES indicators. All the above results were despite the fact that more subjects were from areas with low home ownership or low management workers and high levels of pollution as opposed to subjects from areas with high home ownership or high percentage of workers in management and high levels of pollution. Also, for all four traffic pollutants investigated, low levels of low income were  47 associated with high HRs compared with subjects from areas with high levels, which in turn were associated with lower HRs.  The variable representing transportation by bus, bike, and walk was the only variable that had inconsistent results across pollutants, showing, as was described before, expected results for nitrogen dioxide, while for particulate matter HRs were found to be higher for subjects from areas with low percentages of people that used public transit, walked or biked to work. Neighborhood level covariates As for the DA-level covariates, assumptions were made regarding the behaviour of the HRs for each of the neighborhood level variable investigated. Thus, for the percentage of people with a second language at home, linguistic isolation, university education, family and personal income, and home ownership, the assumption was that the HRs will be higher for the low levels of these variables compared with the high levels. Opposite to this assumption, the assumption for the unemployment variable, low levels of income, governmental transfers variable and neighborhood stress variable was that at higher levels there will be larger HRs compared with the low levels of these variables.  Table 23 presents the HR estimates for the four traffic exposure pollutants stratified by the low and high strata of the ten available SES variables defined at neighborhood levels of aggregation. Appendix III also presents in a graphic format the results from Table 23 as well as the results for ACS and CHF health outcomes from Tables 32 and 34 in Appendix I. As opposed to the results at DA-level, there was more commonality between the results for CCS and ACS at neighbourhood level, with all variables showing comparative results at least for one of the traffic pollutants.  The analyses showed that for each of these variables, at least for one of the traffic pollutants investigated, but in many cases for three or all four of them, there were differences between the low and high levels of the covariates. However, the expected behaviour of the HRs for the low/high levels of the SES covariates did not always matched actual results.  Thus, the expected differences between the low/high levels of the variable materialized for all of the four traffic pollutants (low level corresponding to higher HR) only for the variables representing the use of a second language at home other than French or English and for the linguistic isolation variable. For the variable representing the percentage of university degrees in the area the results were inconclusive for all four traffic pollutants and thus the expected differences between the low/high levels of the variable could not be verified.  Contrary to the assumptions made prior to the stratified analyses, the results for the rest of the neighborhood level variables showed opposite trends than expected. Thus, for the variables representing family and personal income, the results showed that for all four traffic pollutants, low level of family and personal correspond to lower HR. Also for all four pollutants, the variable representing the percent of unemployment in the area showed opposite results than expected, lower levels of unemployment being associated higher HRs. These results are perfectly consistent with the results from the corresponding DA level   48 Table 23. Comparison between traffic pollution HR for low and high levels of Neighborhood-SES variables, when considering CCS health outcomes Health Outcome CCS Pollutant Quartile Neighborhood- level SES Pollutant SES Level 1st Q 2nd Q 3rd Q 4th Q Low 1.00 1.12 (0.95 - 1.33) 1.00 (0.84 - 1.19) 0.91 (0.75 - 1.11) NO High 1.00 0.82 (0.61 - 1.10) 0.53 (0.40 - 0.71) 0.76 (0.58 - 1.00) Low 1.00 1.06 (0.91 - 1.23) 1.05 (0.86 - 1.29) 0.74 (0.59 - 0.93) NO2 High 1.00 0.61 (0.44 - 0.85) 0.54 (0.41 - 0.72) 0.66 (0.50 - 0.87) Low 1.00 1.21 (1.02 - 1.45) 1.28 (1.05 - 1.55) 1.14 (0.95 - 1.36) Black Carbon High 1.00 0.79 (0.63 - 1.00) 0.77 (0.62 - 0.96) 0.82 (0.67 - 1.01) Low 1.00 1.11 (0.93 - 1.33) 1.28 (1.07 - 1.55) 1.03 (0.85 - 1.24) Other language PM2.5 High 1.00 0.77 (0.59 - 1.01) 0.66 (0.51 - 0.85) 0.86 (0.68 - 1.10) Low 1.00 1.11 (0.93 - 1.33) 1.05 (0.88 - 1.25) 0.97 (0.79 - 1.17) NO High 1.00 0.75 (0.55 - 1.02) 0.50 (0.37 - 0.68) 0.67 (0.51 - 0.89) Low 1.00 1.00 (0.85 - 1.18) 1.10 (0.91 - 1.33) 0.72 (0.58 - 0.90) NO2 High 1.00 0.59 (0.42 - 0.81) 0.50 (0.38 - 0.66) 0.58 (0.44 - 0.77) Low 1.00 1.05 (0.87 - 1.27) 1.16 (0.96 - 1.41) 1.13 (0.95 - 1.35) Black Carbon High 1.00 0.87 (0.68 - 1.11) 0.82 (0.64 - 1.04) 0.81 (0.64 - 1.03) Low 1.00 1.15 (0.96 - 1.37) 1.27 (1.05 - 1.54) 1.08 (0.89 - 1.31) Linguistic isolation PM2.5 High 1.00 0.77 (0.59 - 1.01) 0.62 (0.48 - 0.80) 0.75 (0.59 - 0.96) Low 1.00 0.98 (0.85 - 1.13) 0.99 (0.83 - 1.18) 0.86 (0.69 - 1.08) NO High 1.00 1.02 (0.81 - 1.29) 0.90 (0.72 - 1.13) 1.13 (0.92 - 1.39) Low 1.00 0.94 (0.81 - 1.08) 1.05 (0.84 - 1.31) 0.84 (0.69 - 1.04) NO2 High 1.00 0.87 (0.67 - 1.11) 0.92 (0.73 - 1.16) 1.02 (0.82 - 1.28) Low 1.00 1.14 (0.96 - 1.36) 1.02 (0.84 - 1.23) 1.04 (0.87 - 1.25) Black Carbon High 1.00 0.95 (0.75 - 1.20) 0.98 (0.79 - 1.21) 1.15 (0.94 - 1.41) Low 1.00 0.98 (0.81 - 1.18) 1.09 (0.90 - 1.31) 1.02 (0.84 - 1.24) University PM2.5 High 1.00 1.03 (0.82 - 1.30) 1.20 (0.98 - 1.46) 0.80 (0.65 - 0.99) Low 1.00 1.03 (0.86 - 1.24) 1.16 (0.97 - 1.39) 1.21 (1.00 - 1.47) NO High 1.00 0.91 (0.74 - 1.13) 0.65 (0.52 - 0.81) 0.76 (0.62 - 0.95) Low 1.00 1.04 (0.88 - 1.22) 1.31 (1.09 - 1.59) 0.95 (0.75 - 1.19) NO2 High 1.00 0.74 (0.59 - 0.94) 0.57 (0.46 - 0.72) 0.75 (0.62 - 0.91) Low 1.00 1.00 (0.83 - 1.21) 1.24 (1.03 - 1.49) 1.10 (0.91 - 1.33) Black Carbon High 1.00 0.91 (0.72 - 1.15) 0.85 (0.68 - 1.07) 0.89 (0.71 - 1.12) Low 1.00 1.15 (0.97 - 1.36) 1.18 (0.99 - 1.42) 1.13 (0.91 - 1.40) Unemployment PM2.5 High 1.00 1.08 (0.85 - 1.37) 0.90 (0.70 - 1.14) 0.87 (0.68 - 1.12) Low 1.00 0.83 (0.66 - 1.05) 0.54 (0.43 - 0.68) 0.67 (0.54 - 0.84) NO High 1.00 1.07 (0.89 - 1.30) 1.10 (0.91 - 1.34) 1.14 (0.92 - 1.42) Low 1.00 0.85 (0.66 - 1.10) 0.55 (0.43 - 0.69) 0.71 (0.58 - 0.88) NO2 High 1.00 0.96 (0.80 - 1.15) 1.11 (0.92 - 1.35) 0.99 (0.77 - 1.27) Low 1.00 0.94 (0.72 - 1.21) 0.92 (0.72 - 1.18) 0.89 (0.69 - 1.14) Black Carbon High 1.00 1.01 (0.84 - 1.23) 1.09 (0.89 - 1.34) 1.03 (0.84 - 1.26) Low 1.00 0.87 (0.66 - 1.15) 0.75 (0.57 - 0.98) 0.78 (0.60 - 1.03) Family income PM2.5 High 1.00 1.06 (0.87 - 1.29) 1.16 (0.96 - 1.41) 0.95 (0.75 - 1.21)    49 Table 23. Comparison between traffic pollution HR for low and high levels of Neighborhood-SES variables, when considering CCS health outcomes (cont.) Health Outcome CCS Pollutant Quartile Neighborhood- level SES Pollutant SES Level 1st Q 2nd Q 3rd Q 4th Q Low 1.00 0.77 (0.62 - 0.94) 0.53 (0.43 - 0.66) 0.58 (0.47 - 0.72) NO High 1.00 1.12 (0.90 - 1.38) 1.12 (0.91 - 1.38) 1.07 (0.86 - 1.33) Low 1.00 0.78 (0.63 - 0.97) 0.49 (0.40 - 0.61) 0.63 (0.52 - 0.77) NO2 High 1.00 0.85 (0.69 - 1.05) 0.95 (0.78 - 1.17) 0.86 (0.69 - 1.07) Low 1.00 0.90 (0.71 - 1.13) 0.79 (0.62 - 1.00) 0.82 (0.65 - 1.03) Black Carbon High 1.00 0.95 (0.76 - 1.18) 0.98 (0.80 - 1.20) 1.08 (0.89 - 1.30) Low 1.00 0.91 (0.71 - 1.18) 0.76 (0.59 - 0.97) 0.74 (0.57 - 0.96) Personal income PM2.5 High 1.00 1.11 (0.90 - 1.38) 1.26 (1.03 - 1.53) 0.88 (0.70 - 1.09) Low 1.00 0.97 (0.80 - 1.17) 0.97 (0.80 - 1.18) 0.83 (0.66 - 1.04) NO High 1.00 0.88 (0.72 - 1.07) 0.73 (0.60 - 0.89) 0.69 (0.56 - 0.85) Low 1.00 0.82 (0.68 - 0.99) 0.98 (0.80 - 1.19) 0.71 (0.55 - 0.90) NO2 High 1.00 0.84 (0.69 - 1.03) 0.59 (0.48 - 0.72) 0.68 (0.56 - 0.82) Low 1.00 1.14 (0.94 - 1.38) 1.01 (0.82 - 1.25) 1.07 (0.87 - 1.31) Black Carbon High 1.00 1.03 (0.79 - 1.33) 1.00 (0.78 - 1.28) 0.94 (0.73 - 1.21) Low 1.00 1.09 (0.89 - 1.34) 1.13 (0.93 - 1.39) 0.81 (0.64 - 1.01) Governmental transfers PM2.5 High 1.00 0.91 (0.72 - 1.14) 0.78 (0.62 - 0.98) 0.83 (0.66 - 1.05) Low 1.00 1.04 (0.88 - 1.24) 1.24 (1.04 - 1.48) 1.12 (0.91 - 1.38) NO High 1.00 0.99 (0.72 - 1.36) 0.73 (0.54 - 0.99) 0.95 (0.71 - 1.26) Low 1.00 0.95 (0.81 - 1.11) 1.37 (1.11 - 1.69) 0.90 (0.68 - 1.19) NO2 High 1.00 1.11 (0.73 - 1.70) 0.84 (0.57 - 1.24) 1.10 (0.76 - 1.61) Low 1.00 1.19 (1.00 - 1.40) 1.17 (0.96 - 1.43) 1.14 (0.95 - 1.37) Black Carbon High 1.00 0.98 (0.75 - 1.28) 0.96 (0.75 - 1.23) 1.08 (0.84 - 1.38) Low 1.00 1.13 (0.96 - 1.33) 1.21 (1.00 - 1.45) 1.21 (0.97 - 1.51) Low income PM2.5 High 1.00 0.79 (0.61 - 1.03) 0.78 (0.61 - 1.00) 0.77 (0.61 - 0.99) Low 1.00 0.98 (0.77 - 1.26) 0.80 (0.64 - 1.00) 0.91 (0.74 - 1.13) NO High 1.00 0.93 (0.79 - 1.10) 1.01 (0.83 - 1.23) 1.03 (0.82 - 1.30) Low 1.00 0.89 (0.66 - 1.19) 0.80 (0.61 - 1.03) 0.90 (0.71 - 1.14) NO2 High 1.00 0.86 (0.74 - 1.01) 1.29 (1.03 - 1.62) 0.74 (0.54 - 1.01) Low 1.00 1.07 (0.80 - 1.42) 1.03 (0.80 - 1.34) 1.06 (0.82 - 1.38) Black Carbon High 1.00 1.13 (0.96 - 1.34) 1.08 (0.88 - 1.33) 0.95 (0.78 - 1.16) Low 1.00 0.91 (0.72 - 1.15) 1.04 (0.82 - 1.31) 0.80 (0.64 - 1.00) Home ownership PM2.5 High 1.00 1.05 (0.88 - 1.25) 1.26 (1.04 - 1.52) 1.26 (1.00 - 1.58) Low 1.00 1.06 (0.88 - 1.28) 1.18 (0.98 - 1.42) 0.90 (0.73 - 1.11) NO High 1.00 0.90 (0.72 - 1.14) 0.69 (0.55 - 0.87) 0.82 (0.66 - 1.01) Low 1.00 0.92 (0.78 - 1.08) 1.08 (0.89 - 1.31) 0.72 (0.54 - 0.96) NO2 High 1.00 0.83 (0.64 - 1.08) 0.67 (0.52 - 0.87) 0.85 (0.68 - 1.06) Low 1.00 1.28 (1.07 - 1.52) 1.24 (1.01 - 1.52) 1.12 (0.91 - 1.37) Black Carbon High 1.00 0.90 (0.70 - 1.16) 0.91 (0.72 - 1.14) 0.95 (0.75 - 1.19) Low 1.00 1.17 (0.98 - 1.40) 1.15 (0.94 - 1.40) 1.01 (0.82 - 1.26) Neighborhood stress PM2.5 High 1.00 0.89 (0.69 - 1.14) 0.98 (0.77 - 1.25) 0.86 (0.68 - 1.08)   50 variables. For the variables representing percentage of home ownership in the area, contrary to the expectations, for particulate matter, the HRs corresponding to lower levels of this variable were smaller than the HRs for the high levels of this indicator. For the other three pollutants the results were inconclusive.  Also contrary to expectations, for nitrogen oxide, black carbon and particulate matter, low levels of low income, governmental transfer and neighborhood stress were associated with high HRs compared with subjects from areas with high levels for these three variables, which in turn were associated with lower HRs. These results were also consistent with the results for the corresponding variables at DA level.  The distribution of subjects on high/low classes of neighborhood level variables and high levels of pollution matched the distribution of the subjects for the similar variables at the DA-level.  Overall, the results were more distinct when using variables at neighborhood levels compared with the results at DA level, and by distinct I mean clearer (greater) differences between the two strata at all levels of pollution (excluding the benchmark, of course) and more pollutants exhibiting these differences. Road proximity related health outcomes CCS health outcomes considering DA level SES variables Table 24 presents the comparison between HR estimates for CCS morbidity and mortality outcomes in conjunction with road proximity for the low and high levels of the ten available SES variables at DA level of aggregation. Appendix IV also presents in a graphic format the results from Table 24 as well as the results for the ACS and CHF health outcomes from Table 35 in the Appendix. The graphs presented in Appendices IV and V refer only at road proximity type I (subjects living within 50 m from expressways and primary highways), road proximity type IV (between 50 and 150 m from secondary highways and major roads), and road proximity type V (within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads). This choice of only three road proximity types was made in order to create a better contrast between the road type I, considered as the road proximity with the highest levels of traffic exposures and road type IV, which can be considered the proximity with the lowest levels of traffic exposures. Road proximity type V was used as an intermediary between proximity I and proximity IV.  More variables show differences in the CCS health outcomes in conjunction with road proximity type compared with the ACS health outcomes. While for the percentage of people with university education, family income, employment rates, and income variation variables the results are mixed and inconclusive, certain trends are evident for the rest of the DA level variables. Thus, for subjects from areas with a low proportion of Chinese minorities as well as subjects from areas with low levels of low income are at higher risk of developing ACS in relation to traffic proximity. For Chinese minority variable, there is even a protective effect, especially if leaving within 50 m of expressways and highways. Otherwise, the manifested trends are decreasing from road proximity type I to road proximity type V and IV.   51 Table 24. Comparison between HR of different road proximity categories for low and high levels of DA-SES variables, when considering health outcomes Health Outcomes DA-level SES Pollutant SES Level CCS Low 1.44 (1.04 - 2.00) Within 50 m from expressways and primary highways (I)  High 0.71 (0.29 - 1.70) Low 1.30 (0.97 - 1.74) Between 50 and 150 m from expressways and primary highways (II) High 0.87 (0.62 - 1.24) Low 0.99 (0.80 - 1.21) Within 50 m from secondary highways and major roads (III) High 1.08 (0.84 - 1.39) Low 1.05 (0.89 - 1.24) Between 50 and 150 m from secondary highways and major roads (IV) High 0.98 (0.81 - 1.18) Low 1.17 (1.00 - 1.37) Chinese population Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 0.97 (0.78 - 1.20) Low 1.15 (0.83 - 1.60) Within 50 m from expressways and primary highways (I)  High 1.16 (0.60 - 2.24) Low 0.89 (0.68 - 1.17) Between 50 and 150 m from expressways and primary highways (II) High 1.05 (0.75 - 1.46) Low 1.19 (0.99 - 1.44) Within 50 m from secondary highways and major roads (III) High 1.04 (0.82 - 1.33) Low 1.02 (0.86 - 1.21) Between 50 and 150 m from secondary highways and major roads (IV) High 1.21 (1.02 - 1.42) Low 1.14 (0.98 - 1.33) University Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 1.03 (0.84 - 1.27) Low 1.22 (0.85 - 1.76) Within 50 m from expressways and primary highways (I)  High 1.15 (0.63 - 2.08) Low 0.88 (0.67 - 1.17) Between 50 and 150 m from expressways and primary highways (II) High 1.22 (0.88 - 1.68) Low 1.04 (0.86 - 1.26) Within 50 m from secondary highways and major roads (III) High 1.02 (0.80 - 1.30) Low 0.85 (0.72 - 1.00) Between 50 and 150 m from secondary highways and major roads (IV) High 1.06 (0.89 - 1.26) Low 1.02 (0.87 - 1.19) Employment Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 1.09 (0.89 - 1.32) Low 1.01 (0.71 - 1.42) Within 50 m from expressways and primary highways (I)  High 1.29 (0.67 - 2.48) Low 0.89 (0.70 - 1.14) Between 50 and 150 m from expressways and primary highways (II) High 0.95 (0.65 - 1.40) Low 1.17 (0.96 - 1.41) Within 50 m from secondary highways and major roads (III) High 0.97 (0.75 - 1.25) Low 0.98 (0.84 - 1.15) Between 50 and 150 m from secondary highways and major roads (IV) High 1.08 (0.91 - 1.29) Low 1.06 (0.91 - 1.23) Family income Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 0.96 (0.77 - 1.19) Low 1.19 (0.84 - 1.70) Within 50 m from expressways and primary highways (I)  High 1.28 (0.70 - 2.32) Low 0.82 (0.62 - 1.08) Between 50 and 150 m from expressways and primary highways (II) High 1.00 (0.70 - 1.44) Low 1.06 (0.86 - 1.31) Within 50 m from secondary highways and major roads (III) High 1.09 (0.85 - 1.39) Low 0.86 (0.72 - 1.02) Between 50 and 150 m from secondary highways and major roads (IV) High 1.14 (0.97 - 1.35) Low 1.02 (0.86 - 1.20) Personal income Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 1.05 (0.86 - 1.29)         52 Table 24. Comparison between HR of different road proximity categories for low and high levels of DA-SES variables, when considering health outcomes (cont.) Health Outcomes DA-level SES Pollutant Level CCS Low 0.97 (0.57 - 1.65) Within 50 m from expressways and primary highways (I)  High 1.33 (0.93 - 1.89) Low 1.02 (0.72 - 1.45) Between 50 and 150 m from expressways and primary highways (II) High 0.96 (0.74 - 1.25) Low 1.03 (0.83 - 1.28) Within 50 m from secondary highways and major roads (III) High 1.09 (0.89 - 1.35) Low 1.04 (0.86 - 1.24) Between 50 and 150 m from secondary highways and major roads (IV) High 1.10 (0.94 - 1.29) Low 1.04 (0.86 - 1.25) Transportation Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 1.07 (0.90 - 1.26) Low 1.60 (1.01 - 2.52) Within 50 m from expressways and primary highways (I)  High 0.97 (0.66 - 1.44) Low 0.85 (0.57 - 1.28) Between 50 and 150 m from expressways and primary highways (II) High 0.89 (0.68 - 1.17) Low 0.96 (0.76 - 1.21) Within 50 m from secondary highways and major roads (III) High 1.04 (0.84 - 1.28) Low 1.10 (0.92 - 1.31) Between 50 and 150 m from secondary highways and major roads (IV) High 1.00 (0.85 - 1.17) Low 1.00 (0.82 - 1.21) Low income Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 0.99 (0.83 - 1.17) Low 0.93 (0.63 - 1.36) Within 50 m from expressways and primary highways (I)  High 1.46 (0.93 - 2.30) Low 0.94 (0.73 - 1.21) Between 50 and 150 m from expressways and primary highways (II) High 1.12 (0.80 - 1.57) Low 1.10 (0.91 - 1.33) Within 50 m from secondary highways and major roads (III) High 1.10 (0.88 - 1.38) Low 1.02 (0.87 - 1.18) Between 50 and 150 m from secondary highways and major roads (IV) High 1.00 (0.83 - 1.20) Low 1.02 (0.87 - 1.19) Home ownership Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 1.16 (0.97 - 1.40) Low 1.24 (0.87 - 1.78) Within 50 m from expressways and primary highways (I)  High 1.06 (0.50 - 2.24) Low 1.07 (0.83 - 1.39) Between 50 and 150 m from expressways and primary highways (II) High 1.09 (0.77 - 1.54) Low 1.05 (0.86 - 1.28) Within 50 m from secondary highways and major roads (III) High 0.89 (0.70 - 1.14) Low 1.02 (0.86 - 1.21) Between 50 and 150 m from secondary highways and major roads (IV) High 1.12 (0.95 - 1.33) Low 1.08 (0.92 - 1.26) Income variation Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 0.94 (0.76 - 1.16) Low 1.19 (0.82 - 1.73) Within 50 m from expressways and primary highways (I)  High 1.46 (0.90 - 2.36) Low 0.92 (0.70 - 1.22) Between 50 and 150 m from expressways and primary highways (II) High 0.98 (0.72 - 1.34) Low 1.07 (0.86 - 1.32) Within 50 m from secondary highways and major roads (III) High 1.13 (0.90 - 1.41) Low 0.90 (0.76 - 1.07) Between 50 and 150 m from secondary highways and major roads (IV) High 1.15 (0.98 - 1.36) Low 1.03 (0.87 - 1.22) Management Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 1.10 (0.91 - 1.32)  The transportation and home ownership variables have a similar behaviour, the subjects living in the areas with a low proportion of transit or biking usage or walking, or areas with low levels of home ownership being at a low risk of experiencing CCS while subjects from the high level categories are at increasing risk, especially if living within 50 m of expressways or highways. Similar differences can be seen for personal income variable, where the biggest discrepancy between the low/high classes is for subjects living between 50 and 150 m from  53 secondary highways and major roads, while for the management variable the trends are parallel, with subjects from areas with a higher proportion of people working in management being at increased risk of experiencing CCS.  The same assumptions on the behaviour of SES variables that were considered for the traffic pollution were also considered in the case of road proximity. However, these assumptions were checked overall, along the gradient of exposure to traffic represented by the different categories of road proximity and also only for R-I, for subjects living within 50 m of expressways and highways.  When assessing the behaviour of SES variables overall, several of the indicators showed conclusive differences between the low and high levels of the variables and also along the exposure gradient. These variables were the percentage of Chinese population, personal income, transportation, low income, home ownership and the management variable. However, the assumptions posited for each of these variables were met only by the variable representing the percentage of Chinese population and the variable representing the percentage of people that use public transit, walk or bike to work. For the other four variables that showed conclusive differences between the low and high levels the actual results were contrary to the expected results. When looking only at subjects living within 50 m from expressways and highways (R-I), two more variables showed clear differences between the low/high levels. One variable was the family income, but in this case the expected result of high HRs for the low level category did not materialized while for the second variable, income variation, the actual differences matched the expected result, that is low income variation was associated with higher HRs compared with the high levels of income variation.  Table 49 in the Appendix presents the distribution of subjects by the two levels of socioeconomic indicators at DA-level of aggregation and the five road proximity categories. It can be seen from Table 49 that for the variables that showed clear differences between the low/high levels and were in agreement with the a priori assumptions made regarding their behaviour, the distribution of subjects by high/low SES levels and road proximity justify the results, while that is not the case for the variables that had opposite results than expected. CCS health outcomes considering neighborhood level SES variables  Table 25 presents the comparison between HR estimates for all cardiovascular morbidity and mortality outcomes in conjunction with road proximity for the low and high levels of the ten available SES variables at neighborhood level of aggregation. Appendix V also presents in a graphic format the results from Table 31 for CCS health outcomes as well as the results for ACS and CHF health outcomes presented in Table 36 in the Appendix. The results for unemployment rates, governmental transfers, home ownership and neighborhood stress are inconclusive. For a second language used at home, linguistic isolation and low income, subjects living in areas belonging to the lower classes of these covariates experience higher risks of CCS, especially if they live within 50 m of expressways and highways (for the low income covariate the trend lines are though parallel). The situation is reversed for university education, family and personal income where subject living in areas with high levels of university graduates or family and personal income experience increased  54 risks of CCS outcomes, especially for subjects living between 50 and 150 m from secondary highways and major roads.  Table 25. Comparison between HR of different road proximity categories for low and high levels of Neighborhood-SES variables, when considering health outcomes Health Outcomes Neighborhood- level SES Pollutant Level CCS Low 1.16 (0.78 - 1.71) Within 50 m from expressways and primary highways (I)  High 1.08 (0.63 - 1.83) Low 1.16 (0.85 - 1.58) Between 50 and 150 m from expressways and primary highways (II) High 0.79 (0.56 - 1.12) Low 1.02 (0.83 - 1.24) Within 50 m from secondary highways and major roads (III) High 0.95 (0.74 - 1.21) Low 1.00 (0.85 - 1.19) Between 50 and 150 m from secondary highways and major roads (IV) High 0.91 (0.76 - 1.09) Low 1.09 (0.92 - 1.28) Other language Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 0.86 (0.70 - 1.06) Low 1.18 (0.79 - 1.78) Within 50 m from expressways and primary highways (I)  High 0.80 (0.41 - 1.54) Low 1.18 (0.86 - 1.62) Between 50 and 150 m from expressways and primary highways (II) High 0.90 (0.65 - 1.25) Low 1.07 (0.87 - 1.31) Within 50 m from secondary highways and major roads (III) High 0.88 (0.68 - 1.13) Low 1.02 (0.87 - 1.21) Between 50 and 150 m from secondary highways and major roads (IV) High 0.92 (0.77 - 1.09) Low 1.16 (0.98 - 1.37) Linguistic isolation Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 0.84 (0.68 - 1.03) Low 0.97 (0.68 - 1.39) Within 50 m from expressways and primary highways (I)  High 1.35 (0.82 - 2.21) Low 0.97 (0.71 - 1.33) Between 50 and 150 m from expressways and primary highways (II) High 0.98 (0.72 - 1.32) Low 1.15 (0.95 - 1.38) Within 50 m from secondary highways and major roads (III) High 1.07 (0.86 - 1.34) Low 1.03 (0.87 - 1.23) Between 50 and 150 m from secondary highways and major roads (IV) High 1.17 (1.00 - 1.37) Low 1.09 (0.94 - 1.28) University Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 1.06 (0.88 - 1.27) Low 1.15 (0.72 - 1.83) Within 50 m from expressways and primary highways (I)  High 1.35 (0.95 - 1.93) Low 1.14 (0.83 - 1.57) Between 50 and 150 m from expressways and primary highways (II) High 0.90 (0.67 - 1.21) Low 1.02 (0.82 - 1.25) Within 50 m from secondary highways and major roads (III) High 1.21 (0.98 - 1.50) Low 1.11 (0.94 - 1.31) Between 50 and 150 m from secondary highways and major roads (IV) High 0.84 (0.7 - 1.00) Low 1.07 (0.90 - 1.28) Unemployment Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 1.13 (0.95 - 1.34) Low 1.17 (0.76 - 1.79) Within 50 m from expressways and primary highways (I)  High 1.25 (0.72 - 2.16) Low 0.81 (0.60 - 1.10) Between 50 and 150 m from expressways and primary highways (II) High 1.14 (0.81 - 1.60) Low 1.17 (0.94 - 1.46) Within 50 m from secondary highways and major roads (III) High 0.93 (0.73 - 1.19) Low 0.83 (0.69 - 0.99) Between 50 and 150 m from secondary highways and major roads (IV) High 1.21 (1.03 - 1.43) Low 1.01 (0.84 - 1.21) Family income Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 1.01 (0.83 - 1.23)     55 Table 25. Comparison between HR of different road proximity categories for low and high levels of Neighborhood -SES variables, when considering health outcomes (cont.) Health Outcomes Neighborhood- level SES Pollutant Level CCS Low 1.18 (0.80 - 1.74) Within 50 m from expressways and primary highways (I)  High 1.27 (0.73 - 2.20) Low 0.79 (0.58 - 1.08) Between 50 and 150 m from expressways and primary highways (II) High 1.00 (0.73 - 1.38) Low 1.16 (0.92 - 1.45) Within 50 m from secondary highways and major roads (III) High 1.04 (0.83 - 1.31) Low 0.82 (0.68 - 0.98) Between 50 and 150 m from secondary highways and major roads (IV) High 1.14 (0.97 - 1.34) Low 1.02 (0.85 - 1.22) Personal income Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 1.05 (0.87 - 1.27) Low 0.60 (0.19 - 1.88) Within 50 m from expressways and primary highways (I)  High 1.11 (0.76 - 1.62) Low 1.09 (0.72 - 1.63) Between 50 and 150 m from expressways and primary highways (II) High 0.83 (0.63 - 1.09) Low 0.96 (0.74 - 1.25) Within 50 m from secondary highways and major roads (III) High 1.14 (0.94 - 1.39) Low 1.17 (0.98 - 1.39) Between 50 and 150 m from secondary highways and major roads (IV) High 0.79 (0.66 - 0.94) Low 1.00 (0.80 - 1.25) Governmental transfers Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 1.05 (0.89 - 1.24) Low 1.32 (0.90 - 1.96) Within 50 m from expressways and primary highways (I)  High 1.31 (0.86 - 2.01) Low 1.14 (0.81 - 1.59) Between 50 and 150 m from expressways and primary highways (II) High 0.91 (0.68 - 1.23) Low 1.05 (0.86 - 1.29) Within 50 m from secondary highways and major roads (III) High 1.17 (0.94 - 1.45) Low 1.03 (0.86 - 1.23) Between 50 and 150 m from secondary highways and major roads (IV) High 0.99 (0.84 - 1.18) Low 1.13 (0.95 - 1.33) Low income Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 1.05 (0.88 - 1.26) Low 1.35 (0.94 - 1.93) Within 50 m from expressways and primary highways (I)  High 1.40 (0.92 - 2.14) Low 0.90 (0.69 - 1.18) Between 50 and 150 m from expressways and primary highways (II) High 1.09 (0.73 - 1.63) Low 1.13 (0.92 - 1.39) Within 50 m from secondary highways and major roads (III) High 1.01 (0.80 - 1.27) Low 1.09 (0.93 - 1.28) Between 50 and 150 m from secondary highways and major roads (IV) High 1.00 (0.83 - 1.21) Low 1.06 (0.90 - 1.26) Home ownership Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 1.07 (0.89 - 1.30) Low 0.98 (0.47 - 2.06) Within 50 m from expressways and primary highways (I)  High 1.32 (0.93 - 1.89) Low 1.19 (0.85 - 1.69) Between 50 and 150 m from expressways and primary highways (II) High 0.90 (0.68 - 1.19) Low 0.88 (0.69 - 1.13) Within 50 m from secondary highways and major roads (III) High 1.14 (0.93 - 1.39) Low 1.17 (0.99 - 1.39) Between 50 and 150 m from secondary highways and major roads (IV) High 1.00 (0.85 - 1.17) Low 0.99 (0.81 - 1.21) Neighborhood stress Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 1.04 (0.88 - 1.23)  The same assumptions on the behaviour of neighborhood level SES variables that were considered for the traffic pollution were also considered in the case of road proximity. These assumptions were checked overall, along the gradient of exposure to traffic represented by the different categories of road proximity as well as only for R-I, for subjects living within 50 m of expressways and highways.  56 When assessing the behaviour of neighborhood level SES variables overall, several of the indicators showed conclusive differences between the low and high levels of the variables and also along the exposure gradient. These variables were the percentage of subjects speaking a second language at home, other than French or English, linguistic isolation, and family and personal income variables. However, the assumptions posited for each of these variables were met only by the variable representing the percentage of subjects speaking a second language at home, other than French or English, and the variable representing linguistic isolation. For the other two variables that showed conclusive differences between the low and high levels the actual results were contrary to the expected results. When looking only at subjects living within 50 m from expressways and highways (R-I), three more variables showed clear differences between the low/high levels. One variable was university education, but in this case the expected result of high HRs for the low level category did not materialized while for the other two variables, governmental transfers and neighborhood stress, the actual differences matched the expected result, that is high levels of governmental transfers or neighborhood stress were associated with higher HRs compared with the low levels for these two variables.  Table 50 in the Appendix presents the distribution of subjects by the two levels of socioeconomic indicators at neighborhood level of aggregation and the five road proximity categories. It can be seen from Table 50 that for the variables that showed clear differences between the low/high levels and were in agreement with the a priori assumptions made regarding their behaviour, the distribution of subjects by high/low SES levels and road proximity justify the results, while that is not the case for the variables that had opposite results than expected.    57 Discussion The results of this study indicates that residential traffic related air pollution exposure increases the risk of chronic coronary syndrome, even in a region with relatively lower levels of traffic and ambient pollution in a cohort of subjects with no co-morbidities associated with cardiovascular diseases. Even more compelling evidence of the increased risk posed by traffic is produced by the analyses considering road proximity as a surrogate for traffic pollution. My study also looks at the impact of using socioeconomic covariates aggregated at different geographical levels and my results show stronger traffic pollution or road proximity effects when using covariates at medium levels of aggregation provided by the neighbourhood of residence as opposed to a smaller level of aggregation derived from census dissemination areas.  My analyses also found that there is a certain effect modification due to socioeconomic status for several of the available variables on the relationship between some or all of the traffic pollutants investigated and road proximity and chronic coronary syndrome morbidity and mortality. However, for the majority of the variables that showed differences between their low and high levels, these differences were contrary to what was originally expected. Traffic pollution related results I had available217 land-use regression derived exposure for NO, NO2, PM2.5, and black carbon for metro Vancouver area. The adjusted R 2 values for the models ranged from 0.39 to 0.62 and were similar across traffic metrics. Also, the distribution of NO was found to be more heterogeneous than that of NO2, a fact that supports the usefulness of this approach in assessing spatial patterns of traffic-related pollution. Another impetus for using traffic related pollution derived with land-use regression was the fact that the available ambient data was not at a finer enough scale while land use regression derived exposure was derived on a 10 m2 grid and thus had the potential to minimize exposure misclassification. Also, the ambient particulate matter data was collected only on fewer stations and for a shorter period of time and there wasn’t any data collected on ultra fine particles. Furthermore, a recent study conducted by Zanobetti et al.239 found that PM2.5 mass higher in Ni, As, and Cr, as well as Br and OC, which indicate a combustion source from either industrial activities or traffic, significantly increased hospital admissions. This result suggests that particles from industrial combustion sources and traffic may, on average, have greater toxicity and that focusing on the effects of traffic related pollution on cardiovascular diseases might constitute a more focused approach than considering ambient pollution.  Brauer suggested217 that traffic exposure estimates using land use regression may be more appropriate for primary pollutants, such as NO and black carbon, that vary the most spatially, whereas monitor-based estimates are more appropriate for secondary pollutants such as NO2 and PM2.5 that display less spatial heterogeneity. Nevertheless, my study used LUR estimates for all these four pollutants. Knowing that despite their improved spatial resolution of land use regression-based exposure estimates, they lack the more precise temporal information characteristic to exposures determined from ambient monitors244, I tried to improve the yearly average LUR exposure estimates by superimposing the monthly trend derived for the four pollutants (same trend was used for PM2.5 as for black carbon) from ambient data for the period between 1998 - 2003. Originally, it was found that  58 exposure estimates from land use regression and monitoring network data for the same pollutant were only moderately correlated and appeared to be somewhat independent, with each capturing different aspects of spatiotemporal variability in exposure244.  A potential source of exposure misclassification comes from the assignment of residential home exposure as the overall exposure, since people also spent time at work or commuting. Nevertheless, a Canadian study245 on time activity pattern, which corroborates its results with a larger U.S. study, shows that people spend most of their time in or around home, and our restriction of exposure assessment to residential address captures the most relevant part of exposure. As was already mentioned, NO2 and PM2.5 generally display spatially homogeneous distributions across small areas such as dissemination areas and blocks and, as a result, the ambient conditions at postal code and DA levels likely reflect the levels expected at home outdoors246. It is known that PM2.5 of outdoor origin will also penetrate indoors, and it was found that the correlation between long-term outdoor particulate matter concentrations and indoor levels of particulate matter from outdoor origin is high247. Although people spend most of the time indoors, at home, exposures to ambient air pollution while working and during commute are a relevant source of exposure248, even capable of inducing cardiac events91. Our traffic exposure models were also evaluated in conjunction with exposure data derived from ambient measurements against short-term measured personal exposures of pregnant women in Vancouver area249. This evaluation indicated that for NO and NO2, especially for those women who were the least mobile, LUR models were a stronger predictor of personal exposure and better explained between-subject (spatial) variability in exposure. On the other hand, monitor-based estimates were found to better explain within- subject (temporal) variability in exposure. Also, for PM2.5 and black carbon, monitor-based estimates of PM2.5 were more highly correlated with personal exposures than were the LUR models. A European study by Lanki250 with subjects recruited from Helsinki and Amsterdam looked at daily outdoor, indoor, and personal PM2.5 and absorbance (proxy for elemental carbon) concentrations among elderly subjects with cardiovascular disease during the winter and spring of 1998–1999. In Amsterdam, the exposure to environmental tobacco smoke (ETS) indoors was a major source of between-subject variation in PM2.5 exposures, and a strong determinant of PM2.5 and absorbance exposures. When the days with ETS were excluded, within-subject variation accounted for 89% of the total variation in personal PM2.5 and 97% in absorbance in Amsterdam. The respective figures were 66% and 61% in Helsinki. In both cities, outdoor levels of PM2.5 and absorbance were major determinants of personal and indoor levels. Nevertheless, traffic was also an important determinant of absorbance: living near a major street increased exposure by 22%, and every hour spent in a motor vehicle by 13% in Amsterdam. The respective increases were 37% and 9% in Helsinki. Cooking was associated with increased levels of both absorbance and PM2.5.  My results are largely in accordance with previous cohort studies reporting an association between long-term air pollution exposure and cardiopulmonary morbidity and/or mortality82,83,84,85,86,87,87a,87b,88,88a,88b,88c,90. A study in Rome87b that also employed NO2 exposure estimates derived by using LUR produced rather similar results with mine for acute myocardial infarction events. In their analyses, Rosenlund87b and her colleagues have distinguished though between fatal and non-fatal cases of cardiovascular events and they also have results for all their cardiovascular health outcomes grouped together. In their analyses, Rosenlund and her colleagues have used similar SES covariates aggregated at approximately  59 similar size geographical areas as the DAs employed in my study, but they also included subjects with some co-morbidities in the analyses. My study has some refinements compared with the study done in Rome, by being able to follow the residential histories of the subjects in the cohort and for being able to temporally adjust the exposure values prior and throughout the follow-up period, such that each person in the cohort had assigned the previous year average exposure for each month of the follow-up period. Despite experiencing smaller pollution levels than in Rome (Vancouver: mean = 31.1 μg/m3, min = 0.6 μg/m3, and max = 66.8 μg/m3; Rome: mean = 46.8 μg/m3, min = 24 μg/m3, and max = 73 μg/m3), I obtained similar ACS risk estimates for different pollution levels for the models that used SES covariates at neighbourhood level of aggregation. The ACS risk estimates were smaller when using DA-level SES, but comparable with the estimates for non-fatal acute myocardial infarction events. Also, my estimates are slightly higher than the risk estimates from a study12 conducted in the same area that looked among other at cardiovascular mortality rates in a cohort of approximately 550,000 subjects older than 65 year due to short term exposure to ambient pollution.  Rosenlund implies from the results in the Rome study87b that the mechanisms related to short-term effects6 (e.g., arrhythmia) could be of special importance despite the objective of their study being to assess long term effects of air pollution on coronary incidence. The fact that time series70-74 studies have repeatedly reported associations between daily variations in mortality and air pollution concentrations, while cohort studies82-88 have demonstrated increased mortality risks from annual average air pollution levels for people living in different geographic areas obscures the effects of short and long-term air pollution exposure and keeps open the question as to whether variations in air pollution with time or geography, or possibly both, are responsible for the increased mortality risks.  However, my study used something akin to time-series data on a monthly basis at residential level in a large cohort of healthy subjects and looked especially at chronic coronary syndrome group of diseases, which have a longer time frame of development. The risk estimates for chronic coronary syndrome and for three out of four pollutants show a rather consistent increasing trend with increasing levels of pollution. However, to properly disentangle the short-term from the long-term effects, the analyses had to contain lagged daily data for a period of four years, on the top of the average yearly data for the year prior to the current month of analysis. But even if I could have obtain daily coefficients for the LUR estimates for the four years of follow-up, the size of the cohort combined with the size of the exposure data for each individual was above the capacity of the PC used to manipulate the data and perform the analyses.  The general trend in ambient pollution in metro Vancouver area was decreasing for all four traffic pollutants analyzed and especially for NO and NO2 in the period prior and during the follow-up. The fact that I used yearly averages of exposure should help disentangle the long term effect of air pollution from the short term effects. The counterargument87b to this claim is that residents who have high air pollution exposure are probably also those who have a large variability in short-term exposure, whereas those who live in areas with low exposure on the geographical scale are expected also to experience a lower gradient in short-term exposure. The study done by Nethery249 in the same study area and with the same traffic exposure data does not support this claim in the sense that her study found that all pregnant women in the study experienced temporal exposure variability, fact that would bring the  60 estimates towards the null. However, I cannot claim that my study could entirely separate the short term effects of air pollution on the risk estimates found for the long term effects. In my preliminary analyses, I have used the five year average instead of one year prior average exposure and the risk estimates, although showing similar trends, where somewhat smaller than the ones presented here.  The only individual covariates that I had available for analyses were sex and age. For ACS and CCS, there was a striking difference in estimates between males and females, with males at higher risk of experiencing a cardiovascular event compared with women. These results differ from some other studies that have found an increase risk of cardiovascular morbidity and mortality in women88,88a,88c. These studies have focused only on postmenopausal women and it is plausible that in older age, the risk of a cardiovascular event might increase in women as well, a possibility suggested especially by the risk estimates for CHF outcomes, for which there wasn’t a major difference between men and women and there was an overall increased risk with older age groups, where the majority of subjects were women.  My study was based on medical records so no other individual risk factors were available for the subjects in the cohort, especially smoking. However, the rates of smoking for the province of British Columbia, where metro Vancouver is located, are quite low at 18.2%251, B.C. population being ranked first in Canada in terms of combined healthy behaviours. For metro Vancouver, smoking rates are even lower than the overall B.C. rates, some local health authorities reporting rates of 11%. There is a gradient in smoking rates in metro Vancouver local heath authorities, smaller in Vancouver and increasing in the local health authorities from Fraser Valley, east of Vancouver. A calculation of cardiovascular health outcomes by forward sortation area indicated (not shown here) a relative increase in risk for FSAs in the Fraser Valley. The higher smoking rates in Fraser Valley cannot be correlated with pollution, since higher levels of pollution (ambient and LUR) are found in Vancouver. A confirmation of the fact that smoking is not correlated with pollution exposure for the subjects in this study cohort came from a pilot study performed with a subset of subjects (~1,442 subjects) from the cohort that were surveyed through the Canadian Community Health Survey (v. 2001) (CCHS) and reported their smoking habits (15.7% current smokers). Although there were several statistically significant correlations between traffic pollution data and CCHS variables, the highest correlation coefficient in absolute value was only 0.15, while the majority of the coefficients of correlation were in the 10-2 order of magnitude. This indicates that there are no real correlations between traffic pollution and individual CCHS variables, including smoking status. The greater correlations were between individual and family income variables and traffic generated pollutants, and these I found to be inverse correlations.  Nevertheless, similar to other studies87b,93c,93d that employed cohorts based on administrative data, confounding from smoking or other individual risk factors could not be properly evaluated in this study because it was based on administrative registry data. But on the other hand, the approach to employ SES covariates at two different levels of aggregation yielded different results, with risk estimates higher in the case of medium level SES.  As I mentioned previously, my cohort was constructed using administrative databases and subjects’ health status was ascertained using these same databases, making the results prone to potential bias related to the quality of diagnosis. Any such bias would probably affect  61 patients equally, regardless of their air pollution exposure, and thus contribute to an underestimation of any positive associations. From the CCHS data of those subjects in the study cohort, I had available the information on subjects’ self report health status. The level of agreement between CCHS self reported data and medical records for ACS was quite high at 94.5%, while for the CCS and CHF was 85% and 95.5% respectively. One caveat regarding the health outcomes from the electronic administrative data is that only data from 1991 to 2003 was available for creating a health history for any individual, which might explain the imperfect level of concordance between the CCHS answers and the administrative health data.  The reliance on administrative databases for medical history, residential address, and residential history has the potential to bias the results, but given that I used the whole population in the metro Vancouver area it is more likely that the estimates were biased towards the null. On the other hand I had access to exposure at the residential level and I also had available the residential history of every individual in the cohort prior and during the follow-up period thus enhancing the reliability of exposure estimates and reducing the potential of misclassification. Contrary to other studies using CVD events as the outcome (incidence, mortality, or prevalence), my study focuses on the medium and long term contribution of air pollution to the underlying mechanisms that lead to cardiovascular diseases. The importance of focusing on the chronic effects of air pollution was reiterated in the literature, there being a consensus that assessments should not rely on the results of time-series studies but rather should be based on long-term follow-up in cohort studies251a,251b. The use of socioeconomic estimates at different levels of aggregation produced slightly different results, but maintained the trend in the estimates. It is more likely that the use of aggregate level SES has pushed the estimates towards the null, so my results, which are in line with the results from similar studies, might not in fact reveal the true magnitude of the effect of pollution on the cardiovascular health outcomes. Road proximity related results There are an increasing number of studies that use distance from home to major roadways or cumulative traffic intensity as proxies for exposure to traffic-related air93,93a,93b,238. The rationale for using distance to roads or traffic density resides on studies that have found that these measures are significant predictors of outdoor measurements of PM absorbance and NO2 240,241,242. Furthermore, previous investigations have shown that concentrations of traffic-related air pollutants drop off to the local background concentration between 100 and 150 m from the roadside243, indicating that a 150-m buffer is a reasonable size to capture local traffic-related air pollution. A recent study by Hochadel and colleagues (2006)242, after evaluating a wide range of buffer sizes, concluded that cumulative traffic within a 100-m radius buffer was the most predictive of both measured PM2.5 absorbance and NO2. The model that best predicted measured PM2.5 absorbance included both cumulative traffic within a 100-m buffer and the distance to the nearest highway. These studies support the choice of buffer size and road type (which is in fact a proxy for traffic density). More than that, with the five road proximity/traffic density combination used in my analyses, I had not only the ability to detect the most impacted subjects, but also to look at a potential gradient in the risk measures to see if there is any dose response relationship between road proximity and cardiac diseases morbidity and mortality.   62 The results of my large population-based study among subjects free of any cardiovascular health problems or associated co-morbidities indicate an increased risk of experiencing a cardiovascular health outcome with exposure to traffic near the patient’s residence. A greater impact on morbidity and mortality I observed for exposure to traffic closer to major roads and expressways, impact that tapered off with increase distance and decrease traffic density. However, for CCS, I obtained higher risk estimates for subjects living within 50 m of secondary roads (HR = 1.46 95% CI, 1.18 - 1.80). Not as high estimates at a local scale were also obtained by Medina-Ramon et al.93b(HR = 1.30; 95% CI, 1.13–1.49), in a recent study with a cohort or 1,389 subjects in Worchester, Mass. which they explained as due to a different composition of the pollutants mixture deriving from the two categories of roads. But this similarity might be just coincidental, since my CCS health outcomes did not figure among the health outcomes employed in the Worcester study.  Although my study looked at morbidity and mortality combined, the results are comparable with other studies that used proximity to traffic exposure as a surrogate for pollution exposure, but that looked at mortality only. It is possible that my estimates for mortality would be higher if evaluated independently, based on the results from a cohort study in Rome83b where estimates for mortality were higher than the overall estimates for morbidity and mortality combined. In my study, the range of resulting HR estimates for 50 m proximity to expressways and highways was between 1.09 (95% CI: 0.68 - 1.75) for CHF and 1.46 (95% CI: 1.18 - 1.80) for CCS, results that are comparable with the estimates for mortality obtained in a 10-year follow-up of Canadian subjects who underwent pulmonary function tests (HR = 1.18; 95% CI, 1.02–1.38)90, or those obtained in a cohort study among the Dutch general population92 (HR = 1.41; 95% CI, 0.94–2.12), where background levels of air pollution were taken into account, or those in a cohort from Worchester, Massachusetts87b 1.30 (95% CI, 1.13–1.49). The Dutch study, however, found larger effect estimates for cardiopulmonary mortality (HR = 1.95; 95% CI, 1.09–3.51) than for all-cause mortality, consistent with an increased susceptibility to the effects of air pollution in heart failure patients. For the subjects in my cohort residing within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads the HR estimates were 1.20 (95% CI, 1.10 - 1.30) for CCS. Overall, the effects of residential exposure to traffic in my study and the aforementioned studies were of similar magnitude but the differences attributable to varying susceptibility of populations, including the existing health status at the beginning of the follow-up period are hard to distinguish given other differences in methodology and geographic location of these studies.  While some of the studies92,93a that tried to link proximity to traffic and cardiovascular mortality and/or morbidity also included some background measures of air pollution (particulate matter), and then tried to explain the results partially on the difference between background pollution and the particulates generated by the traffic, difference confirmed as being more toxic239, my study looked separately at the effects of road proximity and the combined effects of traffic generated pollutants and road proximity using LUR exposure estimates. The estimates for road proximity measures, at least for subjects living within 50 m from expressways and major highways but also for the subjects residing within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads were higher than the estimates for the highest levels of traffic pollutants derived using land use regression. It is possible that these results are just an artefact of using a binary measure (road proximity) as opposed to a measure that has four levels (quartiles of traffic  63 pollution), but considering that LUR was also constructed considering some measures of road proximity, the highest quartile of traffic pollution should also indicate a close road proximity. These conclusions hold for both sets of analyses, the ones using SES at DA level of aggregation as well as for those using SES covariates at neighbourhood levels of aggregation.  These results seem to indicate that while traffic generated pollutants, like particulate matter, or black carbon, or nitrogen oxides play a role in the development and triggering of cardiovascular morbidity and mortality, proximity to roads might be a surrogate not only for traffic pollution but for some other cardiovascular risk factors. Other factors that can explain the difference between the two sets of analyses such as noise and the resulting psychosocial stress can also affect cardiovascular health. A recent study showed that men, exposed to sound levels of more than 70 dB(A) during the day, had a 1.3-fold increase in risk for myocardial infarction252. Close proximity to traffic may coincide with high traffic-related noise levels, making it difficult to separate effects of air pollution from noise effects. While other studies have tried to adjust for hypertension93 as a means of dealing with noise confounding253, my study included subjects that had no history of cardiovascular diseases or associated co-morbidities like hypertension, diabetes, or COPD. Another factor that might support the added impact of noise, besides that of traffic pollution, is that my study area experience sustained levels of rain all year around and anecdotal evidence suggests that noise levels are higher on wet pavement, residents of Vancouver complaining of increased traffic noise during rainy days, even when they don’t live in close proximity to major roads or bus routes.  My analysis concerning long term exposure to proximity to roads suffers from the same shortcomings as the analysis based on LUR traffic pollutants, in terms of reliability of health diagnostics, lack of personal information regarding smoking status and other individual risk factors, but as I argued before, I expect that my estimates do not suffer from major non- conservative biases but are rather pulled towards the null. On the other hand, different from any of the previous studies I had reliable long term residential history information that mitigated the problem of exposure misclassification. SES related analyses This study is the first to thoroughly examine exposure to traffic pollution or road proximity in the context of social inequalities with an approach akin to a well developed experimental design with two factors, each factor with several levels. All the previous analyses have included, separately, socioeconomic information at two levels of aggregation. Thus, my methodological approach should not be confused with a hierarchical approach where in the same analyses the factors that are included are at different levels of aggregation (i.e. individual, neighbourhood, city, etc) and the analyses are performed taking that in consideration (i.e. random effects). Following the suggestion from Laurent et al.237 (see Figure 4), I performed individual analyses to determine the HR estimates due to traffic pollution and road proximity for the subjects in the lowest, respectively highest of the SES quintiles of each variables available. Thus I was able to investigate the effect of the whole range of exposure values for the subjects from the lowest and highest SES quintiles respectively.  64 As I presented in the results section, the analyses concerning the effect modification of socioeconomic variables on the pollution related cardiovascular health outcomes included only subjects in the highest and the lowest of the five quintiles of those socioeconomic variables investigated. The number of subjects in the four distinct pollution categories (for traffic pollutants) or the two road proximity classes (in/out) varied within each socioeconomic quintile that I investigated while the number of subjects in any of these two distinct levels of socioeconomic status were approximately the same across the variables investigated. A complete summary of the distribution of subjects within the two levels of the socioeconomic variables and the four (two) levels of traffic exposure (road proximity) is presented in Tables 47 to 50 in the Appendix. Because I used the exposure (road proximity) in the analyses as a dynamic variable, the pollution classes are in fact represented in these tables by the first month of follow-up only (similar with Tables 17 and 18).  The twenty variables that I used in the analyses [2 aggregation levels – dissemination area (DA) and neighbourhood (N) – each with ten variables] can be grouped in several categories, for the ease of discussion and also for conceptual reasons. There are five broad classes (identical to those described in the methodological section) that can be construed and that I will discuss separately. These classes are:   Racial/Cultural, including the percentage of Chinese minority variable (DA), percentage of people speaking a second language at home variable (N), and percentage of people experiencing linguistic isolation variable (N);  Income and wealth, which includes the average personal income variable (DA, N), average family income variable (DA, N), percentage of people with low income (DA, N), income variation variable (DA), percentage of income coming from governmental transfers variable (N), percentage of families spending 30% or more of income on shelter costs variable (N), percentage of occupied dwellings that are owner-occupied variable (DA, N);  Education, which consists of the percentage of total population with any university degree variable (DA, N);  Labour, which includes employment rate variable (DA), seasonally adjusted unemployment rate among persons aged 25 years and over (N) variable, and percentage of people working in management variable (DA);  Transportation means, which consists of the percentage of people biking, walking or using public transit in their daily commute to work variable (DA).  The Public Health Agency of British Columbia argues254 that the Province represents a paradox when it comes to health status and social disparities. Thus, while the life expectancy in BC is one of the highest in the world and the province has the lowest rates of smoking and obesity and the highest rate of physical activity in the country, BC has the highest rates of poverty and particularly child poverty in Canada. This presents the aforementioned paradox: despite having by some measures the best overall health outcomes in Canada, BC also has the highest rate of socioeconomic disadvantage in the country. The paradox can be explained partially by looking at the range of values the averages hide, by the fact that certain trends (like high child poverty rates) are only recent, and by the time lag in the causal link between performance on the social determinants of health (including educational achievement, poverty, early childhood development, housing, etc.) and outcomes on health  65 measures such as life expectancy. The full impact of the effects of the upstream determinants of health may not yet be fully realized or apparent from B.C.'s current population health statistics. However, these disparities in income, education, health are evident mostly at the provincial level, while my study area shows a relatively high homogeneity (except in the downtown eastside of Vancouver) in terms of the major socioeconomic indicators as well as health status, in the sense that Vancouver area has the highest levels of education and income in the province and also average outcomes on health measures. Racial/Cultural indicators There is a consistent amount of sociological and epidemiological research, especially in the U.S. that tries to disentangle the role of racial attributes on health outcomes. While there are probably instances that certain genetic characteristics of a particular race might constitute a risk factor, the literature shows that socioeconomic factors like income or education or class trump biological characteristics based on race in relation with health outcomes255. The variables chosen for my analyses as reflecting a certain racial/cultural profile of dissemination areas and neighborhoods in Metro Vancouver do not try to capture some intrinsic genetic characteristics of the population that makes them more or less susceptible to the impact of air pollution and its effects on cardiovascular health outcomes. Instead, it tries to capture material and immaterial aspects of culture as described by Eckersley153 in his paper focused mostly on the western culture. If the variable representing the percentage of Chinese population in the DA might indeed reflect in a distinctive way the influence of a different culture than the traditional Canadian milieu, the other two variables at the neighborhood level are less defined as “cultural” variables, since they refer to the percentage of people that use at home a language that is not English or French and that can be any other language, not necessarily Chinese or to the percentage of people that do not speak either English or French and are assumed to be isolated in their neighborhood. But since the Chinese component in the ethnic make-up of Vancouver is the dominant one, the correlation between the percentage of Chinese people and the percentage of people that speak another language at home or don’t speak English or French is quite high.  A relatively recent study210 that investigated the association between body mass index (BMI) in urban Canada and different socioeconomic variables at individual, neighborhood and metropolitan levels found that there was a strong association between immigrant status and BMI for both men and women, and this association attenuated with length of time in Canada. The study also found that small incremental effects of neighborhood-level environments on the BMI of men and women in urban Canada were related primarily to two neighborhood characteristics: low education levels and the presence of immigrants (for men only). The authors suggested that it is possible to conclude from the neighborhood-level findings that recent immigrants bring with them customs and norms regarding diet or physical activity that become part of local practice and influence behaviors beyond the immigrant community. It is not sure how long this immigrant healthy effect would last since BMI has been shown to increase in Canadian adults with time since immigration256, regardless of self-ascribed ethnicity although while about half of Whites (who constituted more than 80% of the population) were overweight (including people who were obese), East/Southeast Asians had the lowest self-reported prevalence of overweight (22%). Given the relationship between BMI and obesity in general and cardiovascular diseases197,199,202,228, I  66 would then expect lower rates of cardiovascular diseases in DAs and neighborhoods with a higher percentage of Chinese population, or with people speaking a second language at home, even when experiencing higher levels of pollution. In the case of linguistic isolation, the effect of isolation might be conducive to higher stress levels, but on the other hand, if there is a whole community isolated from the overall population and somewhat self-reliant, this might in fact increase the levels of community support and decrease the overall levels of stress and the risk of disease.  This was exactly the case at the dissemination area levels, where, for traffic pollution exposures, I found lower hazard ratios for all three health outcome categories in areas with a higher level of Chinese immigrant population. This was true though especially for PM2.5 and black carbon. These results might carry even more weight due to the fact that areas with a high percentage of Chinese immigrants were also exposed to higher levels of traffic pollution (Appendix – Table 47). I obtained the same pattern of results at the neighborhood level, where, for all three health outcomes investigated, the hazard ratios for particulate matter were markedly smaller in neighborhoods with a higher percentage of a second language speakers or with people linguistically isolated, despite the fact that again, there are more people in these neighborhoods experiencing high levels of air pollution.  As I mentioned before, these differences do not reach significance, the confidence intervals of hazard ratios always intersecting. And there is the interplay between different levels of aggregation (i.e. for the variable indicating the percentage of Chinese population, the maximum value was ~90%, while at the neighborhood level we had maximums of 49 and 25% respectively for a second language and linguistic isolation), different pollutants, and different cardiovascular health outcomes. As I mentioned in the results section, the only consistent results at DA-level of aggregation are for traffic originated particulate matter and for black carbon, for which subjects living in DAs with higher levels of Chinese immigrants experienced a lower risk of developing any of the three cardiovascular health outcomes, even though there more of these people experiencing higher levels of particulate matter or ultra fine particles. For CCS at neighborhood levels (linguistic isolation, second language spoken at home) all four traffic pollutants showed distinct differences between the low/high levels of the two variables belonging to the racial/cultural group.  Similarly, areas with a high percentage of Chinese population and in close proximity to expressways and highways markedly decreased the risk of CCS. I obtained similar results when using the two variables at neighborhood levels of aggregation. These results are more in accord with the ones using the two variables at the neighborhood level of aggregation for traffic pollution. The Canadian study210 that looked at BMI and immigration defined neighborhoods as census tract areas (CTAs), which are geostatistical areas containing about 4,000 people and thus about ten times larger than census dissemination areas and close in size with the neighbourhoods used in my study. At this level of aggregation, I found evidence that, for higher levels of particulates or in closer proximity to traffic, subjects in areas with more immigrants are experiencing lower risk levels for any of the cardiovascular health outcomes investigated. These results are replicated at the smaller level of aggregation represented by dissemination areas.  These results are reinforced by the findings of Anand155, who found in a cohort of 1,227 subjects of different ethnical background (white, Chinese, east-Asians, aboriginal) in  67 Hamilton Ontario that people of Chinese origin had the lowest probability of CVD, although the Caucasian population had the lowest social disadvantage among the ethnic groups investigated. The potential healthy immigrant effect is not though reinforced by high levels of education, the correlation between the three racial/cultural variables and university education being quite low at under 0.2. Income and wealth variables In a review of literature122 concerning income inequalities and health outcomes from 2006, Wilkinson and Pickett looked at 168 analyses in 155 papers reporting research findings on the association between income distribution and population health and concluded that health is less good in societies where income differences are bigger. They also suggest that the studies of income inequality are more supportive in large areas because in that context income inequality serves as a measure of the scale of social stratification, or how hierarchical a society is. Their explanation for lack of evidence in some of the studies that they analyzed rests on the fact that many studies measured inequality in areas too small to reflect the scale of social class differences in a society. Also, they noticed that a number of studies controlled for factors which, rather than being genuine confounders, were likely either to mediate between class and health or to be other reflections of the scale of social stratification. However, there are numerous studies at individual level that link lower income with an increase in negative health outcomes, including cardiovascular diseases128,133,171,175,181. There is also evidence that the “wealth–health gradient” in cardiovascular mortality may be partially ameliorated by more rigorous management of known risk factors among less affluent persons257 or in systems with universal medical coverage180,257.  The time series analysis on the short term effects of air pollution on mortality in a cohort of old people (>65 years) in Vancouver12 found modest increases in the all-cause and cardiovascular mortality due to exposure to NO2, CO, and SO2 (but not in particulate matter) among individuals living in lower relative to higher socioeconomic areas. Due to the many dimensions of my analyses (multiple exposure matrices, two levels of spatial aggregation and more than ten variables, each with two classes, describing various aspects of income and wealth), my results are relatively hard to summarize. Thus, for personal and family income I found contrary to expectations, that for both aggregation levels, subjects in areas with low personal or family income had smaller HRs at higher levels of exposure or proximity to expressways and highways than subjects from areas with higher family or personal income. The results were more consistent at neighborhood level of aggregation, which showed clear differences for all four traffic pollutants These results are not explained by the fact that there were more subjects with low family or personal income in areas with high traffic pollution exposure or in proximity to expressways and highways. Similar conflicting results were available for subjects living in areas with more households below the low-income cut-off, subjects that have in a lower risk of developing a cardiovascular health outcome at higher levels of traffic pollution exposures or when in proximity of expressways or highways. Although more subjects live in areas with a higher proportion of households below the low-income cut-off that were also exposed to higher levels of traffic pollution or proximity to expressways and highways, the lower levels of personal and family income might be still above the low-income cut-off and offer sufficient material protection in absolute terms and mitigate the risk of developing CVD.   68 It is harder to draw an over arching conclusion regarding home ownership, because there are differences in the results at the two different levels of aggregation. My results indicate that for CCS health outcomes, the health risks were more pronounced in neighborhoods with a high proportion of home ownership for both levels of aggregation, and especially at DA- level, where the differences were manifested for all four traffic pollutants. For CCS, higher ownership rates may be associated with longer periods of stay in a particular area and if those areas were associated with higher levels of pollution, this may lead to increased rates of CVD.  It was found that the health effect of income inequality weakens when moving from higher levels of aggregation (such as nations) to lower levels (such as neighborhoods),237,258,259 and even the direction of the effect might change260. The main reasons for such an effect of scale might be fundamentally political and this political effect it is more likely to be perceived at national or state level. In those nations that have great disparities within their borders there is a smaller tendency to have extensive redistributive systems and there can be found lower levels of human capital and social security investments. My analyses were conducted at small and very small levels of aggregation and it is not clear whether these units have enough political autonomy to influence the processes behind income inequality. However, while the dissemination areas were created by Statistics Canada through a process that only tried to equalize the number of people living in the proposed DAs, the neighbourhood boundaries were based upon factors of judgment that went beyond the relatively simple approach considered for the creation of dissemination areas. This fundamental difference in the way the two levels of aggregation were construed might explain the difference in the results between the DA-level and neighbourhood level, with neighbourhood level results of higher magnitude, contrary to the original hypothesis that was derived from Laurent’s review237.  While income dispersion at the regional level may be seen merely as a result of the labor market’s geographic structure, within the smallest units such as neighborhoods, the direction of the association is not evident. Income inequality within a neighborhood is more likely a measure of socioeconomic heterogeneity, a manifestation of the mixture of residents with divergent social and economic characteristics. The residential composition of a neighborhood is partly a result of housing and zoning strategies but it is also driven by wider segregation processes. Furthermore, it has been suggested that economic heterogeneity in urban communities can have beneficial effects. One hypothesis is that poor people benefit from sharing neighborhoods with more affluent families and that a certain proportion of middle and upper class people in urban neighborhoods may be necessary to sustain basic institutions261.  In a study conducted by Stjarne in Sweden208, in the multilevel analyses that were performed, an index of neighborhood heterogeneity was used at the higher level of aggregation (individual/neighborhood) with the assumption that the level of neighborhood homogeneity has an impact on the myocardial infarct (MI) outcomes. Their expectation was that more heterogeneous neighborhoods reduce the risk of MI on the assumption that it is an advantage to society if people from different walks of life share the same neighborhood. Although the study found an increased incidence of MI in low-income neighborhoods that was not due to individual social characteristics, the socioeconomic heterogeneity within a neighborhood seemed to have less effect on MI. In my study I did not use an index of income variation that is usually employed (Gini coefficient) but just the coefficient of  69 variation in personal income and this information was available only at the dissemination area level, which is the smaller of the two levels of aggregation that I used. Given the research in the area, I did expect to have some conclusive results on the assumption that high levels of income variation might be in fact beneficial to health and lower the risk of CVD. My results confirmed this initial assumption for traffic pollution exposure and road proximity as well. The results indicate that overall, for higher levels of exposure (proximity to major roads), there is a decreased risk in experiencing any of the three CVD analyzed, for subjects living in dissemination areas with a greater income heterogeneity  The other two remaining variables that were used to express some aspects of income, governmental transfers and neighborhood stress, are both available only at neighborhood level of aggregation. My results indicate for three out of four pollutants that higher levels of governmental transfers are associated with lower risks of CCS compared with subjects in areas with lower levels of governmental transfers. These results contradict the initial expectations. However, for proximity to expressways and highways the results indicated that subjects living in areas with high levels of governmental transfers or neighborhood stress are at a higher risk of experiencing CCS. Education Besides income, education level is the other socioeconomic factor that was found to be related to health status. Among the measures of SES, such as education, income and occupation, low level of education was most consistently associated with higher coronary risk262,263 and my expectations were that subjects from areas with higher levels of university graduates would manifest a lower level of risk of CVD, at higher levels of traffic pollution or if living in proximity of major roads. My results indicated such a relation for CCS only at the DA-level and only for black carbon. In the few other circumstances (CCS-DA for nitrogen oxides and particulate matter, and CCS-N for nitrogen oxides) the risk was found to be sometimes higher for subjects living in areas with a higher proportion of university graduates but most of the time the trends for low/high levels of university degrees were intersecting, diverging and then converging. This might be due to the fact that the distribution of subjects across the levels of pollution were unbalanced, with much more subjects from areas with more university graduates being exposed to higher levels of traffic pollution (Tables 47 and 48 in the Appendix). This is similar with the distribution that was found by Hoek90 in a cohort study in the Netherlands, where subjects living near a major road were found to have had slightly higher education, worked less frequently in blue-collar jobs, and smoked fewer cigarettes. So, even though higher levels of education might have a protective effect, just the sheer number of highly educated people exposed to higher pollution might produce an artificially higher risk. Nevertheless, for this cohort, there were fewer subjects from areas with high levels of university education living in one of the five road categories investigated as opposed to those with low levels of university education (Tables 49 and 50). Labor and occupation With the objective of helping employers and government to better articulate the tradeoffs between policies that affect economic outcomes of labour market experiences and policies that affect the health consequences of these experiences, Lavis268 has developed a conceptual framework encompassing the availability and nature of work. In this conceptual framework, work availability is characterized by six important experiences. Three of these experiences  70 represent under work: discouraged workers who have withdrawn from the labour market, unemployed workers actively seeking work, and conditions of underemployment - expressed by hours of work and skill utilization. The fourth experience represents those employment circumstances in which there is a constant fear of underemployment. The remaining two experiences in this dimension of the conceptual framework described by Lavis are the experience of full employment and of over-employment or overwork. The dimension of the nature of work has only three defined features: job characteristics, job position within a firm or society, and the organizational characteristics of the firm.  The exploratory mechanisms and pathways for work related health effects go from the material consequences of job loss, to the disruption of social ties that can lead to disturbance in physical and mental health269 and there is a substantial body of research that have estimated the consequences of involuntary unemployment to individual physical and mental health, such that the relationship can be considered causal. A review by Lavis et al.270 did not find studies that conclusively linked unemployment to initiation of disease, although a large body of work had consistently found unemployment associated with the risk of death following the unemployment period.  As indicated in the results section, my study found that subjects from dissemination areas with low level of employment and higher levels of pollution are at lower risk of developing CCS as opposed to subjects where employment rates are higher. Corroborating these results, at neighborhood aggregation levels, subjects from areas with low unemployment were at higher risk of CVD than subjects from areas with high unemployment, for high levels of traffic pollution. Also, subjects from areas with high/low employment/unemployment were found to be at higher risk of CVD if living within 50 m of expressways than subjects from areas with low/high employment/unemployment. It is worth noticing that there were almost double the number of subjects living in areas with high/low unemployment/employment and high pollution levels (proximity to major roads) than otherwise. This might have impacted the results. Also, the census data refer only to the situation when the Census information was collected (2000/2001) and this covers only one year of the whole follow-up period. Unemployment/employment rates are not static and change in time and might not have long lasting effects, like education levels or racial profile.  Starting with the Whitehall II study conducted by Marmot271 and his colleagues, it was shown that people that have less control and are performing highly demanding jobs experience higher mortality than people higher on the hierarchy, that have a high control on their jobs. Similar associations were found by studies that investigated the effects of air pollution and health and have controlled for work type123,183,184,191,212. However, my study has found that at dissemination area level, subjects living in areas with a low percentage of people working in management were less prone to develop CCS, at high levels of traffic pollution (for all four pollutants), compared with subjects living in areas with a high proportion of people working in management. The subjects in the low/high areas were in approximately similar numbers exposed to high levels of traffic pollution. The results considering road proximity were rather inconclusive for all health outcomes investigated.  71 Transportation Only at dissemination area I had available information regarding the proportion of people in an area that use public transit, bike or walk in their daily commute to work. For an individual, the use of a private car for daily work commute might represent an additional stress and Gee195 found in a cohort in Los Angeles that perceived traffic stress was associated with both general health status and depression in multivariate multilevel models, such that persons reporting traffic stress had lower health status and more depressive symptoms. The author reported also that there was an interaction between vehicular burden and traffic stress for both general health status and depression and that persons living in areas with greater vehicular burden and who reported the most traffic stress also had the lowest health status and greatest depressive symptoms. In Canada, metropolitan sprawl, which is associated with higher levels of car use, was found to be associated with higher BMI for Canadian men210. However, Peters91 found that transient exposure to traffic not only as drivers but as passengers in a bus or bikers may increase the risk of myocardial infarction in susceptible persons.  The initial hypothesis that subjects using public transit, biking or walking to work are exposed to higher levels of traffic pollution, especially for the highest quartile was confirmed by the results of my analyses (Tables 47 and 49). The expected consequence of this fact was that these people will have higher rates of CVD than subjects living in areas with low levels of public transit usage or that bike or walk to work less. For subjects living within 50 m from expressways and highways, or within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads I found evidence that for all CVD health outcomes, subjects from areas with a higher percentage of daily commuters that walk, bike or use public transit to work are at higher risk of CVD than subjects from areas with a higher use of private cars. However, these results were not entirely replicated by the analyses involving exposure to particulate matter where the HRs were higher for subjects living in areas with low usage of transit, biking or walking. Nevertheless, for nitrogen dioxide exposure the results were similar with those of road proximity measures. These apparently contradictory results may be explained that the smaller number of subjects living in the proximity to major roads are exposed not only to traffic pollutants but to other factors of stress and they represent only a fraction of the larger number of subjects that although are exposed to high levels of traffic pollution, by using transit, biking or walking to work, manage to reduce their HRs of experiencing cardiovascular health outcomes. Strengths and weaknesses of the study My study is population-based derived from administrative databases and consisting of subjects with no history of cardiovascular diseases or associated co-morbidities. This makes the current study less sensitive to selection bias due to a differential response rate associated with area-level deprivation, or exposure differential, which can be a considerable problem in contextual studies264. More than that, the validity of CVD diagnosis was verified with the self declared health status of those subjects from the cohort that were also captured by the first cycle of Canadian Community Health Survey. By using the incidence of the first event of CVD is considered to be especially relevant when evaluating the etiologic implications of contextual exposures. In particular, it reduces the risk of bias from health-related selection into low-income contexts. Also, because case-fatality rates have been found to be related to neighborhood deprivation265, the inclusion of both fatal and nonfatal cases reduces  72 influences from factors such as availability of acute care or distance to a hospital. However, selecting in the cohort only subjects without any previous medical conditions has the potential to overemphasize the “healthy worker effect”, which may explain some of the results, especially in relation with traffic pollution. Another strength of my study resides in the availability of information on residential mobility and previous contextual exposures, which is rare in long term studies. I had available residential information for more than 8 years prior to the start of the follow-up and for the whole duration of the follow-up and I was able to assign precise traffic exposures (that were themselves adjusted for the period of follow up and one year prior to that to reflect the existing temporal trends in the area) and especially road proximity identifiers.  There is a lot of criticism for using administrative boundaries when estimating contextual attributes. While one of the levels of aggregation used in my study is represented by purely administrative boundaries as defined by Statistics Canada for census purposes, the other level of aggregation used represents neighborhoods as defined by the people living in the respective areas216. For the administratively defined dissemination areas, the summary statistics indicate that they capture the spatial differentiation of socioeconomic recourses reasonably well. In fact, the variances found in the values of covariates at dissemination area level are higher then those corresponding to the neighborhood areas. This probably happens because at a higher level of aggregation, information from sometimes contrasting small residential areas is merged, which hides important contrasts between neighborhoods and causes non differential misclassification of exposures.  The main weakness of this study is the lack of individual level information besides sex and age. This would have given me the possibility to conduct hierarchical analyses. The use of area based variables as a replacement for missing individual level information has been for long criticized as the ecological fallacy and even the use of such data in the presence of individual level information and the drawing of causal inferences from the neighborhood studies so far conducted has recently been questioned266. In line with Diez-Roux14,267, I would argue that it is still productive to use observational data and that the neighborhood effects are not by definition endogenous to the compositional characteristics of neighborhoods. The main question addressed in my study is whether specific aspects of the local social environment have an impact on CVD incidence, and especially how they modify the effect of traffic related pollution on the CVD. Answering this question can be regarded as a first step toward attaining the ultimate goal of understanding the causal role played by context. However, that goal is far from reached, and as Diez Roux states, “associations . . . on neighborhood health effects are what they are: measures of conditional associations under certain assumptions.”267(p.1959) The range of covariates that I used from the available neighborhoods serve as a proxy for a range of circumstances affecting people’s daily life that would have been missed if only social circumstances measured at the individual level were considered. Conclusions My study found significant evidence that even in areas with low levels of traffic pollution and even for healthy people, there is an increased risk of chronic coronary syndrome morbidity or mortality due to an increased exposure to traffic pollution or to living in close  73 proximity to expressways and highways, especially when considering socioeconomic variables at medium levels of aggregation.  While the results were not always very conclusive regarding the extent to which socioeconomic indicators modify the effect of traffic pollution on health on the expected trajectory (individuals from more advantaged areas would be less subjected to the effects of traffic pollution compared with individuals living in more disadvantaged areas) several socio economic attributes stand out, especially in respect to traffic pollution exposure estimates. At dissemination area levels, in the case of exposure to particulate matter, subjects living in areas with a higher percentage of Chinese population were at a lower risk of CCS health outcomes. Subjects living in areas with a higher proportion of university degrees were also at a lower risk of experiencing CCS in conjunction with black carbon exposure. Both of these could signify an immigrant effect and or an education effect, due to immigrants having in general higher education. However, the effect modification of family and individual income in respect with all traffic exposures was inverse than expected, subjects living in areas with high levels of family or personal income being at a higher risk of experiencing CCS related morbidity and/or mortality. Similarly, a high level of home ownership was also associated with increased risk of CCS in conjunction with all traffic pollutants at DA level and with particulate matter only at neighborhood level. Subjects living in areas with a higher proportion of people working in managerial positions were at higher risk of experiencing CCS in relation with all traffic pollutants. As expected, subjects living in areas with a high proportion of people that use transit, bike or walk to work were found to be at higher risks of developing CCS in relation with NO2 and road proximity. Also as expected, subjects living in areas with greater income variation were at a lower risk of experiencing CCS in relation with all traffic pollutants except nitrogen oxide.  A certain level of similarity was found for some of the variables derived at the neighborhood level of aggregation, when compared with the results at dissemination area. Thus, the two covariates representing the racial/cultural axis (other language used at home and linguistic isolation) indicated, as the covariate describing the percentage of Chinese population in a dissemination area, that subjects living in areas with higher percentages of people using a second language at home or higher percentages of linguistically isolated people are at a lower risk of experiencing any of the CVD investigated in conjunction with particulate matter. Also, at neighborhood level, subjects living in areas with higher family incomes were at greater risk of experiencing CVD related morbidity and/or mortality in conjunction with nitrogen oxides exposures. Same was the case for the personal income variable. Nevertheless, subjects living in areas with increased unemployment rates were at lower risk of developing any CVD in relation with the nitrogen oxides.  The results for the stratified analyses were similar for traffic pollution and road proximity. However, only for some of the variables investigated (the racial/cultural variables, education variable, transportation variable and income variation variable), the initial hypotheses regarding their behaviour were confirmed by the analyses. This was not the case for income and wealth related variables. As for the results concerning exposure and CCS outcomes, neighborhood level of aggregation produced higher HRs and clearer separation between strata in the stratified analyses for the majority of the pollutants investigated.   74 These results would need to be studied in conjunction with analyses at individual level data to confirm that these trends are real. Also, a more detailed analysis is required to disentangle the results obtained for ACS and CCS on one hand, and CHF on the other hand, because there were many circumstances where the behavior of the risk estimates for these two groups of CVD health outcomes was totally opposite.   75 References 1. Pearson TA, Blair SN, Daniels SR, et al., for the American Heart Association Science Advisory and Coordinating Committee. 2002. AHA guidelines for primary prevention of cardiovascular disease and stroke: 2002 update: consensus panel guide to comprehensive risk reduction for adult patients without coronary or other atherosclerotic vascular diseases. Circulation. 106:388–391.  2. Smith SC, Blair SN, Bonow RO, et al. 2001. AHA/ACC Scientific Statement: AHA/ACC guidelines for preventing heart attack and death in patients with atherosclerotic cardiovascular disease: 2001 update: a statement for healthcare professionals from the American Heart Association and the American College of Cardiology. Circulation. 104:1577–1579.  2a. Brook RD, Rajagopalan S, Pope CA 3rd, Brook JR, Bhatnagar A, Diez-Roux AV, Holguin F, Hong Y, Luepker RV, Mittleman MA, Peters A, Siscovick D, Smith SC Jr, Whitsel L, Kaufman JD. 2010. Particulate matter air pollution and cardiovascular disease: An update to the scientific statement from the American Heart Association. American Heart Association Council on Epidemiology and Prevention, Council on the Kidney in Cardiovascular Disease, and Council on Nutrition, Physical Activity and Metabolism. Circulation. 2010 Jun 1;121(21):2331-78. Epub 2010 May 10. Review.  3. US Environmental Protection Agency, Office of Research and Development, National Center for Environmental Assessment, Research Triangle Park Office. 2003. Air Quality Criteria for Particulate Matter (Fourth External Review Draft). Research Triangle Park, NC: US Environmental Protection Agency; Publication No. EPA/600/P-99/002aD and bD.  4. N Künzli, R Kaiser, S Medina, M Studnicka, O Chanel, P Filliger, M Herry, F Horak Jr, V Puybonnieux- Texier, P Quénel, J Schneider, R Seethaler, J-C Vergnaud, H Sommer. 2000. Public-health impact of outdoor and traffic-related air pollution a European assessment. Lancet 356: 795–801  5. Burnett RT, Cakmak S, Brook JR. 1998. The effect of the urban ambient air pollution mix on daily mortality rates in 11 Canadian cities. Can J Public Health. 189:152–156.  6. Routledge HC, Ayres JG, Townend JN. 2003. Why cardiologists should be interested in air pollution: Heart. 89:1383–1388.  7. Poloniecki JD, Atkinson RW, de Leon AP, et al. 1997. Daily time series for cardiovascular hospital admissions and previous day’s air pollution in London, UK. Occup Environ Med. 54:535–540.  8. World Health Organization. World Health Report. 2002. Geneva: World Health Organization; Available at: http:www.who.int/whr/2002/en/.  9. Vitaliano, P. P. , J. M. Scanlan, Jianping Zhang, Margaret V. Savage, Irl B. Hirsch, and Ilene C. Siegler. 2002. A Path Model of Chronic Stress, the Metabolic Syndrome, and Coronary Heart Disease. Psychosomatic Medicine 64:418–435  10. Malstrom M., Jan Sundquist, and Sven-Erik Johansson. 1999. Neighborhood environment and self- reported health-status: a multilevel analysis. Am. J. Publ. Health 89 (8): 1181-1186.  11. Sampson R.J., J. D. Morenoff, and T. Gannon-Rowley 2002. Assessing “Neighborhood Effects”: Social processes and new direction in research. Annu. Rev. Sociol. 28:443–78  12. Villeneuve, Paul J., Richard T. Burnett, Yuanli Shi, Daniel Krewski, Mark S. Goldberg, Clyde Hertzman, Yue Chen, And Jeff Brook. 2003. A time-series study of air pollution, socioeconomic status, and mortality in Vancouver, Canada. Journal of Exposure Analysis and Environmental Epidemiology 13: 427–435   76 13. Jerrett, M., Richard T. Burnett, Renjun Ma, C. Arden Pope III, Daniel Krewski, K. Bruce Newbold, George Thurston, Yuanli Shi, Norm Finkelstein, Eugenia E. Calle, and Michael J. Thun. 2005. Spatial Analysis of Air Pollution and Mortality in Los Angeles. Epidemiology 16: 727–736  14. Diez-Roux AV. Bringing context back into epidemiology: variables and fallacies in multilevel analysis. Am J Public Health 1998;88:216–22.  15. Schwartz S. The fallacy of the ecological fallacy: the potential misuse of a concept and the consequences. Am J Public Health 1994;84:819–24. 16. Brunekreef B, Holgate ST. Air pollution and health. Lancet. 2002;360: 1233–1242.  17. Pope CA. Epidemiology of fine particulate air pollution and human health: biologic mechanisms and who’s at risk? Environ Health Perspect. 2000;108:713–723.  18. Shy, Carl M. 1989. Review, discussion, and summary of epidemiological studies. Environ Health Perspect 79: 187-190  19. Aunan, Kristin. 1996. Exposure-Response Functions for Health Effects of Air Pollutants Based on Epidemiological Findings. Risk Analysis 16( 5): 693-709  20. Pope, III C. Arden. 2000. Epidemiology of Fine Particulate Air Pollution and Human Health Biologic Mechanisms and Who's at Risk. Environ Health Perspect 108(suppl 4) :713-723  21. Pope III, C. Arden, David V. Bates, and Mark E. Raizenne. 2000. Health Effects of Particulate Air Pollution: Time for Reassessment. Environ Health Perspect 103(5): 472-480  22. Morris, Robert D. 2001. Airborne Particulates and Hospital Admissions for Cardiovascular Disease: A Quantitative Review of the Evidence. Environ Health Perspect 109(suppl 4):495–500  23. Dockery, Douglas W. 2001. Epidemiologic Evidence of Cardiovascular Effects of Particulate Air Pollution. Environ Health Perspect 109(suppl 4):483–486  24. Katsouyanni, Klea. 2003. Ambient air pollution and health. BMJ 68: 143–156  25. Brook, Robert D., Barry Franklin, Wayne Cascio, Yuling Hong, George Howard, Michael Lipsett, Russell Luepker, Murray Mittleman, Jonathan Samet, Sidney C. Smith, Jr, Ira Tager. 2004. Air Pollution and Cardiovascular Disease. Circulation 109:2655-2671  26. Bhatnagar, Aruni. 2004. Cardiovascular pathophysiology of environmental pollutants. J Physiol Heart Circ Physiol 286: H479–H485  27. Bell, Michelle L., Jonathan M. Samet, and Francesca Dominici. 2004. Time-Series Studies of Particulate Matter. Annu. Rev. Public Health. 25:247–80  28. HEI Health Review Committee and Scientific Staff. Understanding the health effects of components of the particulate matter mix: progress and next steps. HEI Perspectives. Cambridge, Mass: Health Effects Institute; April 2002.  29. Wichmann HE, Mueller W, Allhoff P, et al. 1989. Health effects during a smog episode in West Germany in 1985. Environ Health Perspect. 79:89–99.  30. Schwartz J, Marcus A. 1990. Mortality and air pollution in London: a time series analysis. Am J Epidemiol. 131(1) :185–94.  31. Katsouyanni K, Karakatsani A, Messari I, et al. 1990. Air pollution and cause specific mortality in Athens. J Epidemiol Community Health. 44:321–4.   77 32. Kinney PL, Ozkaynak H. 1991. Associations of daily mortality and air pollution in Los Angeles County. Environ Res 54:99–120.  33. Schwartz J, Dockery DW. 1992. Increased mortality in Philadelphia associated with daily air pollution concentrations. Am Rev Respir Dis. 145:600–4.  34. Schwartz J. 1994. What are people dying of on high air pollution days? Environ Res. 64:26–35.  35. Schwartz J. 1994. Air pollution and daily mortality: a review and meta analysis. Environ Res. 64:36–52. 36. Anderson HR, Ponce de Leon A, Bland JM, et al. 1996. Air pollution and daily mortality in London: 1987– 92. BMJ 312:665–9.  37. Ponka A, Savela M, Virtanen M. 1998. Mortality and air pollution in Helsinki. Arch Environ Health 53:281– 6.  38. Zmirou D, Schwartz J, Saez M, et al. 1998. Time-series analysis of air pollution and cause-specific mortality. Epidemiology 9:495–503.  39. Ostro B, Chestnut L, Vichit-Vadakan N, et al. 1999. The impact of particulate matter on daily mortality in Bangkok, Thailand. J Air Waste Manag Assoc. 49:100–7.  40. Rossi G, Vigotti MA, Zanobetti A, et al. 1999. Air pollution and cause-specific mortality in Milan, Italy, 1980–1989. Arch Environ Health 54:158–64.  41. Samet JMD, Curriero F, Coursac FC, et al. 2000. Fine particulate air pollution and mortality in 20 U.S.cities 1987–1994. N Engl J Med 343:1742–9.  42. Checkoway, H., Drew Levy, Lianne Sheppard, Joel Kaufman, Jane Koenig, and David Siscovick. 2000. A Case-Crossover Analysis of Fine Particulate Matter Air Pollution and Out-of-Hospital Sudden Cardiac Arrest. Health Effects Institute, Research Report 99  43. Peters, A., Emerson Liu, Richard L. Verrier, Joel Schwartz, Diane R. Gold, Murray Mittleman, Jeff Baliff, J. Annie Oh, George Allen, Kevin Monahan, and Douglas W. Dockery. 2000. Air Pollution and Incidence of Cardiac Arrhythmia. Epidemiology 11: 11-17  44. Kwon HJ, Cho SH, Nyberg F, et al. 2001. Effects of ambient air pollution on daily mortality in a cohort of patients with congestive heart failure. Epidemiology 12:413–9.  45. Katsouyanni K, Touloumi G, Samoli E, et al. 2001. 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 12:521–31.  46. Peters, A., Douglas W. Dockery, James E. Muller, Murray A. Mittleman. 2001. Increased Particulate Air Pollution and the Triggering of Myocardial Infarction. Circulation 103:2810-2815  47. Goldberg, Mark S., Richard T. Burnett, Jeffrey Brook, John C. Bailar III, Marie-France Valois, and Renaud Vincent. 2001. Associations between Daily Cause-specific Mortality and Concentrations of Ground-level Ozone in Montreal, Quebec. Am J Epidemiol 154:817–26  48. Goldberg, Mark S., Richard T. Burnett, John C. Bailar, III, Jeffrey Brook, Yvette Bonvalot, Robyn Tamblyn, Ravinder Singh, Marie-France Valois, and Renaud Vincent. 2001. The Association between Daily Mortality and Ambient Air Particle Pollution in Montreal, Quebec 2. Cause-Specific Mortality. Environmental Research Section A 86: 26-36  49. Braga, Alfésio Luís Ferreira, Antonella Zanobetti, Joel Schwartz. 2001. The Lag Structure Between Particulate Air Pollution and Respiratory and Cardiovascular Deaths in 10 US Cities. J Occup Environ Med. 43:927–933  78  50. Hong, Yun-Chul, Jong-Tae Lee, Ho Kim and Ho-Jang Kwon. 2002. Air Pollution A New Risk Factor in Ischemic Stroke Mortality. Stroke 33:2165-2169  51. D’Ippoliti, D., Francesco Forastiere, Carla Ancona, Nera Agabiti, Danilo Fusco, Paola Michelozzi, and Carlo A. Perucci. 2003. Air Pollution and Myocardial Infarction in Rome: A Case-Crossover Analysis. Epidemiology 14: 528–535  52. Sullivan, J., N. Ishikawa, L. Sheppard, D. Siscovick, H. Checkoway, and J. Kaufman. 2003. Exposure to Ambient Fine Particulate Matter and Primary Cardiac Arrest among Persons With and Without Clinically Recognized Heart Disease. Am J Epidemiol 157:501–509  53. Dominici, F., Aidan McDermott, Scott L. Zeger, and Jonathan M. Samet. 2003. National Maps of the Effects of Particulate Matter on Mortality Exploring Geographical Variation. Environ Health Perspect 111:39– 43  54. Bateson, Thomas F. and Joel Schwartz. 2004. Who is sensitive to the effects of particulate air pollution on mortality? A Case-Crossover Analysis of effect Modifiers. Epidemiology 15: 143–149  55. Zanobetti, A., Marina Jacobson Canner, Peter H. Stone, Joel Schwartz, David Sher, Elizabeth Eagan- Bengston, Karen A. Gates, L. Howard Hartley, Helen Suh and Diane R. Gold. 2004. Ambient Pollution and Blood Pressure in Cardiac Rehabilitation Patients. Circulation 110:2184-2189.  56. Daniels, Michael J, Francesca Dominici, Scott L Zeger, and Jonathan M Samet. 2004. National Morbidity, Mortality, and Air Pollution Study. Part III Concentration-Response Curves and Thresholds for the 20 Largest US Cities. Health Effects Institute Research Paper 94, Part III.  57. Bell, Michelle L., Jonathan M. Samet, Francesca Dominici. Ozone and Mortality: A Meta-Analysis of Time- Series Studies and Comparison to a Multi-City Study (The National Morbidity, Mortality and Air Pollution Study).2004. Johns Hopkins University, Dept. of Biostatistics Working Papers Paper 57 26 pp.  58. Schwartz, Joel. 2004. The effects of particulate air pollution on daily deaths a multi-city case crossover analysis. Occup. Environ. Med. 61: 956-961  59. Forastiere, F., Massimo Stafoggia, Sally Picciotto, Tom Bellander, Daniela D’Ippoliti, Timo Lanki, Stephanie von Klot, Fredrik Nyberg, Pentti Paatero, Annette Peters, Juha Pekkanen, Jordi Sunyer, and Carlo A. Perucci. 2005. A Case-Crossover Analysis of Out-of-Hospital Coronary Deaths and Air Pollution in Rome, Italy. Am J Respir Crit Care Med 172: 1549–1555  60. Schwartz, Joel. 2005. How Sensitive Is the Association between Ozone and Daily Deaths to Control for Temperature. Am J Respir Crit Care Med 171: 627–631  61. Ruidavets, Jean-Bernard, Maxime Cournot, Sylvie Cassadou, Michel Giroux, Mariam Meybeck, Jean Ferrières. 2005. Ozone Air Pollution Is Associated With Acute Myocardial Infarction. Circulation 111:563-569  62. Zeka, A., A Zanobetti and J Schwartz. 2005. Short term effects of particulate matter on cause specific mortality: effects of lags and modification by city characteristics. Occup. Environ. Med. 62:718-725  63. Murakami, Yoshitaka and Masaji Ono. 2006. Myocardial infarction deaths after high level exposure to particulate matter.J. Epidemiol. Community Health 60:262-266  64. Martins L.C., L.A.A Pereira, CA Lin, UP Santos, G Prioli, O do Carmo Luiz, PHN Saldiva, AL Ferreira Braga. 2006. The effects of air pollution on cardiovascular diseases: lag structures. Rev Saúde Pública 40(4):XX- XX    79 65. Analitis, A., Klea Katsouyanni, Konstantina Dimakopoulou, Evangelia Samoli, Aristidis K. Nikoloulopoulos, Yannis Petasakis, Giota Touloumi, Joel Schwartz, Hugh Ross Anderson, Koldo Cambra, Francesco Forastiere, Denis Zmirou, Judith M. Vonk, Luke Clancy, Bohumir Kriz, Janos Bobvos, and Juha Pekkanen. 2006. Short-Term Effects of Ambient Particles on Cardiovascular and Respiratory Mortality. Epidemiology 17(2): 230-233  66. Schwartz J, Morris R. 1995. Air pollution and hospital admissions for cardiovascular disease in Detroit, Michigan. Am J Epidemiol. 142:23–35.  67. Burnett RT, Dales R, Krewski D, et al. 1995. Associations between ambient particulate sulfate and admissions to Ontario hospitals for cardiac and respiratory diseases. Am J Epidemiol. 142:15–22.  68. Morris RD, Naumova EN, Munasinghe RL. 1995. Ambient air pollution and hospitalization for congestive heart failure among elderly people in seven large US cities. Am J Public Health 85:1361–5.  69. Wordley J, Walters S, Ayres JG. 1997. Short term variations in hospital admissions and mortality and particulate air pollution. Occup Environ Med 54:108–16.  70. Schwartz J. 1997. Air pollution and hospital admissions for cardiovascular disease in Tucson. Epidemiology 8:371–7.  71. Burnett RT, Dales RE, Brook JR, et al. 1997. Association between ambient carbon monoxide levels and hospitalizations for congestive heart failure in the elderly in 10 Canadian cities. Epidemiology 8:162–7.  72. Burnett RT, Cakmak S, Raizenne ME, et al. 1998. The association between ambient carbon monoxide levels and daily mortality in Toronto, Canada. J Air Waste Manage Assoc 48:689–700.  73. Poloniecki JD, Atkinson RW, de LAP, et al. 1997. Daily time series for cardiovascular hospital admissions and previous day’s air pollution in London, UK. Occup Environ Med 54:535–40.  74. Schwartz J. 1999. Air pollution and hospital admissions for heart disease in eight U.S.counties. Epidemiology 10:17–22.  75. Wong, Chit-Ming, Richard W Atkinson, H. Ross Anderson, Anthony Johnson Hedley, Stefan Ma, Patsy Yuen-Kwan Chau, and Tai-Hing Lam. 2002. A tale of two cities effects of air pollution on hospital admissions in Hong Kong and London compared .Environ Health Perspect 110:67–77  76. Jordi Sunyera, Ferran Ballester, Alain Le Tertre, Richard Atkinson, Jon G. Ayres, Francesco Forastiere, Bertil Forsberg, Judith M. Vonk, Luigi Bisanti, Jose M. Tenıas, Sylvia Medina, Joel Schwartz, Klea Katsouyanni. 2003. The association of daily sulphur dioxide air pollution levels with hospital admissions for cardiovascular diseases in Europe (The Aphea-II study). European Heart Journal 24: 752–760  77. Zanobetti, Antonella and Joel Schwartz. 2005. The Effect of Particulate Air Pollution On Emergency Admissions for Myocardial Infarction: A Multi-city Case-crossover Analysis. Environ Health Perspect 113:978– 982  78. Symons, J. M., L. Wang, E. Guallar, E. Howell, F. Dominici, M. Schwab, B. A. Ange, J. Samet, J. Ondov, D. Harrison, and A. Geyh. 2006. A Case-Crossover Study of Fine Particulate Matter Air Pollution and Onset of Congestive Heart Failure Symptom Exacerbation Leading to Hospitalization. Am J Epidemiol 164:421–433  79. Clancy, Luke, Pat Goodman, Hamish Sinclair, Douglas W DockeryPat Goodman, Hamish Sinclair, Douglas W Dockery. 2002. Effect of air-pollution control on death rates in Dublin, Ireland: an intervention study. Lancet 360: 1210–14.  80. Hedley, Anthony Johnson, Chit-Ming Wong, Thuan Quoc Thach, Stefan Ma, Tai-Hing Lam, Hugh Ross Anderson. 2002. Cardiorespiratory and all-cause mortality after restrictions on sulphur content of fuel in Hong Kong: an intervention study. Lancet 360: 1646–52.  80 81. Laden, Francine, Joel Schwartz, Frank E. Speizer, and Douglas W. Dockery. 2006. Reduction in fine particulate air pollution and mortality: extended follow-up of the Harvard Six Cities study. Am J Respir Crit Care Med 173: 667–672.  82. Dockery, Douglas W., C. Arden Pope, Xiping Xu, John D. Spengler, James H. Ware, Martha E. Fay, Benjamin G. Ferris, and Frank E. Speizer. 1993. An Association between Air Pollution and Mortality in Six U.S. Cities. N Engl J Med 329(24):1753-1759.  83. Pope, CA, MJ Thun, MM Namboodiri, DW Dockery, JS Evans, FE Speizer and CW Heath Jr. 1995. Particulate air pollution as a predictor of mortality in a prospective study of U.S. adults. Am J Respir Crit Care Med. 151(3):669–674.  84. Pope III, C. Arden, Richard T. Burnett, Michael J. Thun, Eugenia E. Calle, Daniel Krewski, Kazuhiko Ito, and George D. Thurston. 2002. Lung cancer, cardiopulmonary mortality, and long-term exposure to fine particulate air pollution. JAMA 287: 1132-1141  85. Pope, III C. Arden, Richard T. Burnett, George D. Thurston, Michael J. Thun, Eugenia E. Calle, Daniel Krewski and John J. Godleski. 2004. Cardiovascular Mortality and Long-Term Exposure to Particulate Air Pollution Epidemiological Evidence of General Pathophysiological Pathways of Disease. Circulation 109:71-77  86. Nafstad, P., Lise Lund Håheim, Torbjørn Wisløff, Frederick Gram, Bente Oftedal, Ingar Holme, Ingvar Hjermann, and Paul Leren. 2004. Urban air pollution and mortality in a cohort of Norwegian men. Environ Health Perspect 112:610–615  87. Künzli, N., Michael Jerrett, Wendy J. Mack, Bernardo Beckerman, Laurie LaBree, Frank Gilliland, Duncan Thomas, John Peters, and Howard N. Hodis. 2005. Ambient Air Pollution and Atherosclerosis in Los Angeles. Environ Health Perspect 113:201–206  87a. Jerrett, M., Richard T. Burnett, Renjun Ma, C. Arden Pope III, Daniel Krewski, K. Bruce Newbold, George Thurston, Yuanli Shi, Norm Finkelstein, Eugenia E. Calle, and Michael J. Thun. 2005. Spatial Analysis of Air Pollution and Mortality in Los Angeles. Epidemiology. 16: 727–736.  87b. Rosenlund, M., Sally Picciotto, Francesco Forastiere, Massimo Stafoggia, and Carlo A. Perucci. 2008. Traffic-Related Air Pollution in Relation to Incidence and Prognosis of Coronary Heart Disease. Epidemiology. 19: 121–128.  87c. Künzli N, Jerrett M, Garcia-Esteban R, Basagaña X, Beckermann B, Gilliland F, Medina M, Peters J, Hodis HN, Mack WJ. 2010. Ambient air pollution and the progression of atherosclerosis in adults. PLoS One. 2010 Feb 8;5(2):e9096.  88. Miller, Kristin A., David S. Siscovick, Lianne Sheppard, Kristen Shepherd, Jeffrey H. Sullivan, Garnet L. Anderson, and Joel D. Kaufman. 2007. Long-term exposure to air pollution and incidence of cardiovascular events in women. N Engl J Med 356:447-58.  88a. Gehring Ulrike, Joachim Heinrich, Ursula Kramer, Veit Grote, Matthias Hochadel, Dorothea Sugiri, Martin Kraft, Knut Rauchfuss, Hans Georg Eberwein, and H.-Erich Wichmann. 2006. Long-Term Exposure to Ambient Air Pollution and Cardiopulmonary Mortality in Women. Epidemiology 17: 545-551.  88b. Beelen, R, Gerard Hoek, Piet A. van den Brandt, R. Alexandra Goldbohm, Paul Fischer, Leo J. Schouten, Michael Jerrett, Edward Hughes, Ben Armstrong, and Bert Brunekreef. 2008. Long-Term Effects of Traffic- Related Air Pollution on Mortality in a Dutch Cohort (NLCS-AIR Study). Environ Health Perspect 116:196–202;  88c. Chen Lie Hong, Synnove F. Knutsen, David Shavlik, W. Lawrence Beeson, Floyd Petersen, Mark Ghamsary, and David Abbey. 2005. The Association between Fatal Coronary Heart Disease and Ambient Particulate Air Pollution: Are Females at Greater Risk? Environ Health Perspect 113:1723–1729   81 89. Rijnders, Esther, Nicole A.H. Janssen, Patricia H.N. van Vliet, and Bert Brunekreef. 2001. Personal and Outdoor Nitrogen Dioxide Concentrations in Relation to Degree of Urbanization and Traffic Density. Environ Health Perspect 109(suppl 3):411–417.  90. Hoek G., Bert Brunekreef, Sandra Goldbohm, Paul Fischer, Piet A van den Brandt. 2002. Association between mortality and indicators of traffic-related air pollution in the Netherlands- a cohort study. Lancet 360: 1203–09.  91. Peters, A., Stephanie von Klot, Margit Heier, Ines Trentinaglia, Allmut Hörmann, H. Erich Wichmann, and Hannelore Löwel. 2004. Exposure to traffic and the onset of myocardial infarction. N Engl J Med 351:1721-30.  92. Finkelstein, Murray M., Michael Jerrett, and Malcolm R. Sears. 2004. Traffic Air Pollution and Mortality Rate Advancement Periods. Am J Epidemiol 160:173–177.  93. Hoffmann, B., Susanne Moebus, Andreas Stang, Eva-Maria Beck, Nico Dragano, Stephan Mohlenkamp, Axel Schmermund, Michael Memmesheimer, Klaus Mann, Raimund Erbel, and Karl-Heinz Jockel. 2006. Residence close to high traffic and prevalence of coronary heart disease. European Heart Journal 27: 2696– 2702  93a. Hoffmann, B., S. Moebus, S. Möhlenkamp, A. Stang, N. Lehmann, N. Dragano, A. Schmermund, M. Memmesheimer, K. Mann, R. Erbel, K.-H. Jöckel. 2007. Residential Exposure to Traffic Is Associated With Coronary Atherosclerosis. Circulation. 116:1-8.  93b. Medina-Ramón, M., Robert Goldberg, Steven Melly, Murray A. Mittleman, and Joel Schwartz. 2008. Residential Exposure to Traffic-Related Air Pollution and Survival after Heart Failure. Environ Health Perspect 116:481–485.  93c. Tonne, C ., J Yanosky, A Gryparis, S Melly,2 M Mittleman, R Goldberg, S von Klot, J Schwartz. 2009. Traffic particles and occurrence of acute myocardial infarction: a case–control analysis. Occup Environ Med. 66:797–804.  93d. Tonne, C. Steve Melly, Murray Mittleman, Brent Coull, Robert Goldberg, and Joel Schwartz. 2007. A Case–Control Analysis of Exposure to Traffic and Acute Myocardial Infarction. Environ Health Perspect. 115:53–57.  94. Seaton A, MacNee W, Donaldson K, et al. 1995. Particulate air pollution and acute health effects. Lancet 345:176–8.  95. Ferreiros EB, Boissonnet CP, Pizarro R, et al. 1999. Independant prognostic value of elevated C-reactive protein in unstable angina. Circulation 100:1958–63.  97. Biasucci LLG. 1999. Elevated levels of C-reactive protein at discharge in patients with unstable angina predict recurrent instability. Circulation 99:855–60.  97. Ridker PM, Hennekens CH, Buring JE, et al. 2000. C-reactive protein and other markers of inflammation in the prediction of cardiovascular disease in women. N Engl J Med 342:836–43.  98. Nemmar A, Vanbilloen H, Hoylaerts MF, et al. 2001. Passage of intratracheally instilled ultrafine particles from the lung into the systemic circulation in hamster. Am J Respir Crit Care Med. 164:1665–1668.  99. Nemmar A, Hoet PH, Vanquickenborne B, et al. 2002. Passage of inhaled particles into the blood circulation in humans. Circulation. 105:411–414.  100. Oberdorster G, Sharp Z, Atudorei V, et al. 2002. Extrapulmonary translocation of ultrafine carbon particles following whole-body inhalation exposure of rats. J Toxicol Environ Health A. 65:1531–1543.   82 101. Schwartz P, Vanoli E, Stramba-Badiale M, et al. 1988. Autonomic mechanisms and sudden death: new insights from analysis of baroreceptor reflexes in conscious dogs with and without a myocardial infarction. Circulation 78:969–79.  102. Godleski JJ, Verrier RL, Koutrakis P, et al. 2000. Mechanisms of morbidity and mortality from exposure to ambient air particles. Res Rep Health Eff Inst Feb:5–88.  103. La Rovere M, Bigger JT Jr, Marcus FI, et al. 1998. Baroreflex sensitivity and heartrate variability in prediction of total cardiac mortality after myocardial infarction. ATRAMI (autonomic tone and reflexes after myocardial infarction) Investigators. Lancet 351:478–84.  104. Nolan J, Batin PD, Andrews R, et al. 1998. Prospective study of heart rate variability and mortality in chronic heart failure: results of the United Kingdom heart failure evaluation and assessment of risk trial (UK- heart). Circulation 98:1510–6.  105. Anderson, Norman B. and Armstead, Cheryl A. 1995. Toward Understanding the Association of Socioeconomic Status and Health: A New Challenge for the Biopsychosocial Approach. Psychosom Med 57(3):213-225  106. Fox AJ, Goldblatt P, Jones D. 1985.Social class mortality differentials: Artefact, selection or life circumstance. J Epidemiol Community Health 39:1-8  107. Pincus T, Kallahan L, Burkhauser R. 1987. Most chronic diseases are reported more frequently by individuals with fewer than 12 years of formal education in the ages 18-64 in the United States population. J Chronic Dis 40:865-874  108. Ruberman W, Weinblatt E, Goldberg JD, et al. 1984. Psychosocial influences on mortality after myocardial infarction. N Engl J Med 11:552-559  109. Williams RB, Barefoot JC, Kaliff RM, et al.1992. Prognostic significance of social and economic resources among medically treated patients with angiographically documented coronary artery disease. JAMA 267:520- 524  110. Marmot MG, Kogevinas M, Elston MA. 1987. Social/economic status and disease. Annu Rev Public Health 8:111-135  111. Williams, DR. 1990. Socioeconomic differentials in health: A review and redirection. Soc Psychol Q 32:81- 99,  112. Kaplan, GA and JE Keil. 1993. Socioeconomic factors and cardiovascular disease: a review of the literature. Circulation 88:1973-1998  113. Carroll D, Bennett P, Davey Smith G. 1993. Socio-economic health inequities: Their origins and implications. Psychol Health 8:295-316  114. Adler N, Boyce T, Chesney M, et al. 1993. Socioeconomic inequities in health: No easy solution. JAMA 269:3140-3145  115. Adler N, Boyce T, Chesney M, et al. 1994. Socioeconomic status and health: The challenge of the gradient. Am Psychol 49:15-24  116. Yen, I. H. and S. L. Syme. 1999. The Social Environment and Health: A Discussion of the Epidemiologic Literature. Annu. Rev. Public Health. 20:287–308  117. Marmot, Michael. 2001. Inequalities in health. N Engl J Med 345(2): 134-136   83 118. Small, Mario Luis and Katherine Newmann. 2001. Urban Poverty after The Truly Disadvantaged The Rediscovery of the Family, the Neighborhood, and Culture. Annual Review of Sociology 27: 23-45.  119. Steptoe, A. and M. Marmot. 2002. The role of psychobiological pathways in socio-economic inequalities in cardiovascular disease risk. European Heart Journal 23:13–25  120. Isaacs, Stephen L. and Steven A. Schroeder. 2004. Class — The Ignored Determinant of the Nation’s Health. N Engl J Med 351(11): 1137-1142  121. Benjamin, Emelia J., Mariell Jessup, John M. Flack, Harlan M. Krumholz, Kiang Liu, Vinay M. Nadkarni, Dorothy A. Rhoades, Beatriz L. Rodriguez, Rosalyn P. Scott, Malcolm P. Taylor, Eric J. Velazquez and Marilyn A. Winkleby. 2005. Discovering the full spectrum of cardiovascular disease: Minority Health Summit 2003: Report of the Outcomes Writing Group. Circulation 111:124-133  122. Wilkinson, Richard G. and Kate E. Pickett. 2006. Income inequality and population health: A review and explanation of the evidence. Social Science & Medicine 62: 1768–1784 123. Krieger, Nancy. 1992. Overcoming the Absence of Socioeconomic Data in Medical Records: Validation and Application of a Census-Based Methodology. Am J Public Health 92:703-710  124. Wing, S., Michele Casper, Wilson Riggan, Carl Hayes, and H. A. Tyroler. 1988. Socioenvironmental Characteristics Associated with the Onset of Decline of Ischemic Heart Disease Mortality in the United States. Am J Public Health 78: 928-926  125. Luepker, RV., WD Rosamond, R Murphy, JM Sprafka, AR Folsom, PG McGovern and H Blackburn. 1993. Socioeconomic status and coronary heart disease risk factor trends. The Minnesota Heart Survey. Circulation 88:2172-2179  126. Anderson, Norman B. and Armstead, Cheryl A. 1995. Toward Understanding the Association of Socioeconomic Status and Health: A New Challenge for the Biopsychosocial Approach. Psychosom Med 57(3):213-225  127. Mackenbach, Johan P., Anton E Kunst, Adriënne E J M Cavelaars, Feikje Groenhof, José J M Geurts. 1997. Socioeconomic inequalities in morbidity and mortality in Western Europe. Lancet 349: 1655–59  128. Lynch, J., Niklas Krause, George A. Kaplan, Jaakko Tuomilehto, and Jukka T Salonen. 1997. Workplace conditions, socioeconomic status, and the risk of mortality and acute myocardial infarction the Kuopio Ischemic Heart Disease Risk Factor Study. Am J Public Health 87: 617-622  129. Ribisl, Kurt M., Marilyn A. Winkleby, Stephen P. Fortmann and June A. Flora. 1998. The interplay of socioeconomic status and ethnicity on Hispanic and White men's cardiovascular disease risk and health communication patterns . Health Education Researh. Theory&Practice 13(3): 407-417  130. Hemingway, Harry and Michael Marmot. 1999. Psychosocial factors in the aetiology and prognosis of coronary heart disease - systematic review of prospective cohort studies. BMJ 318: 1460-1467  131. Tyroler, Herman A.. 1999. The influence of socioeconomic factors on cardiovascular disease risk factor development. Preventive Medicine 29:S36–S40  132. Ross, Nancy A., Michael C Wolfson, James R Dunn, Jean­Marie Berthelot, George A Kaplan, John W Lynch. 2000. Relation between income inequality and mortality in Canada and in the United States - cross sectional assessment using census data and vital statistics. BMJ 320: 898-902       84 133. Veikko Salomaa, Matti Niemela, Heikki Miettinen, Matti Ketonen, Pirjo Immonen-Raiha, Seppo Koskinen, Markku Mahonen, Seppo Lehto, Tapio Vuorenmaa, Pertti Palomaki, Harri Mustaniemi, Esko Kaarsalo, Matti Arstila, Jorma Torppa, Kari Kuulasmaa, Pekka Puska, Kalevi Pyorala, Jaakko Tuomilehto. 2000. Relationship of Socioeconomic Status to the Incidence and Prehospital, 28-Day, and 1-Year Mortality Rates of Acute Coronary Events in the FINMONICA Myocardial Infarction Register Study. Circulation 101:1913-1918  134. Bartley, M., R Fitzpatrick, D Firth and M Marmot. 2000. Social distribution of cardiovascular disease risk factors: change among men in England 1984-1993. J Epidemiol Community Health 54:806-814  135. Chandola, Tarani and Crispin Jenkinson. 2000. The UK National Statistic SE Classification. Social class and health. Journal of Public Health Medicine 22(2): 182-190  136. Anand, Sonia S., Salim Yusuf, Ruby Jacobs, A Darlene Davis, Qilong Yi, Hertzel Gerstein, Patricia A Montague, Eva Lonn,. 2001. Risk factors, atherosclerosis, and cardiovascular disease among Aboriginal people in Canada: the Study of Health Assessment and Risk Evaluation in Aboriginal Peoples (SHARE-AP). Lancet 358: 1147–53  137. Cooper, Richard S. 2001. Social inequality, ethnicity and cardiovascular disease. Int. J. Epidemiol 30: S48- S52  138. Ferrario, M., Roberto Sega, Liliana Chatenoud, Giuseppe Mancia, Paolo Mocarelli, Carla Crespi, and Giancarlo Cesana. 2001. Time trends of major coronary risk factors in a northern Italian population (1986- 1994). How remarkable are socioeconomic differences in an industrialized low CHD incidence country? International Journal of Epidemiology 30: 285-291  139. Rooks, Ronica N., Eleanor M. Simonsick, Toni Miles, Anne Newman, Stephen B. Kritchevsky, Richard Schulz, and Tamara Harris. 2002. The association of race and socioeconomic status with cardiovascular disease indicators among older adults in the health, aging, and body composition study. Journal of Gerontology 57B(4): S247–S256  140. Diez Roux, Ana V. 2003. Residential environments and cardiovascular risk. Journal of Urban Health: Bulletin of the New York Academy of Medicine 80(4): 569-589  141. von dem Knesebeck, O., G.unther L.uschen, William C. Cockerham, Johannes Siegrist. 2003. Socioeconomic status and health among the aged in the United States and Germany: a comparative cross- sectional study. Social Science & Medicine 57: 1643–1652  142. Jerrett, M., Richard T. Burnett, Alette Willis, Daniel Krewski, Mark S. Goldberg, Patrick DeLuca, Norm Finkelstein. 2003. Spatial Analysis of the Air Pollution-Mortality Relationship in the Context of Ecologic Confounders. Journal of Toxicology and Environmental Health, Part A, 66:1735–1777  143. Nordstrom, Cheryl K., Ana V. Diez Roux, Sharon A. Jackson, Julius M. Gardin. 2004. The association of personal and neighborhood socioeconomic indicators with subclinical cardiovascular disease in an elderly cohort. The cardiovascular health study. Social Science & Medicine 59: 2139–2147  144. Roos, Leslie L., Randy Walld, Julia Uhanova, and Ruth Bond. 2005. Physician visits, hospitalizations, and socioeconomic status ambulatory care sensitive conditions in a Canadian setting. HSR: Health Services Research 40(4):1167-1185  145. Cubbin, Catherine and Marilyn A. Winkleby. 2005. Protective and harmful effects of neighborhood-level deprivation on individual-level health knowledge, behavior changes, and risk of coronary heart disease. Am J Epidemiol 162:559–568  146. Davidson, Pamela L., Roshan Bastani, Terry T. Nakazono, Daisy C. Carreon. 2005. Role of community risk factors and resources on breast carcinoma stage at diagnosis. Cancer 103:922–30.   85 147. Schulz, Amy J., Srimathi Kannan, J. Timothy Dvonch, Barbara A. Israel, Alex Allen III, Sherman A. James, James S. House, and James Lepkowski. 2005. Social and Physical Environments and Disparities in Risk for Cardiovascular Disease: The Healthy Environments Partnership Conceptual Model. Environ Health Perspect 113:1817–1825  148. Braveman, Paula A., Catherine Cubbin, Susan Egerter, Sekai Chideya, Kristen S. Marchi, Marilyn Metzler, Samuel Posner. 2005. Socioeconomic status in health research. One size does not fit all. JAMA. 294:2879-2888  149. Mensah, George A., Ali H. Mokdad, Earl S. Ford, Kurt J. Greenlund and Janet B. Croft. 2005. State of Disparities in Cardiovascular Health in the United States . Circulation 111:1233-1241  150. Veenstra, Garry. 2005. Location, location, location: contextual and compositional health effects of social capital in British Columbia, Canada. Social Science & Medicine 60: 2059–2071  151. Janes, Craig R. 2006. `Culture', cultural explanations and causality. International Journal of Epidemiology 35:261–263  152. Glass, Thomas A. 2006. Culture in epidemiology-the 800 pound gorilla. International Journal of Epidemiology 35:259–261  153. Eckersley, Richard. 2006. Is modern Western culture a health hazard? International Journal of Epidemiology 35:252–258  154. Chen, Jarvis T., David H. Rehkopf, Pamela D. Waterman, S. V. Subramanian, Brent A. Coull, Bruce Cohen, Mary Ostrem, and Nancy Krieger. 2006. Mapping and Measuring Social Disparities in Premature Mortality: The Impact of Census Tract Poverty within and across Boston Neighborhoods, 1999–2001. Journal of Urban Health: Bulletin of the New York Academy of Medicine, Vol. 83, No. 6 1063-1084  155. Anand, Sonia S., Fahad Razak, AD Davis, Ruby Jacobs, Vlad Vuksan, Koon Teo1, and Salim Yusuf. 2006. Social disadvantage and cardiovascular disease development of an index and analysis of age, sex, and ethnicity effects. International Journal of Epidemiology 35:1239–1245  156. Kohlhuber, M., Andreas Mielck, Stephan K. Weiland, Gabriele Bolte. 2006. Social inequality in perceived environmental exposures in relation to housing conditions in Germany. Environmental Research 101: 246–255  157. Messer, Lynne C., Barbara A. Laraia, Jay S. Kaufman, Janet Eyster, Claudia Holzman, Jennifer Culhane, Irma Elo, Jessica G. Burke, and Patricia O’Campo. 2006. The development of a standardized neighborhood deprivation index . Journal of Urban Health: Bulletin of the New York Academy of Medicine, 83(6): 1041-1062  158. Matheson, Flora I., Rahim Moineddin, James R. Dunn, Maria Isabella Creatore, Piotr Gozdyra, Richard H. Glazier. 2006. Urban neighborhoods, chronic stress, gender and depression. Social Science & Medicine 63: 2604–2616  159. Warra, Deborah J., Theonie Tacticosb, Margaret Kelaherb, Harald Klein. 2007. `Money, stress, jobs' Residents' perceptions of health-impairing factors in `poor' neighbourhoods. Health & Place 13: 743–756  160. Cerina, E., Eva Leslie, Lorinne du Toit, Neville Owen, Lawrence D. Frank. 2007. Destinations that matter: Associations with walking for transport. Health & Place 13: 713–724  161. Safaei, Jalil. 2007. Income and health inequality across Canadian provinces. Health & Place 13:629–638  162. Chaix, B., Maria Rosvall, and Juan Merlo. 2007. Recent increase of neighborhood socioeconomic effects on ischemic heart disease mortality: a multilevel survival analysis of two large Swedish cohorts. Am J Epidemiol 165:22–26   86 163. Turrell, G., John W Lynch, Claudia Leite, Trivellore Raghunathan, George A Kaplan. 2007. Socioeconomic disadvantage in childhood and across the life course and all-cause mortality and physical function in adulthood: evidence from the Alameda County Study. J Epidemiol Community Health 61:723-730  164. Pickett, KE, and M Pearl. 2001. Multilevel analyses of neighbourhood socioeconomic context and health outcomes: a critical review. J Epidemiol Community Health 55:111-122  165. Sampson, Robert J., Jeffrey D. Morenoff, and Thomas Gannon-Rowley. 2002. Assessing neighborhood effects: Social Processes and new Directions in Research. Annu. Rev. Sociol. 28:443–78  166. Briggs, D., Daniela Fecht and Kees de Hoogh. 2007. Census data issues for epidemiology and health risk assessment- experiences from the Small Area Health Statistics Unit. J. R. Statist. Soc. A 170 Part 2: 355–378  167. Carpiano, Richard M.. 2006. Toward a neighborhood resource-based theory of social capital for health. Can Bourdieu and sociology help? Social Science & Medicine 62: 165–175  168. Carpiano, Richard M. 2007. Neighborhood social capital and adult health: An empirical test of a Bourdieu- based model. Health & Place 13: 639–655  169. Robert, Stephanie A. 1999. Socioeconomic Position and Health - The Independent Contribution of Community Socioeconomic Context. Annu. Rev. Sociol. 25:489-516  170. Cutchin, Malcolm P. 2007. The need for the "new health geography" in epidemiologic studies of environment and health. Health & Place 13: 725–742  171. Lynch, John W., George A. Kaplan, Richard D. Cohen, Jaakko Tuomilehto, and Jukka T. Salonen. 1996. Do Cardiovascular Risk Factors Explain the Relation between Socioeconomic Status, Risk of All-Cause Mortality, Cardiovascular Mortality, and Acute Myocardial Infarction? Am J Epidemiol 144:934-42  172. Vargas, Clemencia M., Vicki L. Burt, and Richard F. Gillum. 1997. Cardiovascular disease in the NHANES III. Ann Epidemiol 7: 523-525  173. Marmot, MG, H Bosma, H Hemingway, E Brunner, S Stansfeld. 1997. Contribution of job control and other risk factors to social variations in coronary heart disease incidence. Lancet 350: 235–39  174. Harrell, Joanne S. and Shirley V. Gore. 1998. Cardiovascular risk factors and socioeconomic status in African American and Caucasian women. Research in Nursing & Health 21: 285–295  175. Diez-Roux, Ana V., Bruce G. Link, Mary E. Northridge. 2000. A multilevel analysis of income inequality and cardiovascular disease risk factors. Social Science & Medicine 50: 673-687  176. Leander, K., Johan Hallqvist, Christina Reuterwall, Anders Ahlbom, and Ulf de Faire. 2001. Family History of Coronary Heart Disease, a Strong Risk Factor for Myocardial Infarction Interacting with Other Cardiovascular Risk Factors. Epidemiology 12: 215-221  177. Vitaliano, Peter P., James M. Scanlan, Jianping Zhang, Margaret V. Savage, Irl B. Hirsch, And Ilene C. Siegler. 2002. A Path Model of Chronic Stress, the Metabolic Syndrome, and Coronary Heart Disease. Psychosomatic Medicine 64:418–435  178. Mayer, Jr O., J Simon, J Heidrich, D V Cokkinos and D De Bacquer. 2004. Educational level and risk profile of cardiac patients in the EUROASPIRE II substudy. J. Epidemiol. Community Health 58:47-52  179. Warrena, John Robert, Peter Hoonakker, Pascale Carayon, Jennie Brand. 2004. Job characteristics as mediators in SES-health relationships. Social Science & Medicine 59: 1367–1378   87 180. Munoz, Miguel-Angel, Izabella Rohlfs, Sandra Masuet, Carolina Rebato, Marta Cabanero, Jaume Marrugat. 2005. Analysis of inequalities in secondary prevention of coronary heart disease in a universal coverage health care system. European Journal of Public Health 16(4): 361–367  181. Yarnell, J., S Yu, E McCrum, D Arveiler, B Haas, J Dallongeville, M Montaye, P Amouyel, J Ferrières, J-B Ruidavets, A Evans, A Bingham, P Ducimetière. 2005. Education, socioeconomic and lifestyle factors, and risk of coronary heart disease: the PRIME study. International Journal of Epidemiology 34:268–275  182. Eaker, Elaine D., Lisa M. Sullivan, Margaret Kelly-Hayes, Ralph B. D’Agostino, Sr, and Emelia J. Benjamin. 2007. Marital Status, Marital Strain, and Risk of Coronary Heart Disease or Total Mortality The Framingham Offspring Study. Psychosom Med. 69: 509-513  183. Wing, S., Elizabeth Barnett, Michele Casper, and H. A. Tyroler. 1992. Geographic and Socioeconomic Variation in the Onset of Decline of Coronary Heart Disease Mortality in White Women. Am J Public Health 82: 204-209  184. Diez-Roux, Ana V., F. Javier Nieto, Carles Muntaner, Herman A. Tyroler, George W. Comstock, Eyal Shahar, Lawton S. Cooper, Robert L. Watson, and Moyses Szklo. 1997. Neighborhood Environments and Coronary Heart Disease: A multilevel analysis. Am J. Epidemiol 146(1): 48-63  185. Hart, C., Russell Ecob, and George Davey Smith. 1997. People, places and coronary heart disease risk factors: A multilevel analysis of the Scottish heart health study archive. Soc. Sci. Med. 45(6): 893-902  186. Sundquist, J., Marianne Malmstrom and Sven-Erik Johansson. 1999. Cardiovascular risk factors and the neighbourhood environment: a multilevel analysis. International Journal of Epidemiology 28: 841-845  187. Marianne Malmstrom, Jan Sundquist, and Sven-Erik Johansson. 1999. Neighborhood Environment and Self-Reported Health Status: A Multilevel Analysis. Am J Public Health 89: 1181-1186  188. Diez-Roux, Ana V., Bruce G. Link, Mary E. Northridge. 2000. A multilevel analysis of income inequality and cardiovascular disease risk factors. Social Science & Medicine 50: 673-687  189. Lynch, John W., George Davey Smith, George A Kaplan, James S House. 2000. Income inequality and mortality - importance to health of individual income, psychosocial environment, or material conditions. BMJ 320:1200–4  190. Diez-Roux, Ana V., Sharon Stein Merkin, Donna Arnett, Lloyd Chambless, Mark Massing, Javier Nieto, Paul Sorlie, Moyses Szklo, Herman A. Tyroler, and Robert L. Watson. 2001. Neighborhood of Residence and Incidence of Coronary Heart Disease. N Engl J Med 345:99-106  191. Jerrett, M., R T Burnett, J Brook, P Kanaroglou, C Giovis, N Finkelstein, B Hutchison. 2004. Do socioeconomic characteristics modify the short term mortality? Evidence from a zonal time series in Hamilton, Canada. J Epidemiol Community Health 58:31–40  192. Allacci MaryAnn Sorensen. 2004. Identifying environmental risk factors for asthma emergency care- a multilevel approach for ecological study. J Ambulatory Care Manage 28(1): 2–15  193. Lopez, Russ. 2004. Income inequality and self-rated health in US metropolitan areas: A multi-level analysis. Social Science & Medicine 59: 2409–2419  194. Sundquist, K., Marilyn Winkleby, Helena Ahlén, and Sven-Erik Johansson. 2004. Neighborhood Socioeconomic Environment and Incidence of Coronary Heart Disease: A Follow-up Study of 25,319 Women and Men in Sweden. Am J Epidemiol 159:655–662  195. Gee, Gilbert C. and David T. Takeuchi. 2004. Traffic stress, vehicular burden and well-being: A multilevel analysis. Social Science & Medicine 59: 405–414   88 196. Subramanian, SV, Daniel Kim and Ichiro Kawachi. 2005. Covariation in the socioeconomic determinants of self rated health and happiness: a multivariate multilevel analysis of individuals and communities in the USA . J. Epidemiol. Community Health 59:664-669  197. Mujahid, Mahasin S., Ana V. Diez Roux, Luisa N. Borrell, and F. Javier Nieto. 2005. Cross-sectional and longitudinal associations of BMI with socioeconomic characteristics. Obesity Research 13(8): 1412-1421  198. Cakmak, S., Robert E. Dales, Stan Judek, and Frances Coates. 2005. Does socio-demographic status influence the effect of pollens and molds on hospitalization for asthma? Results from a time-series study in 10 Canadian cities. Ann Epidemiol 15:214–218.  199. Pickett, Kate E., Shona Kelly, Eric Brunner, Tim Lobstein and Richard G Wilkinson. 2005. Wider income gaps, wider waistbands: An ecological study of obesity and income inequality. J. Epidemiol. Community Health 59:670-674  200. Sloane, D., Lori Nascimento, Gwendolyn Flynn, LaVonna Lewis, Joyce Jones Guinyard, Lark Galloway- Gilliam, Allison Diamant, Antronette Yancey. 2006. Assessing resource environments to target prevention interventions in community chronic disease control . Journal of Health Care for the Poor and Underserved 17: 146–158.  201. Powell, Lisa M., Sandy Slater, Frank J. Chaloupka, and Deborah Harper. 2006. Availability of Physical Activity-Related Facilities and Neighborhood Demographic and Socioeconomic Characteristics. Am J Public Health 96: 1676-1680 202. Mobley, Lee R., Elisabeth D. Root, Eric A. Finkelstein, Olga Khavjou, Rosanne P. Farris, Julie C. Will. 2006. Environment, obesity, and cardiovascular disease risk in low-income women. Am J Prev Med 30(4):327– 332  203. Dahl, E., Jon Ivar Elstad, Dag Hofoss, Melissa Martin-Mollard. 2006. For whom is income inequality most harmful? A multi-level analysis of income inequality and mortality in Norway. Social Science & Medicine 63: 2562–2574  204. Wilkinson, Richard G., and Kate E. Pickett. 2006. Income inequality and population health: A review and explanation of the evidence . Social Science & Medicine 62: 1768–1784  205. McGrath, Jennifer J., Karen A. Matthews, Sonya S. Brady. 2006. Individual versus neighborhood socioeconomic status and race as predictors of adolescent ambulatory blood pressure and heart rate. Social Science & Medicine 63: 1442–1453  206. Cagney, Kathleen A.. 2006. Neighborhood age structure and its implications for health. Journal of Urban Health: Bulletin of the New York Academy of Medicine, 83(5): 827-834  207. Taylor, C. Barr, David Ahn, Marilyn A. Winkleby. 2006. Neighborhood and individual socioeconomic determinants of hospitalization. Am J Prev Med 31(2):127–134  208. Stjarne, Maria K., Johan Fritzell, Antonio Ponce De Leon, and Johan Hallqvist. 2006. Neighborhood socioeconomic context, individual income and myocardial infarction. Epidemiology 17: 14–23  209. Wen, M., Louise C. Hawkley, John T. Cacioppo. 2006. Objective and perceived neighborhood environment, individual SES and psychosocial factors, and self-rated health: An analysis of older adults in Cook County, Illinois. Social Science & Medicine 63: 2575–2590  210. Ross, Nancy A., Stephane Tremblay, Saeeda Khan, Daniel Crouse, Mark Tremblay, Jean-Marie Berthelot. 2007. Body mass index in urban Canada - Neighborhood and Metropolitan area effects. Am J Public Health 97: 500-508   89 211. Chang, Wei-Ching, Padma Kaul, Cynthia M. Westerhout, Michelle M. Graham, Paul W. Armstrong. 2007. Effects of socioeconomic status on mortality after acute myocardial infarction. The American Journal of Medicine 120: 33-39  212. Lisabeth, L.D., A. V. Diez Roux, J. D. Escobar, M. A. Smith, and L. B. Morgenstern. 2007. Neighborhood environment and risk of ischemic stroke. The brain attack surveillance in Corpus Christi (BASIC) Project. Am J Epidemiol 165:279–287  213. Chaix, B., Maria Rosvall, and Juan Merlo. 2007. Neighborhood socioeconomic deprivation and residential instability effects on incidence of ischemic heart disease and survival after myocardial infarction. Epidemiology 18: 104–111  214. Tsongalis, Gregory J.., William B. Coleman. 2006. Clinical genotyping: The need for integration of single nucleotide polymorphism and mutations in the clinical laboratory. Clinica Chimica Acta 363: 127-137.  215. Puderer, Henry. 2001. Introducing the Dissemination Area for the 2001 Census: an update. Geography Working Paper Series No 2000-4. Geography Division, Statistics Canada  216. Kershaw, P., Lori Irwin, Kate Trafford, Clyde Hertzman et al. 2005. The British Columbia Atlas of Child Development. Human Early Learning Partnership, UBC, Vancouver. Canadian western Geographical series 1203-1178, v. 40.  216a. Brauer M, Lencar C, Tamburic L, Koehoorn M, Demers P, Karr C. A Cohort Study of Traffic-Related Air Pollution Impacts on Birth Outcomes. Environmental Health Perspectives. 2008; 116:680–686.  217. Henderson, Sarah B., Bernardo Beckerman, Michael Jerrett, Michael Brauer. 2007. Application of land use regression to estimate long-term concentrations of traffic-related nitrogen oxides and fine particulate matter. Environ. Sci. Technol. 41: 2422-2428.  218. Gilbert NL, Woodhouse S, Stieb DM, Brook JR. 2003. Ambient nitrogen dioxide and distance from a major highway. Sci Total Environ 312(1-3):43-46.  219. Goldberg, Mark S., John C Bailar III, Richard T Burnett, Jeffrey R Brook, Robyn Tamblyn, Yvette Bonvalot, Pierre Ernst, Kenneth M Flegel, Ravinder K Singh, and Marie-France Valois. 2000. Identifying Subgroups of the General Population That May Be Susceptible to Short-Term Increases in Particulate Air Pollution: A Time-Series Study in Montreal,Quebec. Health effects Institute, research report No 97.  220. Krewski Daniel, Richard T Burnett, Mark S Goldberg, Kristin Hoover, Jack Siemiatycki, Michael Jerrett, Michal Abrahamowicz, Warren H White, et al. 2000. Reanalysis of the Harvard Six Cities and the American Cancer Society Study of Particulate Air Pollution and Mortality, Phase II: Sensitivity Analysis. Health Effects Institute  221. Jerrett M, Burnett R, Willis A, Krewski D, Goldberg MS, DeLuca p,  N. 2003. Spatial analysis of the air pollution mortality association in the context of ecologic confounders. J Toxicol Environ Health. 66:1735– 1777  222. Ma R, Krewski D, Burnett RT. 2003. Effects Cox models: a Poisson regression modelling approach. Biometrika. 90:157–169.  223. Evans GW, Kantrowitz E. 2002. Socioeconomic status and health: the potential role of environmental risk exposure. Annu Rev Public Health 23:303–31.  224. Woodward A, Boffetta P. 1997. Environmental exposure, social class, and cancer risk. IARC Sci Publ 138:361–7.  225. Chen Y, Tang M, Krewski D, et al. 2001. Relationship between asthma prevalence and income among Canadians. JAMA. 286(8):919–20.  90 226. Finkelstein MM, Jerrett M, Sears MR. 2005. Environmental inequality and circulatory disease mortality gradients. J Epidemiol Community Health. 59(6):481–7.  227. American Lung Association. 2001. Urban air pollution and health inequities: a workshop report. Environ Health Perspect. 109(Suppl 3):357–74.  228 O’Neill MS, Jerrett M, Kawachi I, et al. 2003. Health, wealth, and air pollution: advancing theory and methods. Environ Health Perspect. 111(16):1861–70.  229 Sexton K, Gong H Jr, Bailar JC III, et al. 1993. Air pollution health risks: do class and race matter? Toxicol Ind Health. 9(5):843–78.  230 Bowen W. 2002. An analytical review of environmental justice research: what do we really know? Environ Manage;29(1):3–15. 231 Dolinoy DC, Miranda ML. 2004. GIS modeling of air toxics releases from TRI-reporting and non-TRI- reporting facilities: impacts for environmental justice. Environ Health Perspect. 112(17):1717–24.  232 Stroh E, Oudin A, Gustafsson S, et al. 2005. Are associations between socio-economic characteristics and exposure to air pollution a question of study area size? An example from Scania, Sweden. Int J Health Geogr. 4:30.  233 Wheeler BW, Ben-Shlomo Y. 2005. Environmental equity, air quality, socioeconomic status, and respiratory health: a linkage analysis of routine data from the Health Survey for England. J Epidemiol Community Health. 59(11):948–54. 234 Brajer V, Hall JV. 2005. Changes in the distribution of air pollution exposure in the Los Angeles basin from 1990 to 1999. Contemporary Economic Policy. 23(1):50–8.  235 Filleul L, Harrabi I. 2004. Do socioeconomic conditions reflect a high exposure to air pollution or more sensitive health conditions? J Epidemiol Community Health. 58(9):802.  236 Jerrett M, Burnett RT, Brook J, et al. 2004. Do socioeconomic characteristics modify the short term association between air pollution and mortality? Evidence from a zonal time series in Hamilton, Canada. J Epidemiol Community Health. 58(1):31–40.  237. Laurent, O., Denis Bard, Laurent Filleul and Claire Segala. 2007. Effect of the socioeconomic status on the relationship between atmospheric pollution and mortality. J Epidemiol Community Health. 61:665-675.  238. Dragano, N., B. Hoffmann, S. Moebus, S. Mohlenkamp, A. Stang, P. E. Verde, K-H. Jockel, R. Erbel, and J. Siegrist. 2009. Traffic exposure and subclinical cardiovascular disease: is the association modified by socioeconomic characteristics of individuals and neighbourhoods? Results from a multilevel study in an urban region. Occup Environ Med. 66:628–635.  238b. Ren Cizao, Steve Melly, Joel Schwartz. 2010. Modifiers of short-term effects of ozone on mortality in eastern Massachusetts - A case-crossover analysis at individual level. Environmental Health 2010, 9:3 doi:10.1186/1476-069X-9-3.  239. Zanobetti Antonella, Meredith Franklin, Petros Koutrakis, and Joel Schwartz. 2009. Fine particulate air pollution and its components in association with cause-specific emergency admissions. Environmental Health 2009, 8:58 doi:10.1186/1476-069X-8-58.  240. Brauer M, Hoek G, van Vliet P, Meliefste K, Fischer P, Gehring U, et al. 2003. Estimating long-term average particulate air pollution concentrations: application of traffic indicators and geographic information systems. Epidemiology 14:228–239.  241. Hochadel M, Heinrich J, Gehring U, Morgenstern V, Kuhlbusch T, Link E, et al. 2006. Predicting long- term average concentrations of traffic-related air pollutants using GIS-based information. Atmos Environ 40:542.  91 242. Hoek G, Fischer P, Van Den Brandt P, Goldbohm S, Brunekreef B. 2001. Estimation of long-term average exposure to outdoor air pollution for a cohort study on mortality. J Expo Anal Environ Epidemiol 11:459–469.  243. Roorda-Knape MC, Janssen NA, de Hartog JJ, van Vliet PHN, Harssema H, Brunekreef B. 1998. Air pollution from traffic in city districts near major motorways. Atmos Environ 32:1921–1930.  244. Marshall JD, Nethery EN, Brauer M. 2008. Within-urban variability in ambient air pollution: comparison of estimation methods. Atmos Environ 42(6):1359–1369.  245. Leech JA, Nelson WC, Burnett RT, Aaron S, Raizenne ME. 2002. It’s about time: a comparison of Canadian and American time-activity patterns. J Expo Anal Environ Epidemiol 12(6):427–432.  246. Roosli M, Braun-Fahrlander C, Künzli N, Oglesby L, Theis G, Camenzind M, et al. 2000. Spatial variability of different fractions of particulate matter within an urban environment and between urban and rural sites. J Air Waste Manag Assoc 50(7):1115–1124.  247. Sarnat J, Koutrakis P, Suh H. 2000. Assessing the relationship between personal particulate and gaseous exposures of senior citizens living in Baltimore, MD. J Air Waste Manag Assoc 50(7):1184–1198.  248. Riediker M, Williams R, Devlin R, Griggs T, Bromberg P. 2003. Exposure to particulate matter, volatile organic compounds, and other air pollutants inside patrol cars. Environ Sci Technol 37(10):2084–2093.  249. Nethery EN, Sara Leckie, Kay Teschke, Michael Brauer. 2007. From measures to models: An evaluation of air pollution exposure assessment for epidemiologic studies of pregnant women. Occup Environ Med. 65:579‐586.  250. Lanki, T., Anne Ahokas, Sari Alm, Nicole A. H. Janssen, Gerard Hoek, Jeroen J. De Hartog, Bert Brunekreef, Juha Pekkanen. 2007. Determinants of personal and indoor PM2.5 and absorbance among elderly subjects with coronary heart disease. J Expo Sci Environ Epidemiol. 17(2): 124-133, doi: 0.1038/sj.jes.7500470  251. Canada Heart and stroke Foundation. 2010. A perfect storm of heart diseases looming on our horizon. 2010 Hear and Stroke Foundation Annual Report on Canadians' Health. January 25, 2010. 8p.  251a. 12. Kunzli N, Medina S, Kaiser R, et al. 2001. Assessment of deaths attributable to air pollution: should we use risk estimates based on time series or on cohort studies? Am J Epidemiol.153: 1050–5.  251b. McMichael AJ, Anderson HR, Brunekreef B, et al. 1998. Inappropriate use of daily mortality analyses to estimate longer-term mortality effects of air pollution. Int J Epidemiol. 27:450–3.  252. Babisch WF, Beule B, Schust M, Kersten N, Ising H. 2005. Traffic noise and risk of myocardial infarction. Epidemiology.16:33–40.  253. Davies HW, Teschke K, Kennedy SM, Hodgson MR, Hertzman C, Demers PA. 2005. Occupational exposure to noise and mortality from acute myocardial infarction. Epidemiology. 16:25–32.  254. Health Inequities in British Columbia – Discussion paper. November 2008. PHABC. 83p  255. Cooper RS, JS Kaufman. 1998. Race and hypertension: Science or nescience? Hypertension. 32: 813-816.  256. Tremblay MS, Perez CE. Aidem CI, Bryan SN, Katzmarzyk PT. 2005. Obesity, overweight and ethnicity: evidence from the Canadian community health survey. Health Reports. 16(4):23-34.  257. Alter, David A. , Alice Chong, Peter C. Austin, Cameron Mustard, Karey Iron, Jack I. Williams, Christopher D. Morgan, Jack V. Tu, Jane Irvine,and C. David Naylor. 2006. Socioeconomic Status and Mortality after Acute Myocardial Infarction. Ann Intern Med. 144:82-93.  92 258. Soobader MJ, LeClere FB. 1999. Aggregation and the measurement of income inequality: effects on morbidity. Soc Sci Med. 48:733–744.  259. Blakely TA, Lochner K, Kawachi I. 2002. Metropolitan area income inequality and self-rated health—a multi-level study. Soc Sci Med. 54:65–77.  260. Hou F, Myles J. 2004. Neighborhood Inequality, Relative Deprivation and Self-Perceived Health Status. Analytical studies branch research paper series no. 228. Ottawa, Canada.  261. Wilson WJ. 1987. The Truly Disadvantaged: The Inner City, the Underclass, and Public Policy. Chicago: University of Chicago Press.  262. Liu K, Cedres LD, Stamler J, et al.1982. Relationship of education to major risk factor and death from coronary heart disease, cardiovascular diseases and all causes. Circulation. 66:308–14.  263. Muller A. 2002. Education, income inequality, and mortality: a multiple regression analysis. BMJ. 324:23–5. 264. Macintyre K, Stewart S, Chalmers J. 2001. Relation between socioeconomic deprivation and death from a first myocardial infarction in Scotland: population based analysis. BMJ.322:1152–1153.  265. Leiland A, Finlayson A, Clark D, et al. 2004. Assessing the representativeness of health surveys Abstract. Eur J Public Health S. 14:45.  266. Oakes JM. 2004. The (mis)estimation of neighbourhood effects: casual inference for a practicable social epidemiology. Soc Sci Med. 58:1929–1952.  267. Diez Roux AV. 2004. Estimating neighbourhood health effects: the challenges of casual inference in a complex world. Soc Sci Med. 58:1953–1960.  268. Lavis J.N. 1997. The links between labour-market experiences and health: Towards a research framework. Toronto: Institute for Work & Health. Working Paper 63.  269. Bartley M. 1994. Unemployment and ill-health: Understanding the relationship. Journal of Epidemiology and Community Health. 48: 333-337.  270. Lavis, J.N., C.A. Mustard, J.I. Payne, and M.S.R. Farrant. 1998. Employment, working conditions and health: Towards a set of population-level indicators. Toronto: Institute for Work & Health. Working Paper 99.  271. Marmot, M.G., G. Davey Smith, S. Stansfeld, C. Patel, F. North, J. Head, I. White, E. Brunner, and A. Fenney. 1991. Health inequalities among British civil servants: The Whitehall II Study. Lancet 337: 1387-1393.  272. Chiu, Maria, Peter C. Austin, Douglas G. Manuel, Jack V. Tu. Comparison of cardiovascular risk profiles among ethnic groups using population health surveys between 1996 and 2007. CMAJ 2010. DOI:10.1503/cmaj.091676   93 Appendix – Traffic pollution, road proximity and ACS and CHF outcomes  Table 26. Traffic exposure and relative risk for CCS health outcomes Traffic exposure Health Outcome RR of health outcome for subjects in the 4th vs. 1st quartiles of traffic pollution exposure ACS 0.95 (0.87 – 1.05) NO CHF 1.08 (0.89 – 1.30) ACS 0.93 (0.85 – 1.01) NO2 CHF 1.04 (0.88 – 1.29) ACS 1.08 (0.98 – 1.18) Black Carbon CHF 1.36 (1.11 – 1.65) ACS 0.98 (0.89 – 1.07) PM2.5 CHF 1.00 (0.82 - 1.24)                    * Note: exposure is determined based on the first month of follow-up, January 1999   Table 27. Road proximity* and relative risk for CVD health outcomes Road proximity Health Outcome RR of health outcome for subjects in road proximity vs. subjects not in road proximity ACS 1.46 (1.18 – 1.80) Subjects living within 50 m from expressways and primary highways (R-I) CHF 1.39 (0.88 – 2.19) ACS 1.09 (0.94 – 1.26) Subjects living between 50 and 150 m from expressways and primary highways (R-II) CHF 1.19 (0.88 – 1.60) ACS 1.11 (1.00 – 1.23) Subjects living within 50 m from secondary highways and major roads (R-III) CHF 1.41 (1.16 – 1.72) ACS 1.05 (0.97 – 1.13) Subjects living between 50 and 150 m from secondary highways and major roads (R-IV) CHF 0.95 (0.80 – 1.13) ACS 1.15 (1.06 – 1.25) Subjects living within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (R-V) CHF 1.43 (1.21 – 1.69)    * Note: exposure is determined based on the first month of follow-up, January 1999            94 Table 28. Crude hazard ratios for traffic pollutants Crude HR and 95% CI Pollutant Health Outcome 1st 2nd 3rd 4th ACS 1.00 1.01 (0.93 - 1.11) 0.93 (0.85 - 1.02) 0.95 (0.87 - 1.04) NO CHF 1.00 0.79 (0.64 - 0.96) 0.88 (0.73 - 1.08) 1.07 (0.89 - 1.29) ACS 1.00 0.96 (0.88 - 1.05) 0.80 (0.72 - 0.88) 0.94 (0.86 - 1.03) NO2 CHF 1.00 0.86 (0.71 - 1.05) 0.92 (0.75 - 1.12) 1.07 (0.89 - 1.30) ACS 1.00 0.99 (0.90 - 1.08) 1.02 (0.93 - 1.13) 1.10 (1.00 - 1.20) PM2.5 CHF 1.00 1.34 (1.09 - 1.64) 1.28 (1.03 - 1.57) 1.48 (1.21 - 1.81) ACS 1.00 1.04 (0.95 - 1.14) 0.95 (0.87 - 1.05) 1.00 (0.91 - 1.10) Black Carbon CHF 1.00 1.12 (0.91 - 1.36) 1.07 (0.87 - 1.31) 1.06 (0.86 - 1.30)  Table 29. Hazard ratios for traffic pollutants adjusted for DA and neighborhood levels SES covariates DA SES Adjusted1 HR and 95% CI Neighborhood SES Adjusted2 HR and 95% CI Pollutant Health Outcome 2nd 3rd 4th 2nd 3rd 4th ACS 1.04 (0.95 - 1.14) 0.99 (0.89 - 1.09) 1.03 (0.93 - 1.14) 1.05 (0.95 - 1.15) 1.03 (0.93 - 1.14) 1.08 (0.97 - 1.20) NO CHF 0.73 (0.59 - 0.90) 0.81 (0.66 - 1.00) 0.98 (0.79 - 1.20) 0.80 (0.64 - 0.99) 0.91 (0.73 - 1.14) 1.11 (0.89 - 1.37) ACS 1.00 (0.92 - 1.10) 0.91 (0.82 - 1.01) 1.01 (0.91 - 1.13) 1.01 (0.92 - 1.11) 0.93 (0.83 - 1.05) 1.09 (0.98 - 1.22) NO2 CHF 0.83 (0.68 - 1.02) 0.86 (0.68 - 1.07) 0.91 (0.73 - 1.15) 0.92 (0.74 - 1.15) 0.97 (0.76 - 1.24) 1.10 (0.87 - 1.40) ACS 0.94 (0.86 - 1.04) 0.97 (0.88 - 1.07) 0.99 (0.90 - 1.09) 0.96 (0.87 - 1.06) 1.02 (0.92 - 1.13) 1.06 (0.96 - 1.17) PM2.5 CHF 1.25 (1.02 - 1.54) 1.11 (0.89 - 1.38) 1.20 (0.97 - 1.48) 1.21 (0.97 - 1.50) 1.11 (0.89 - 1.40) 1.18 (0.95 - 1.47) ACS 1.06 (0.96 - 1.16) 1.01 (0.91 - 1.11) 1.07 (0.96 - 1.18) 1.07 (0.97 - 1.18) 1.03 (0.93 - 1.14) 1.09 (0.98 - 1.21) Black Carbon CHF 1.06 (0.86 - 1.3) 1.03 (0.83 - 1.27) 0.95 (0.77 - 1.18) 1.12 (0.91 - 1.38) 1.09 (0.88 - 1.36) 0.99 (0.79 - 1.24) 1 The adjustment was done for sex, age class, and 10 DA level SES covariates; 2 The adjustment was done for sex, age class, and 10 Neighborhood level SES covariates; SES variables grouped in quintiles  Table 30. Crude and adjusted hazard ratios for road proximity adjustment done using SES covariates at different levels of aggregation Analyses HR and 95% CI Pollutant Health Outcome Crude HR DA SES Adjusted1 HR Neighborhood SES Adjusted2 HR ACS 1.46 (1.18 - 1.80) 1.24 (1.00 - 1.53) 1.25 (1.02 - 1.55) Within 50 m from expressways and primary highways (R-I) CHF 1.32 (0.82 - 2.10) 1.17 (0.73 - 1.87) 1.09 (0.68 - 1.75) ACS 1.12 (0.97 - 1.29) 1.08 (0.93 - 1.24) 1.10 (0.95 - 1.27) Between 50 and 150 m from expressways and primary highways (R-II) CHF 1.21 (0.90 - 1.63) 1.13 (0.83 - 1.52) 1.11 (0.82 - 1.50) ACS 1.16 (1.05 - 1.28) 1.06 (0.96 - 1.17) 1.07 (0.97 - 1.19) Within 50 m from secondary highways and major roads (R-III) CHF 1.61 (1.33 - 1.95) 1.31 (1.09 - 1.59) 1.34 (1.11 - 1.62) ACS 1.01 (0.93 - 1.09) 1.02 (0.94 - 1.11) 1.04 (0.96 - 1.13) Between 50 and 150 m from secondary highways and major roads (R-IV) CHF 0.96 (0.81 - 1.14) 0.89 (0.75 - 1.06) 0.91 (0.76 - 1.09) ACS 1.20 (1.10 - 1.30) 1.09 (1.01 - 1.19) 1.11 (1.02 - 1.21) Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (R-V) CHF 1.56 (1.33 - 1.83) 1.31 (1.11 - 1.55) 1.32 (1.12 - 1.55) 1 The adjustment was done for sex, age class, and 10 DA level SES covariates; 2 The adjustment was done for sex, age class, and 10 Neighborhood level SES covariates; SES variables grouped in quintiles  95 Appendix – Traffic pollution, road proximity and ACS and CHF outcomes in relation with low/high levels of  SES Table 31. Comparison between traffic pollution HR for low and high levels of DA- SES variables, when considering ACS health outcomes Health Outcome ACS Pollutant Quartile DA-level SES Pollutant SES Level 1st Q 2nd Q 3rd Q 4th Q Low 1.00 1.03 (0.88 - 1.22) 1.05 (0.88 - 1.25) 0.97 (0.79 - 1.19) NO High 1.00 1.10 (0.79 - 1.53) 0.89 (0.65 - 1.22) 1.08 (0.79 - 1.46) Low 1.00 1.03 (0.88 - 1.21) 1.00 (0.82 - 1.22) 0.87 (0.71 - 1.08) NO2 High 1.00 0.99 (0.67 - 1.44) 0.88 (0.62 - 1.25) 0.96 (0.68 - 1.37) Low 1.00 0.93 (0.78 - 1.12) 0.97 (0.80 - 1.17) 0.92 (0.76 - 1.10) Black Carbon High 1.00 1.09 (0.85 - 1.40) 0.98 (0.76 - 1.25) 1.07 (0.84 - 1.37) Low 1.00 1.08 (0.90 - 1.29) 1.05 (0.87 - 1.26) 1.10 (0.91 - 1.34) Chinese population PM2.5 High 1.00 0.96 (0.72 - 1.27) 0.66 (0.50 - 0.88) 0.84 (0.64 - 1.10) Low 1.00 0.99 (0.84 - 1.18) 1.02 (0.85 - 1.23) 0.94 (0.77 - 1.16) NO High 1.00 0.90 (0.70 - 1.16) 0.89 (0.70 - 1.13) 0.94 (0.75 - 1.18) Low 1.00 1.07 (0.91 - 1.26) 0.99 (0.80 - 1.21) 1.08 (0.89 - 1.30) NO2 High 1.00 0.82 (0.63 - 1.06) 0.83 (0.65 - 1.06) 0.86 (0.68 - 1.10) Low 1.00 1.01 (0.83 - 1.23) 0.95 (0.78 - 1.16) 1.06 (0.87 - 1.28) Black Carbon High 1.00 0.84 (0.66 - 1.08) 0.97 (0.77 - 1.21) 1.01 (0.81 - 1.27) Low 1.00 1.16 (0.94 - 1.42) 1.04 (0.85 - 1.28) 1.19 (0.96 - 1.46) University PM2.5 High 1.00 0.87 (0.67 - 1.13) 0.93 (0.74 - 1.18) 0.95 (0.76 - 1.18) Low 1.00 0.89 (0.73 - 1.10) 0.86 (0.70 - 1.05) 0.82 (0.67 - 1.00) NO High 1.00 0.94 (0.76 - 1.16) 0.82 (0.66 - 1.02) 0.78 (0.62 - 0.98) Low 1.00 0.87 (0.71 - 1.07) 0.68 (0.55 - 0.83) 0.80 (0.66 - 0.97) NO2 High 1.00 0.84 (0.68 - 1.03) 0.83 (0.66 - 1.04) 0.74 (0.59 - 0.92) Low 1.00 0.91 (0.74 - 1.13) 0.87 (0.71 - 1.07) 0.98 (0.81 - 1.19) Black Carbon High 1.00 0.86 (0.69 - 1.07) 0.91 (0.73 - 1.13) 0.85 (0.68 - 1.06) Low 1.00 0.96 (0.77 - 1.18) 0.86 (0.69 - 1.07) 0.99 (0.81 - 1.22) Employment PM2.5 High 1.00 1.11 (0.90 - 1.37) 0.91 (0.72 - 1.14) 0.79 (0.62 - 1.00) Low 1.00 0.90 (0.73 - 1.10) 0.75 (0.61 - 0.92) 0.74 (0.61 - 0.90) NO High 1.00 0.94 (0.76 - 1.16) 1.09 (0.89 - 1.33) 0.95 (0.76 - 1.20) Low 1.00 0.90 (0.73 - 1.10) 0.75 (0.61 - 0.92) 0.74 (0.61 - 0.90) NO2 High 1.00 0.89 (0.73 - 1.07) 0.70 (0.56 - 0.86) 0.75 (0.58 - 0.98) Low 1.00 0.73 (0.57 - 0.93) 0.82 (0.66 - 1.02) 0.93 (0.76 - 1.14) Black Carbon High 1.00 0.79 (0.65 - 0.98) 0.87 (0.69 - 1.08) 1.03 (0.84 - 1.26) Low 1.00 0.98 (0.77 - 1.25) 0.85 (0.67 - 1.08) 0.99 (0.78 - 1.24) Family income PM2.5 High 1.00 0.98 (0.80 - 1.20) 0.92 (0.74 - 1.13) 0.83 (0.66 - 1.06) Low 1.00 0.97 (0.77 - 1.22) 0.76 (0.61 - 0.95) 0.80 (0.64 - 0.99) NO High 1.00 0.90 (0.72 - 1.12) 1.06 (0.86 - 1.30) 0.96 (0.77 - 1.20) Low 1.00 0.94 (0.74 - 1.19) 0.69 (0.55 - 0.87) 0.80 (0.64 - 0.99) NO2 High 1.00 0.87 (0.71 - 1.05) 0.75 (0.61 - 0.93) 0.75 (0.59 - 0.96) Low 1.00 0.84 (0.66 - 1.07) 0.92 (0.73 - 1.15) 0.97 (0.78 - 1.20) Black Carbon High 1.00 0.81 (0.66 - 1.01) 0.91 (0.73 - 1.13) 1.00 (0.82 - 1.24) Low 1.00 1.01 (0.77 - 1.32) 0.81 (0.62 - 1.06) 1.04 (0.80 - 1.34) Personal income PM2.5 High 1.00 0.89 (0.72 - 1.11) 0.97 (0.79 - 1.20) 0.85 (0.67 - 1.07)  96 Table 31. Comparison between traffic pollution HR for low and high levels of DA- SES variables, when considering ACS health outcomes (cont.) Health Outcome ACS Pollutant Quartile DA-level SES Pollutant SES Level 1st Q 2nd Q 3rd Q 4th Q Low 1.00 1.00 (0.83 - 1.20) 1.12 (0.93 - 1.36) 0.92 (0.75 - 1.14) NO High 1.00 0.99 (0.71 - 1.38) 0.80 (0.58 - 1.11) 0.85 (0.62 - 1.16) Low 1.00 0.92 (0.78 - 1.09) 0.72 (0.56 - 0.91) 0.92 (0.73 - 1.17) NO2 High 1.00 0.96 (0.59 - 1.56) 0.89 (0.57 - 1.40) 1.02 (0.66 - 1.59) Low 1.00 0.95 (0.79 - 1.14) 1.01 (0.82 - 1.24) 0.98 (0.80 - 1.19) Black Carbon High 1.00 0.89 (0.66 - 1.20) 0.95 (0.72 - 1.26) 0.97 (0.74 - 1.28) Low 1.00 1.14 (0.95 - 1.38) 1.01 (0.82 - 1.24) 1.14 (0.92 - 1.40) Transportation PM2.5 High 1.00 1.05 (0.79 - 1.39) 0.96 (0.73 - 1.27) 1.04 (0.80 - 1.36) Low 1.00 1.00 (0.83 - 1.21) 1.18 (0.97 - 1.42) 1.21 (0.99 - 1.49) NO High 1.00 0.84 (0.66 - 1.06) 0.83 (0.67 - 1.04) 0.76 (0.61 - 0.93) Low 1.00 0.76 (0.61 - 0.93) 1.02 (0.86 - 1.21) 1.03 (0.83 - 1.27) NO2 High 1.00 1.02 (0.86 - 1.21) 1.03 (0.83 - 1.27) 1.00 (0.78 - 1.28) Low 1.00 1.00 (0.84 - 1.21) 1.01 (0.81 - 1.24) 1.10 (0.91 - 1.34) Black Carbon High 1.00 0.86 (0.67 - 1.10) 0.94 (0.75 - 1.17) 1.04 (0.84 - 1.30) Low 1.00 1.12 (0.94 - 1.33) 1.06 (0.87 - 1.29) 1.03 (0.82 - 1.30) Low income PM2.5 High 1.00 1.02 (0.80 - 1.31) 0.92 (0.72 - 1.18) 1.00 (0.79 - 1.26) Low 1.00 0.95 (0.74 - 1.22) 0.74 (0.58 - 0.95) 0.68 (0.54 - 0.86) NO High 1.00 0.89 (0.74 - 1.07) 1.09 (0.90 - 1.32) 1.02 (0.82 - 1.25) Low 1.00 0.94 (0.71 - 1.25) 0.65 (0.50 - 0.85) 0.76 (0.59 - 0.97) NO2 High 1.00 0.93 (0.79 - 1.10) 0.87 (0.70 - 1.07) 1.06 (0.83 - 1.34) Low 1.00 0.77 (0.59 - 1.01) 0.82 (0.65 - 1.04) 0.85 (0.67 - 1.06) Black Carbon High 1.00 1.17 (0.97 - 1.41) 1.16 (0.94 - 1.43) 1.31 (1.07 - 1.61) Low 1.00 0.99 (0.77 - 1.27) 0.91 (0.71 - 1.16) 0.99 (0.79 - 1.25) Home ownership PM2.5 High 1.00 0.95 (0.79 - 1.15) 1.10 (0.90 - 1.34) 1.01 (0.81 - 1.25) Low 1.00 1.06 (0.88 - 1.28) 0.82 (0.67 - 1.01) 0.82 (0.68 - 0.99) NO High 1.00 0.94 (0.75 - 1.18) 1.04 (0.83 - 1.30) 0.99 (0.79 - 1.24) Low 1.00 0.99 (0.81 - 1.20) 0.84 (0.68 - 1.04) 0.87 (0.73 - 1.04) NO2 High 1.00 0.80 (0.64 - 0.99) 0.70 (0.56 - 0.86) 0.78 (0.62 - 1.00) Low 1.00 1.07 (0.87 - 1.33) 1.08 (0.88 - 1.32) 1.06 (0.87 - 1.30) Black Carbon High 1.00 0.76 (0.61 - 0.94) 0.85 (0.68 - 1.06) 0.86 (0.69 - 1.06) Low 1.00 1.14 (0.92 - 1.40) 1.07 (0.87 - 1.34) 0.96 (0.78 - 1.18) Income variation PM2.5 High 1.00 0.96 (0.77 - 1.19) 0.95 (0.76 - 1.19) 1.01 (0.81 - 1.27) Low 1.00 0.87 (0.72 - 1.05) 0.68 (0.55 - 0.84) 0.79 (0.64 - 0.96) NO High 1.00 1.07 (0.87 - 1.32) 0.95 (0.77 - 1.17) 0.94 (0.76 - 1.16) Low 1.00 0.90 (0.74 - 1.10) 0.67 (0.54 - 0.82) 0.80 (0.66 - 0.96) NO2 High 1.00 0.87 (0.72 - 1.05) 0.77 (0.62 - 0.95) 0.75 (0.60 - 0.94) Low 1.00 0.86 (0.70 - 1.06) 0.79 (0.64 - 0.97) 0.90 (0.73 - 1.10) Black Carbon High 1.00 0.94 (0.77 - 1.15) 0.96 (0.78 - 1.19) 0.87 (0.71 - 1.07) Low 1.00 1.09 (0.88 - 1.36) 0.99 (0.79 - 1.24) 1.04 (0.83 - 1.30) Management PM2.5 High 1.00 1.10 (0.90 - 1.34) 0.89 (0.72 - 1.11) 0.96 (0.78 - 1.19)     97 Table 32. Comparison between traffic pollution HR for low and high levels of Neighborhood-SES variables, when considering ACS health outcomes Health Outcome ACS Pollutant Quartile Neighborhood- level SES Pollutant SES Level 1st Q 2nd Q 3rd Q 4th Q Low 1.00 1.03 (0.85 - 1.23) 1.19 (1.00 - 1.42) 0.99 (0.81 - 1.21) NO High 1.00 0.84 (0.62 - 1.13) 0.63 (0.47 - 0.85) 0.76 (0.57 - 1.01) Low 1.00 1.19 (1.01 - 1.39) 1.03 (0.83 - 1.28) 0.94 (0.76 - 1.17) NO2 High 1.00 0.83 (0.59 - 1.15) 0.59 (0.44 - 0.80) 0.78 (0.58 - 1.04) Low 1.00 1.01 (0.84 - 1.21) 1.10 (0.90 - 1.33) 0.96 (0.80 - 1.16) Black Carbon High 1.00 0.82 (0.65 - 1.03) 0.72 (0.57 - 0.90) 0.90 (0.74 - 1.11) Low 1.00 1.05 (0.87 - 1.27) 1.19 (0.98 - 1.45) 1.19 (0.98 - 1.43) Other language PM2.5 High 1.00 0.97 (0.74 - 1.27) 0.63 (0.48 - 0.82) 0.88 (0.68 - 1.12) Low 1.00 1.02 (0.85 - 1.24) 1.03 (0.86 - 1.24) 0.96 (0.78 - 1.17) NO High 1.00 0.86 (0.62 - 1.20) 0.64 (0.47 - 0.88) 0.79 (0.58 - 1.07) Low 1.00 1.14 (0.96 - 1.34) 0.82 (0.66 - 1.03) 0.81 (0.65 - 1.00) NO2 High 1.00 0.82 (0.58 - 1.15) 0.60 (0.44 - 0.81) 0.75 (0.55 - 1.01) Low 1.00 0.94 (0.77 - 1.14) 1.06 (0.87 - 1.29) 0.95 (0.79 - 1.15) Black Carbon High 1.00 0.82 (0.64 - 1.04) 0.73 (0.57 - 0.93) 0.87 (0.69 - 1.09) Low 1.00 1.03 (0.85 - 1.24) 1.15 (0.94 - 1.41) 1.10 (0.91 - 1.33) Linguistic isolation PM2.5 High 1.00 0.97 (0.75 - 1.27) 0.69 (0.53 - 0.90) 0.81 (0.63 - 1.05) Low 1.00 0.90 (0.77 - 1.06) 1.08 (0.90 - 1.29) 0.81 (0.64 - 1.03) NO High 1.00 0.83 (0.65 - 1.06) 0.74 (0.59 - 0.93) 0.84 (0.67 - 1.04) Low 1.00 1.03 (0.88 - 1.20) 1.00 (0.78 - 1.27) 0.95 (0.77 - 1.17) NO2 High 1.00 0.77 (0.59 - 0.99) 0.70 (0.55 - 0.88) 0.75 (0.60 - 0.95) Low 1.00 1.03 (0.86 - 1.23) 1.03 (0.85 - 1.25) 0.95 (0.79 - 1.15) Black Carbon High 1.00 0.76 (0.59 - 0.98) 0.90 (0.72 - 1.13) 0.99 (0.80 - 1.23) Low 1.00 1.12 (0.92 - 1.37) 1.02 (0.83 - 1.24) 1.20 (0.98 - 1.47) University PM2.5 High 1.00 0.87 (0.67 - 1.12) 0.92 (0.73 - 1.14) 0.90 (0.73 - 1.12) Low 1.00 0.95 (0.78 - 1.15) 1.15 (0.96 - 1.39) 1.15 (0.94 - 1.41) NO High 1.00 0.96 (0.77 - 1.18) 0.74 (0.59 - 0.92) 0.80 (0.64 - 0.99) Low 1.00 1.10 (0.93 - 1.29) 0.94 (0.76 - 1.16) 0.90 (0.71 - 1.14) NO2 High 1.00 1.04 (0.83 - 1.31) 0.65 (0.52 - 0.82) 0.93 (0.76 - 1.14) Low 1.00 0.90 (0.74 - 1.09) 1.07 (0.88 - 1.30) 0.99 (0.82 - 1.20) Black Carbon High 1.00 0.92 (0.73 - 1.16) 0.87 (0.70 - 1.10) 0.94 (0.75 - 1.17) Low 1.00 1.06 (0.89 - 1.26) 0.98 (0.81 - 1.19) 1.05 (0.84 - 1.31) Unemployment PM2.5 High 1.00 1.00 (0.86 - 1.39) 0.89 (0.70 - 1.14) 1.02 (0.80 - 1.31) Low 1.00 0.90 (0.71 - 1.14) 0.64 (0.51 - 0.80) 0.72 (0.57 - 0.89) NO High 1.00 0.94 (0.77 - 1.15) 1.07 (0.88 - 1.30) 1.20 (0.97 - 1.49) Low 1.00 1.09 (0.85 - 1.41) 0.63 (0.50 - 0.80) 0.88 (0.71 - 1.10) NO2 High 1.00 0.91 (0.75 - 1.09) 0.88 (0.72 - 1.08) 1.00 (0.78 - 1.27) Low 1.00 0.87 (0.67 - 1.12) 0.85 (0.67 - 1.08) 0.88 (0.69 - 1.12) Black Carbon High 1.00 0.83 (0.68 - 1.02) 1.00 (0.81 - 1.24) 1.02 (0.83 - 1.24) Low 1.00 1.02 (0.77 - 1.34) 0.80 (0.61 - 1.06) 0.93 (0.71 - 1.22) Family income PM2.5 High 1.00 0.96 (0.78 - 1.17) 1.01 (0.83 - 1.24) 0.95 (0.75 - 1.20)      98 Table 32. Comparison between traffic pollution HR for low and high levels of Neighborhood-SES variables, when considering ACS health outcomes (cont.) Health Outcome ACS Pollutant Quartile Neighborhood- level SES Pollutant SES Level 1st Q 2nd Q 3rd Q 4th Q Low 1.00 0.87 (0.71 - 1.08) 0.60 (0.48 - 0.74) 0.68 (0.55 - 0.84) NO High 1.00 0.93 (0.74 - 1.16) 0.93 (0.75 - 1.16) 0.93 (0.75 - 1.16) Low 1.00 0.97 (0.78 - 1.21) 0.55 (0.44 - 0.68) 0.78 (0.64 - 0.95) NO2 High 1.00 0.82 (0.66 - 1.01) 0.76 (0.61 - 0.94) 0.81 (0.65 - 1.01) Low 1.00 0.86 (0.68 - 1.08) 0.75 (0.59 - 0.95) 0.82 (0.65 - 1.03) Black Carbon High 1.00 0.75 (0.59 - 0.95) 0.92 (0.75 - 1.14) 0.96 (0.78 - 1.16) Low 1.00 1.13 (0.87 - 1.46) 0.81 (0.63 - 1.05) 0.87 (0.67 - 1.14) Personal income PM2.5 High 1.00 0.91 (0.72 - 1.14) 0.95 (0.77 - 1.18) 0.89 (0.72 - 1.11) Low 1.00 0.92 (0.76 - 1.13) 0.84 (0.68 - 1.04) 0.83 (0.66 - 1.04) NO High 1.00 0.89 (0.72 - 1.09) 0.75 (0.62 - 0.92) 0.71 (0.57 - 0.87) Low 1.00 0.76 (0.63 - 0.92) 0.75 (0.60 - 0.93) 0.72 (0.57 - 0.92) NO2 High 1.00 1.11 (0.90 - 1.37) 0.67 (0.54 - 0.83) 0.88 (0.72 - 1.07) Low 1.00 0.85 (0.68 - 1.05) 0.98 (0.79 - 1.21) 0.94 (0.76 - 1.16) Black Carbon High 1.00 0.94 (0.73 - 1.22) 0.99 (0.78 - 1.27) 0.88 (0.69 - 1.13) Low 1.00 1.10 (0.89 - 1.36) 1.02 (0.82 - 1.27) 0.80 (0.63 - 1.01) Governmental transfers PM2.5 High 1.00 1.05 (0.83 - 1.33) 0.79 (0.63 - 1.00) 0.97 (0.76 - 1.22) Low 1.00 1.05 (0.87 - 1.26) 1.30 (1.08 - 1.56) 1.13 (0.91 - 1.41) NO High 1.00 1.06 (0.77 - 1.47) 0.86 (0.63 - 1.16) 0.86 (0.64 - 1.16) Low 1.00 1.12 (0.95 - 1.31) 1.04 (0.81 - 1.33) 1.08 (0.83 - 1.41) NO2 High 1.00 1.33 (0.84 - 2.09) 0.97 (0.63 - 1.49) 1.20 (0.79 - 1.81) Low 1.00 1.02 (0.85 - 1.23) 1.12 (0.91 - 1.38) 1.04 (0.86 - 1.26) Black Carbon High 1.00 0.81 (0.62 - 1.07) 0.90 (0.70 - 1.16) 1.07 (0.84 - 1.37) Low 1.00 1.10 (0.93 - 1.31) 1.13 (0.93 - 1.38) 1.14 (0.91 - 1.44) Low income PM2.5 High 1.00 1.11 (0.84 - 1.47) 0.87 (0.66 - 1.15) 1.13 (0.87 - 1.47) Low 1.00 1.25 (0.96 - 1.61) 0.99 (0.78 - 1.27) 0.91 (0.72 - 1.15) NO High 1.00 0.93 (0.78 - 1.11) 1.05 (0.86 - 1.29) 1.02 (0.80 - 1.30) Low 1.00 1.22 (0.89 - 1.66) 0.90 (0.67 - 1.20) 1.04 (0.79 - 1.35) NO2 High 1.00 0.92 (0.78 - 1.09) 1.00 (0.77 - 1.31) 0.93 (0.70 - 1.25) Low 1.00 1.08 (0.78 - 1.48) 1.16 (0.87 - 1.55) 1.27 (0.95 - 1.69) Black Carbon High 1.00 0.95 (0.79 - 1.14) 1.17 (0.95 - 1.44) 0.95 (0.77 - 1.17) Low 1.00 1.00 (0.77 - 1.28) 0.99 (0.77 - 1.28) 1.02 (0.81 - 1.29) Home ownership PM2.5 High 1.00 1.04 (0.87 - 1.24) 1.17 (0.96 - 1.43) 1.24 (0.97 - 1.57) Low 1.00 0.98 (0.81 - 1.19) 1.18 (0.97 - 1.43) 0.95 (0.77 - 1.18) NO High 1.00 1.06 (0.83 - 1.34) 0.86 (0.67 - 1.09) 0.77 (0.61 - 0.97) Low 1.00 1.02 (0.86 - 1.20) 0.82 (0.66 - 1.02) 0.93 (0.71 - 1.21) NO2 High 1.00 1.21 (0.92 - 1.61) 0.82 (0.62 - 1.09) 1.02 (0.79 - 1.30) Low 1.00 1.01 (0.84 - 1.22) 1.13 (0.91 - 1.39) 0.99 (0.81 - 1.22) Black Carbon High 1.00 0.86 (0.66 - 1.12) 0.93 (0.73 - 1.19) 1.03 (0.81 - 1.31) Low 1.00 1.08 (0.90 - 1.31) 1.16 (0.95 - 1.42) 0.92 (0.73 - 1.16) Neighborhood stress PM2.5 High 1.00 1.12 (0.86 - 1.46) 1.00 (0.76 - 1.31) 1.19 (0.93 - 1.54)      99 Table 33. Comparison between traffic pollution HR for low and high levels of DA- SES variables, when considering CHF health outcomes Health Outcome CHF Pollutant Quartile DA-level SES Pollutant SES Level 1st Q 2nd Q 3rd Q 4th Q Low 1.00 0.82 (0.58 - 1.16) 0.62 (0.41 - 0.94) 0.70 (0.45 - 1.10) NO High 1.00 0.63 (0.31 - 1.27) 0.67 (0.35 - 1.27) 0.94 (0.52 - 1.73) Low 1.00 0.85 (0.60 - 1.19) 0.70 (0.44 - 1.11) 0.88 (0.57 - 1.35) NO2 High 1.00 0.98 (0.41 - 2.33) 0.93 (0.42 - 2.05) 1.05 (0.47 - 2.32) Low 1.00 1.75 (1.17 - 2.61) 1.37 (0.88 - 2.12) 0.90 (0.57 - 1.42) Black Carbon High 1.00 0.96 (0.53 - 1.73) 1.19 (0.69 - 2.07) 1.74 (1.04 - 2.91) Low 1.00 1.34 (0.90 - 1.99) 1.16 (0.76 - 1.77) 1.10 (0.70 - 1.72) Chinese population PM2.5 High 1.00 0.88 (0.47 - 1.64) 0.78 (0.43 - 1.42) 0.92 (0.52 - 1.63) Low 1.00 0.99 (0.69 - 1.42) 0.85 (0.57 - 1.26) 0.86 (0.56 - 1.32) NO High 1.00 0.64 (0.38 - 1.11) 0.70 (0.43 - 1.15) 0.97 (0.62 - 1.53) Low 1.00 0.70 (0.48 - 1.01) 0.67 (0.43 - 1.06) 0.98 (0.68 - 1.43) NO2 High 1.00 0.64 (0.37 - 1.11) 0.65 (0.39 - 1.08) 0.94 (0.59 - 1.52) Low 1.00 1.83 (1.14 - 2.94) 1.53 (0.95 - 2.48) 1.43 (0.88 - 2.31) Black Carbon High 1.00 0.75 (0.45 - 1.23) 0.76 (0.48 - 1.21) 0.88 (0.57 - 1.37) Low 1.00 0.99 (0.63 - 1.55) 1.13 (0.74 - 1.74) 0.99 (0.64 - 1.56) University PM2.5 High 1.00 1.29 (0.78 - 2.11) 1.05 (0.64 - 1.72) 0.95 (0.59 - 1.53) Low 1.00 0.72 (0.48 - 1.08) 0.81 (0.55 - 1.2) 0.82 (0.56 - 1.21) NO High 1.00 0.87 (0.54 - 1.38) 0.85 (0.54 - 1.36) 0.98 (0.63 - 1.53) Low 1.00 0.70 (0.46 - 1.06) 0.65 (0.44 - 0.95) 0.73 (0.51 - 1.04) NO2 High 1.00 0.90 (0.58 - 1.4) 0.79 (0.48 - 1.3) 0.94 (0.6 - 1.46) Low 1.00 1.21 (0.78 - 1.87) 1.04 (0.68 - 1.59) 1.33 (0.89 - 1.99) Black Carbon High 1.00 1.29 (0.81 - 2.06) 1.31 (0.81 - 2.11) 1.05 (0.64 - 1.72) Low 1.00 0.89 (0.6 - 1.32) 0.80 (0.53 - 1.2) 0.89 (0.61 - 1.31) Employment PM2.5 High 1.00 1.04 (0.65 - 1.67) 1.14 (0.72 - 1.81) 0.80 (0.48 - 1.32) Low 1.00 0.56 (0.35 - 0.89) 0.70 (0.47 - 1.05) 0.76 (0.52 - 1.12) NO High 1.00 0.84 (0.52 - 1.38) 1.03 (0.65 - 1.64) 1.37 (0.87 - 2.18) Low 1.00 0.51 (0.32 - 0.83) 0.49 (0.32 - 0.74) 0.68 (0.47 - 0.97) NO2 High 1.00 0.92 (0.59 - 1.45) 1.00 (0.64 - 1.59) 1.17 (0.70 - 1.97) Low 1.00 1.75 (1.00 - 3.07) 1.35 (0.78 - 2.33) 1.61 (0.95 - 2.73) Black Carbon High 1.00 0.89 (0.58 - 1.37) 0.92 (0.58 - 1.46) 0.72 (0.44 - 1.16) Low 1.00 1.06 (0.65 - 1.72) 0.92 (0.57 - 1.50) 0.95 (0.59 - 1.53) Family income PM2.5 High 1.00 1.16 (0.74 - 1.83) 1.26 (0.80 - 1.99) 1.08 (0.65 - 1.79) Low 1.00 0.48 (0.29 - 0.78) 0.65 (0.43 - 0.98) 0.63 (0.42 - 0.95) NO High 1.00 0.70 (0.42 - 1.14) 0.72 (0.45 - 1.16) 1.03 (0.66 - 1.62) Low 1.00 0.44 (0.27 - 0.73) 0.46 (0.30 - 0.71) 0.62 (0.42 - 0.92) NO2 High 1.00 0.65 (0.40 - 1.05) 0.83 (0.52 - 1.31) 0.97 (0.59 - 1.58) Low 1.00 1.66 (0.93 - 2.97) 1.65 (0.94 - 2.89) 1.76 (1.02 - 3.04) Black Carbon High 1.00 0.82 (0.51 - 1.29) 0.79 (0.49 - 1.29) 0.83 (0.52 - 1.31) Low 1.00 1.06 (0.61 - 1.84) 0.78 (0.45 - 1.35) 1.02 (0.61 - 1.73) Personal income PM2.5 High 1.00 1.19 (0.74 - 1.93) 1.15 (0.70 - 1.90) 1.25 (0.76 - 2.04)      100 Table 33. Comparison between traffic pollution HR for low and high levels of DA- SES variables, when considering CHF health outcomes (cont.) Health Outcome CHF Pollutant Quartile DA-level SES Pollutant SES Level 1st Q 2nd Q 3rd Q 4th Q Low 1.00 0.85 (0.57 - 1.27) 0.82 (0.53 - 1.27) 0.65 (0.40 - 1.06) NO High 1.00 0.92 (0.43 - 1.96) 1.13 (0.56 - 2.29) 1.21 (0.61 - 2.42) Low 1.00 0.73 (0.49 - 1.09) 0.64 (0.38 - 1.10) 1.18 (0.74 - 1.90) NO2 High 1.00 1.07 (0.40 - 2.84) 0.97 (0.39 - 2.44) 1.01 (0.41 - 2.48) Low 1.00 1.58 (1.05 - 2.40) 1.45 (0.92 - 2.30) 0.96 (0.59 - 1.58) Black Carbon High 1.00 0.61 (0.33 - 1.12) 0.83 (0.49 - 1.41) 0.98 (0.58 - 1.66) Low 1.00 1.35 (0.87 - 2.10) 1.44 (0.92 - 2.27) 1.08 (0.65 - 1.79) Transportation PM2.5 High 1.00 1.23 (0.72 - 2.11) 0.94 (0.54 - 1.63) 0.94 (0.55 - 1.58) Low 1.00 0.73 (0.47 - 1.12) 0.61 (0.38 - 0.99) 1.11 (0.72 - 1.71) NO High 1.00 0.63 (0.37 - 1.07) 0.86 (0.54 - 1.37) 0.81 (0.52 - 1.28) Low 1.00 0.65 (0.43 - 0.97) 0.89 (0.57 - 1.41) 0.72 (0.40 - 1.27) NO2 High 1.00 0.70 (0.40 - 1.23) 0.59 (0.35 - 0.99) 0.71 (0.44 - 1.14) Low 1.00 1.39 (0.92 - 2.08) 1.10 (0.68 - 1.79) 0.90 (0.55 - 1.47) Black Carbon High 1.00 2.58 (1.28 - 5.18) 2.27 (1.16 - 4.45) 2.69 (1.38 - 5.23) Low 1.00 1.24 (0.82 - 1.89) 1.24 (0.78 - 1.97) 1.04 (0.59 - 1.82) Low income PM2.5 High 1.00 1.39 (0.78 - 2.49) 1.18 (0.66 - 2.12) 1.38 (0.80 - 2.38) Low 1.00 0.79 (0.47 - 1.33) 0.80 (0.49 - 1.30) 0.77 (0.49 - 1.23) NO High 1.00 0.78 (0.5 - 1.22) 0.81 (0.50 - 1.30) 0.79 (0.47 - 1.30) Low 1.00 0.79 (0.46 - 1.35) 0.55 (0.34 - 0.91) 0.58 (0.37 - 0.92) NO2 High 1.00 0.87 (0.57 - 1.33) 0.65 (0.37 - 1.13) 1.37 (0.83 - 2.25) Low 1.00 1.46 (0.83 - 2.57) 1.06 (0.61 - 1.82) 1.35 (0.80 - 2.27) Black Carbon High 1.00 1.51 (0.96 - 2.38) 1.28 (0.77 - 2.15) 1.26 (0.76 - 2.10) Low 1.00 1.36 (0.80 - 2.29) 1.15 (0.68 - 1.94) 1.10 (0.67 - 1.81) Home ownership PM2.5 High 1.00 1.15 (0.74 - 1.77) 1.14 (0.70 - 1.84) 0.57 (0.29 - 1.10) Low 1.00 0.77 (0.50 - 1.17) 0.78 (0.50 - 1.20) 0.88 (0.59 - 1.30) NO High 1.00 0.74 (0.45 - 1.22) 0.90 (0.56 - 1.44) 1.28 (0.82 - 1.99) Low 1.00 0.77 (0.05 - 1.18) 0.52 (0.32 - 0.87) 0.90 (0.62 - 1.30) NO2 High 1.00 0.91 (0.56 - 1.47) 0.96 (0.61 - 1.52) 0.99 (0.60 - 1.65) Low 1.00 1.67 (1.03 - 2.73) 1.43 (0.88 - 2.3) 1.29 (0.80 - 2.06) Black Carbon High 1.00 0.78 (0.50 - 1.2) 0.71 (0.45 - 1.14) 0.81 (0.52 - 1.27) Low 1.00 1.28 (0.79 - 2.06) 1.36 (0.84 - 2.2) 1.07 (0.67 - 1.70) Income variation PM2.5 High 1.00 0.82 (0.52 - 1.29) 1.08 (0.70 - 1.67) 0.82 (0.51 - 1.31) Low 1.00 0.74 (0.49 - 1.13) 0.70 (0.45 - 1.08) 0.93 (0.62 - 1.39) NO High 1.00 0.89 (0.55 - 1.45) 0.99 (0.63 - 1.57) 1.08 (0.68 - 1.71) Low 1.00 0.65 (0.42 - 1.01) 0.66 (0.43 - 1.01) 0.76 (0.52 - 1.13) NO2 High 1.00 0.80 (0.49 - 1.29) 1.13 (0.72 - 1.79) 1.16 (0.72 - 1.87) Low 1.00 1.14 (0.71 - 1.85) 1.07 (0.67 - 1.72) 1.36 (0.86 - 2.15) Black Carbon High 1.00 1.01 (0.65 - 1.57) 0.90 (0.56 - 1.45) 0.92 (0.58 - 1.45) Low 1.00 0.76 (0.48 - 1.19) 0.88 (0.57 - 1.35) 0.74 (0.47 - 1.16) Management PM2.5 High 1.00 1.36 (0.87 - 2.10) 1.04 (0.64 - 1.69) 1.11 (0.70 - 1.77)      101 Table 34. Comparison between traffic pollution HR for low and high levels of Neighborhood-SES variables, when considering CHF health outcomes Health Outcome CHF Pollutant Quartile Neighborhood- level SES Pollutant SES Level 1st Q 2nd Q 3rd Q 4th Q Low 1.00 0.73 (0.48 - 1.10) 0.58 (0.37 - 0.89) 0.92 (0.61 - 1.38) NO High 1.00 0.55 (0.26 - 1.14) 0.88 (0.46 - 1.70) 0.93 (0.49 - 1.77) Low 1.00 0.83 (0.58 - 1.19) 0.81 (0.50 - 1.31) 0.73 (0.45 - 1.16) NO2 High 1.00 0.77 (0.35 - 1.67) 0.79 (0.40 - 1.55) 0.75 (0.38 - 1.46) Low 1.00 1.38 (0.89 - 2.14) 1.48 (0.94 - 2.32) 1.19 (0.77 - 1.84) Black Carbon High 1.00 1.09 (0.65 - 1.83) 0.94 (0.57 - 1.58) 1.39 (0.89 - 2.19) Low 1.00 1.25 (0.80 - 1.95) 1.77 (1.13 - 2.76) 1.51 (0.97 - 2.33) Other language PM2.5 High 1.00 1.10 (0.62 - 1.94) 0.86 (0.49 - 1.52) 0.97 (0.57 - 1.66) Low 1.00 0.54 (0.34 - 0.86) 0.58 (0.37 - 0.91) 0.99 (0.66 - 1.47) NO High 1.00 0.69 (0.30 - 1.56) 0.98 (0.47 - 2.06) 1.11 (0.53 - 2.3) Low 1.00 0.62 (0.41 - 0.94) 0.80 (0.51 - 1.26) 0.70 (0.45 - 1.10) NO2 High 1.00 0.79 (0.35 - 1.76) 0.80 (0.40 - 1.62) 0.76 (0.38 - 1.54) Low 1.00 1.11 (0.71 - 1.75) 1.08 (0.68 - 1.71) 1.11 (0.73 - 1.70) Black Carbon High 1.00 1.51 (0.83 - 2.76) 1.27 (0.69 - 2.35) 2.04 (1.17 - 3.56) Low 1.00 1.40 (0.89 - 2.21) 1.82 (1.13 - 2.91) 1.80 (1.14 - 2.83) Linguistic isolation PM2.5 High 1.00 1.37 (0.74 - 2.51) 1.10 (0.60 - 1.99) 1.14 (0.64 - 2.05) Low 1.00 0.90 (0.65 - 1.26) 0.76 (0.51 - 1.14) 0.54 (0.30 - 0.94) NO High 1.00 0.67 (0.40 - 1.12) 0.72 (0.46 - 1.14) 1.03 (0.68 - 1.57) Low 1.00 0.86 (0.62 - 1.19) 0.58 (0.31 - 1.08) 0.70 (0.43 - 1.13) NO2 High 1.00 0.56 (0.32 - 0.99) 0.70 (0.43 - 1.13) 0.93 (0.59 - 1.47) Low 1.00 2.10 (1.36 - 3.22) 1.61 (1.01 - 2.57) 1.19 (0.74 - 1.91) Black Carbon High 1.00 0.92 (0.57 - 1.49) 0.93 (0.60 - 1.45) 0.94 (0.61 - 1.44) Low 1.00 1.18 (0.77 - 1.81) 1.07 (0.69 - 1.64) 0.95 (0.60 - 1.49) University PM2.5 High 1.00 1.09 (0.68 - 1.74) 1.04 (0.68 - 1.59) 0.92 (0.60 - 1.41) Low 1.00 0.86 (0.56 - 1.32) 0.89 (0.57 - 1.38) 1.09 (0.70 - 1.69) NO High 1.00 0.50 (0.30 - 0.82) 0.74 (0.47 - 1.14) 0.81 (0.53 - 1.24) Low 1.00 0.70 (0.47 - 1.05) 0.91 (0.58 - 1.43) 0.90 (0.57 - 1.44) NO2 High 1.00 0.56 (0.34 - 0.93) 0.59 (0.38 - 0.91) 0.51 (0.34 - 0.78) Low 1.00 1.21 (0.78 - 1.87) 1.05 (0.66 - 1.66) 1.04 (0.67 - 1.60) Black Carbon High 1.00 1.80 (1.00 - 3.21) 1.13 (0.62 - 2.06) 1.74 (0.98 - 3.09) Low 1.00 0.85 (0.56 - 1.29) 1.11 (0.73 - 1.66) 0.70 (0.41 - 1.22) Unemployment PM2.5 High 1.00 1.67 (0.95 - 2.94) 1.27 (0.72 - 2.25) 1.23 (0.68 - 2.22) Low 1.00 0.48 (0.27 - 0.84) 0.69 (0.44 - 1.07) 0.82 (0.53 - 1.26) NO High 1.00 0.72 (0.46 - 1.12) 0.65 (0.41 - 1.02) 1.22 (0.80 - 1.86) Low 1.00 0.55 (0.31 - 0.97) 0.64 (0.41 - 1.00) 0.67 (0.44 - 1.01) NO2 High 1.00 0.73 (0.48 - 1.11) 0.83 (0.54 - 1.28) 1.22 (0.76 - 1.96) Low 1.00 1.54 (0.85 - 2.76) 1.09 (0.60 - 1.96) 1.53 (0.87 - 2.70) Black Carbon High 1.00 0.99 (0.66 - 1.48) 0.75 (0.46 - 1.23) 0.74 (0.48 - 1.16) Low 1.00 1.35 (0.70 - 2.60) 1.27 (0.67 - 2.38) 1.32 (0.71 - 2.46) Family income PM2.5 High 1.00 1.09 (0.70 - 1.68) 1.10 (0.71 - 1.69) 0.84 (0.49 - 1.42)      102 Table 34. Comparison between traffic pollution HR for low and high levels of Neighborhood-SES variables, when considering CHF health outcomes (cont.) Health Outcome CHF Pollutant Quartile Neighborhood- level SES Pollutant SES Level 1st Q 2nd Q 3rd Q 4th Q Low 1.00 0.57 (0.35 - 0.92) 0.69 (0.46 - 1.06) 0.84 (0.55 - 1.26) NO High 1.00 0.70 (0.43 - 1.15) 0.87 (0.55 - 1.37) 1.14 (0.73 - 1.77) Low 1.00 0.67 (0.42 - 1.08) 0.66 (0.44 - 0.99) 0.62 (0.42 - 0.93) NO2 High 1.00 0.82 (0.51 - 1.34) 0.86 (0.55 - 1.37) 1.16 (0.73 - 1.83) Low 1.00 1.62 (0.92 - 2.85) 1.07 (0.60 - 1.93) 1.57 (0.91 - 2.71) Black Carbon High 1.00 0.98 (0.63 - 1.52) 0.67 (0.41 - 1.08) 0.94 (0.63 - 1.41) Low 1.00 1.91 (1.04 - 3.52) 1.35 (0.73 - 2.48) 1.47 (0.80 - 2.73) Personal income PM2.5 High 1.00 1.33 (0.83 - 2.14) 1.16 (0.73 - 1.84) 1.13 (0.72 - 1.77) Low 1.00 0.69 (0.43 - 1.13) 0.62 (0.37 - 1.03) 0.93 (0.58 - 1.49) NO High 1.00 0.61 (0.39 - 0.97) 0.74 (0.49 - 1.11) 0.95 (0.63 - 1.42) Low 1.00 0.71 (0.44 - 1.15) 0.61 (0.36 - 1.05) 1.29 (0.81 - 2.05) NO2 High 1.00 0.70 (0.45 - 1.10) 0.69 (0.46 - 1.04) 0.74 (0.50 - 1.09) Low 1.00 0.88 (0.54 - 1.44) 0.82 (0.49 - 1.37) 0.91 (0.57 - 1.46) Black Carbon High 1.00 1.92 (1.00 - 3.68) 1.42 (0.74 - 2.72) 1.93 (1.02 - 3.65) Low 1.00 0.63 (0.35 - 1.14) 0.78 (0.46 - 1.32) 0.93 (0.58 - 1.49) Governmental transfers PM2.5 High 1.00 1.44 (0.83 - 2.51) 1.37 (0.80 - 2.37) 1.63 (0.95 - 2.81) Low 1.00 0.63 (0.40 - 0.99) 0.84 (0.53 - 1.32) 1.04 (0.66 - 1.63) NO High 1.00 0.77 (0.34 - 1.73) 1.08 (0.52 - 2.22) 1.39 (0.69 - 2.77) Low 1.00 0.70 (0.47 - 1.05) 1.12 (0.68 - 1.85) 0.81 (0.44 - 1.49) NO2 High 1.00 0.98 (0.35 - 2.73) 1.10 (0.44 - 2.77) 1.12 (0.45 - 2.77) Low 1.00 1.38 (0.93 - 2.05) 0.97 (0.58 - 1.61) 0.89 (0.56 - 1.41) Black Carbon High 1.00 1.80 (0.91 - 3.57) 1.48 (0.76 - 2.88) 2.02 (1.06 - 3.86) Low 1.00 0.95 (0.62 - 1.44) 1.49 (0.98 - 2.29) 0.98 (0.55 - 1.73) Low income PM2.5 High 1.00 1.17 (0.66 - 2.07) 0.99 (0.57 - 1.74) 0.97 (0.56 - 1.69) Low 1.00 0.62 (0.32 - 1.20) 1.28 (0.77 - 2.12) 1.29 (0.79 - 2.10) NO High 1.00 0.67 (0.43 - 1.05) 0.86 (0.52 - 1.42) 0.99 (0.58 - 1.70) Low 1.00 1.24 (0.62 - 2.46) 0.98 (0.52 - 1.85) 1.23 (0.69 - 2.19) NO2 High 1.00 0.69 (0.45 - 1.07) 1.00 (0.54 - 1.84) 1.14 (0.63 - 2.06) Low 1.00 2.28 (0.95 - 5.52) 2.37 (1.03 - 5.49) 2.91 (1.26 - 6.72) Black Carbon High 1.00 1.32 (0.87 - 2.00) 0.95 (0.56 - 1.64) 0.72 (0.41 - 1.24) Low 1.00 1.38 (0.79 - 2.43) 1.81 (1.04 - 3.16) 1.47 (0.87 - 2.50) Home ownership PM2.5 High 1.00 0.81 (0.51 - 1.27) 1.40 (0.90 - 2.18) 0.73 (0.37 - 1.43) Low 1.00 0.58 (0.35 - 0.97) 0.80 (0.50 - 1.28) 0.98 (0.62 - 1.54) NO High 1.00 0.65 (0.37 - 1.14) 0.88 (0.53 - 1.46) 0.96 (0.61 - 1.53) Low 1.00 0.74 (0.49 - 1.11) 0.91 (0.56 - 1.46) 0.60 (0.31 - 1.19) NO2 High 1.00 0.60 (0.33 - 1.10) 0.70 (0.42 - 1.18) 0.71 (0.44 - 1.14) Low 1.00 1.14 (0.74 - 1.75) 0.97 (0.57 - 1.64) 0.89 (0.54 - 1.46) Black Carbon High 1.00 2.37 (1.15 - 4.88) 1.91 (0.94 - 3.87) 2.30 (1.14 - 4.64) Low 1.00 1.11 (0.69 - 1.78) 1.70 (1.07 - 2.71) 0.90 (0.49 - 1.63) Neighborhood stress PM2.5 High 1.00 1.08 (0.63 - 1.86) 1.09 (0.64 - 1.87) 1.02 (0.61 - 1.71)      103 Table 35. Comparison between HR of different road proximity categories for low and high levels of DA-SES variables, when considering ACS and CHF health outcomes Health Outcomes DA-level SES Pollutant SES Level ACS CHF Low 1.40 (1.00 - 1.96) 0.51 (0.16 - 1.60) Within 50 m from expressways and primary highways (I)  High 0.76 (0.31 - 1.82) 1.32 (0.33 - 5.32) Low 1.56 (1.19 - 2.05) 0.81 (0.38 - 1.71) Between 50 and 150 m from expressways and primary highways (II) High 0.70 (0.47 - 1.04) 1.57 (0.88 - 2.79) Low 0.93 (0.75 - 1.15) 1.17 (0.78 - 1.75) Within 50 m from secondary highways and major roads (III) High 1.24 (0.96 - 1.58) 1.60 (1.01 - 2.53) Low 1.14 (0.97 - 1.34) 0.69 (0.46 - 1.04) Between 50 and 150 m from secondary highways and major roads (IV) High 0.97 (0.80 - 1.18) 0.87 (0.58 - 1.32) Low 1.17 (0.99 - 1.38) 0.93 (0.65 - 1.35) Chinese population Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 1.03 (0.83 - 1.28) 1.74 (1.18 - 2.55) Low 1.13 (0.81 - 1.59) 0.83 (0.37 - 1.87) Within 50 m from expressways and primary highways (I)  High 1.22 (0.61 - 2.46) 0.60 (0.08 - 4.29) Low 0.99 (0.76 - 1.29) 0.54 (0.27 - 1.10) Between 50 and 150 m from expressways and primary highways (II) High 0.96 (0.66 - 1.40) 1.04 (0.51 - 2.13) Low 1.03 (0.84 - 1.27) 1.37 (0.93 - 2.00) Within 50 m from secondary highways and major roads (III) High 1.09 (0.84 - 1.41) 1.33 (0.84 - 2.12) Low 1.03 (0.86 - 1.22) 0.89 (0.60 - 1.32) Between 50 and 150 m from secondary highways and major roads (IV) High 1.03 (0.86 - 1.24) 1.09 (0.76 - 1.56) Low 1.07 (0.91 - 1.25) 1.04 (0.74 - 1.45) University Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 1.05 (0.85 - 1.31) 1.13 (0.74 - 1.71) Low 1.25 (0.86 - 1.81) 1.58 (0.81 - 3.07) Within 50 m from expressways and primary highways (I)  High 1.24 (0.66 - 2.31) 0.00 (0.00 - ∞) Low 1.06 (0.82 - 1.38) 0.85 (0.49 - 1.49) Between 50 and 150 m from expressways and primary highways (II) High 1.22 (0.86 - 1.75) 1.09 (0.51 - 2.33) Low 1.10 (0.91 - 1.33) 1.32 (0.94 - 1.86) Within 50 m from secondary highways and major roads (III) High 0.96 (0.74 - 1.26) 1.65 (1.06 - 2.57) Low 0.94 (0.80 - 1.10) 0.83 (0.60 - 1.15) Between 50 and 150 m from secondary highways and major roads (IV) High 0.89 (0.73 - 1.10) 0.71 (0.45 - 1.12) Low 1.14 (0.98 - 1.33) 1.28 (0.95 - 1.71) Employment Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 1.05 (0.85 - 1.31) 1.29 (0.86 - 1.94) Low 1.38 (1.02 - 1.87) 1.35 (0.72 - 2.56) Within 50 m from expressways and primary highways (I)  High 1.23 (0.61 - 2.47) 0.66 (0.09 - 4.73) Low 0.97 (0.76 - 1.24) 1.12 (0.70 - 1.80) Between 50 and 150 m from expressways and primary highways (II) High 1.06 (0.73 - 1.55) 0.82 (0.34 - 2.01) Low 1.04 (0.85 - 1.26) 1.33 (0.91 - 1.93) Within 50 m from secondary highways and major roads (III) High 1.00 (0.77 - 1.3) 1.12 (0.67 - 1.86) Low 1.01 (0.86 - 1.18) 0.69 (0.48 - 0.99) Between 50 and 150 m from secondary highways and major roads (IV) High 1.03 (0.86 - 1.24) 0.96 (0.65 - 1.42) Low 1.08 (0.92 - 1.26) 1.48 (1.10 - 1.99) Family income Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 1.04 (0.84 - 1.29) 1.02 (0.65 - 1.59) Low 1.44 (1.03 - 2.02) 1.72 (0.91 - 3.25) Within 50 m from expressways and primary highways (I)  High 1.52 (0.86 - 2.68) 0.56 (0.08 - 4.03) Low 0.87 (0.66 - 1.15) 0.85 (0.49 - 1.50) Between 50 and 150 m from expressways and primary highways (II) High 0.86 (0.58 - 1.29) 1.11 (0.52 - 2.38) Low 0.98 (0.79 - 1.22) 1.42 (0.96 - 2.10) Within 50 m from secondary highways and major roads (III) High 0.94 (0.72 - 1.23) 1.05 (0.61 - 1.80) Low 0.98 (0.83 - 1.16) 0.79 (0.55 - 1.14) Between 50 and 150 m from secondary highways and major roads (IV) High 1.11 (0.93 - 1.32) 1.12 (0.76 - 1.63) Low 1.01 (0.86 - 1.20) 1.37 (1.00 - 1.89) Personal income Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 0.97 (0.78 - 1.20) 1.04 (0.66 - 1.64)           104   Table 35. Comparison between HR of different road proximity categories for low and high levels of DA-SES variables, when considering ACS and CHF health outcomes (cont.) Health Outcomes DA-level SES Pollutant Level ACS CHF Low 1.09 (0.64 - 1.85) 1.23 (0.39 - 3.84) Within 50 m from expressways and primary highways (I)  High 1.44 (1.00 - 2.06) 1.48 (0.73 - 3.01) Low 1.11 (0.79 - 1.57) 1.10 (0.54 - 2.25) Between 50 and 150 m from expressways and primary highways (II) High 1.00 (0.76 - 1.31) 1.36 (0.84 - 2.19) Low 1.02 (0.81 - 1.28) 0.74 (0.42 - 1.28) Within 50 m from secondary highways and major roads (III) High 1.21 (0.98 - 1.50) 1.47 (1.00 - 2.16) Low 0.99 (0.81 - 1.20) 1.02 (0.67 - 1.57) Between 50 and 150 m from secondary highways and major roads (IV) High 1.00 (0.85 - 1.19) 0.91 (0.65 - 1.27) Low 1.07 (0.88 - 1.29) 0.90 (0.59 - 1.38) Transportation Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 1.18 (0.99 - 1.40) 1.65 (1.20 - 2.27) Low 1.73 (1.11 - 2.70) 0.86 (0.21 - 3.46) Within 50 m from expressways and primary highways (I)  High 1.23 (0.86 - 1.75) 1.35 (0.63 - 2.88) Low 1.29 (0.93 - 1.80) 1.27 (0.62 - 2.59) Between 50 and 150 m from expressways and primary highways (II) High 1.02 (0.79 - 1.32) 1.45 (0.89 - 2.37) Low 1.03 (0.82 - 1.30) 1.30 (0.82 - 2.06) Within 50 m from secondary highways and major roads (III) High 1.08 (0.88 - 1.33) 1.54 (1.05 - 2.28) Low 1.06 (0.88 - 1.27) 0.87 (0.56 - 1.35) Between 50 and 150 m from secondary highways and major roads (IV) High 1.07 (0.92 - 1.25) 0.88 (0.62 - 1.24) Low 1.17 (0.97 - 1.41) 1.21 (0.81 - 1.83) Low income Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 1.09 (0.93 - 1.28) 1.77 (1.29 - 2.44) Low 1.19 (0.84 - 1.70) 1.28 (0.63 - 2.59) Within 50 m from expressways and primary highways (I)  High 1.83 (1.19 - 2.79) 1.16 (0.37 - 3.65) Low 1.06 (0.83 - 1.36) 1.31 (0.84 - 2.07) Between 50 and 150 m from expressways and primary highways (II) High 1.13 (0.80 - 1.59) 0.63 (0.23 - 1.69) Low 0.99 (0.81 - 1.21) 1.31 (0.91 - 1.87) Within 50 m from secondary highways and major roads (III) High 1.07 (0.84 - 1.36) 1.18 (0.70 - 2.00) Low 0.98 (0.84 - 1.15) 0.95 (0.70 - 1.29) Between 50 and 150 m from secondary highways and major roads (IV) High 1.15 (0.96 - 1.39) 0.95 (0.60 - 1.52) Low 1.02 (0.87 - 1.20) 1.46 (1.09 - 1.97) Home ownership Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 1.20 (0.99 - 1.45) 1.03 (0.65 - 1.62) Low 1.17 (0.80 - 1.72) 1.15 (0.51 - 2.6) Within 50 m from expressways and primary highways (I)  High 1.20 (0.57 - 2.53) 0.74 (0.10 - 5.30) Low 0.98 (0.74 - 1.29) 0.83 (0.44 - 1.57) Between 50 and 150 m from expressways and primary highways (II) High 1.11 (0.78 - 1.6) 1.02 (0.48 - 2.18) Low 1.10 (0.9 - 1.34) 1.54 (1.06 - 2.23) Within 50 m from secondary highways and major roads (III) High 1.04 (0.81 - 1.33) 1.19 (0.76 - 1.88) Low 0.95 (0.79 - 1.13) 0.66 (0.43 - 1.01) Between 50 and 150 m from secondary highways and major roads (IV) High 1.01 (0.84 - 1.21) 0.95 (0.65 - 1.38) Low 1.07 (0.91 - 1.27) 1.39 (1.01 - 1.93) Income variation Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 1.10 (0.89 - 1.35) 1.16 (0.78 - 1.73) Low 1.20 (0.83 - 1.76) 1.35 (0.63 - 2.87) Within 50 m from expressways and primary highways (I)  High 1.06 (0.59 - 1.93) 0.94 (0.23 - 3.8) Low 1.00 (0.76 - 1.32) 0.98 (0.56 - 1.72) Between 50 and 150 m from expressways and primary highways (II) High 0.96 (0.69 - 1.34) 1.38 (0.75 - 2.56) Low 1.06 (0.85 - 1.31) 1.55 (1.05 - 2.28) Within 50 m from secondary highways and major roads (III) High 1.08 (0.85 - 1.38) 1.29 (0.81 - 2.08) Low 0.94 (0.80 - 1.12) 0.75 (0.51 - 1.10) Between 50 and 150 m from secondary highways and major roads (IV) High 0.87 (0.73 - 1.05) 0.97 (0.66 - 1.43) Low 1.05 (0.89 - 1.25) 1.40 (1.01 - 1.95) Management Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 1.07 (0.88 - 1.30) 1.26 (0.85 - 1.88)        105  Table 36. Comparison between HR of different road proximity categories for low and high levels of Neighborhood-SES variables, when considering ACS and CHF health outcomes Health Outcomes Neighborhood- level SES Pollutant Level ACS CHF Low 1.15 (0.77 - 1.72) 0.69 (0.22 - 2.15) Within 50 m from expressways and primary highways (I)  High 1.23 (0.75 - 2.02) 1.65 (0.67 - 4.02) Low 1.36 (1.01 - 1.81) 0.69 (0.30 - 1.55) Between 50 and 150 m from expressways and primary highways (II) High 0.79 (0.55 - 1.12) 1.24 (0.69 - 2.24) Low 0.98 (0.79 - 1.2) 1.42 (0.95 - 2.10) Within 50 m from secondary highways and major roads (III) High 1.09 (0.86 - 1.37) 1.45 (0.94 - 2.23) Low 1.06 (0.89 - 1.26) 0.60 (0.38 - 0.94) Between 50 and 150 m from secondary highways and major roads (IV) High 0.94 (0.79 - 1.13) 0.91 (0.63 - 1.33) Low 1.12 (0.95 - 1.32) 1.24 (0.87 - 1.77) Other language Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 0.97 (0.80 - 1.19) 1.47 (1.03 - 2.11) Low 1.22 (0.81 - 1.85) 1.02 (0.38 - 2.75) Within 50 m from expressways and primary highways (I)  High 1.15 (0.66 - 1.98) 1.93 (0.79 - 4.69) Low 1.27 (0.93 - 1.74) 0.49 (0.18 - 1.33) Between 50 and 150 m from expressways and primary highways (II) High 0.81 (0.57 - 1.15) 1.24 (0.69 - 2.23) Low 1.02 (0.82 - 1.26) 1.67 (1.13 - 2.48) Within 50 m from secondary highways and major roads (III) High 1.07 (0.85 - 1.35) 1.42 (0.93 - 2.17) Low 1.06 (0.89 - 1.25) 0.58 (0.37 - 0.92) Between 50 and 150 m from secondary highways and major roads (IV) High 0.94 (0.79 - 1.12) 0.86 (0.60 - 1.25) Low 1.13 (0.95 - 1.34) 1.37 (0.95 - 1.96) Linguistic isolation Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 0.96 (0.79 - 1.17) 1.48 (1.04 - 2.11) Low 1.12 (0.78 - 1.59) 0.60 (0.22 - 1.62) Within 50 m from expressways and primary highways (I)  High 1.19 (0.67 - 2.11) 0.73 (0.18 - 2.95) Low 1.10 (0.81 - 1.50) 0.97 (0.48 - 1.96) Between 50 and 150 m from expressways and primary highways (II) High 1.03 (0.74 - 1.42) 0.88 (0.45 - 1.72) Low 1.00 (0.82 - 1.22) 1.09 (0.73 - 1.62) Within 50 m from secondary highways and major roads (III) High 1.07 (0.85 - 1.36) 1.42 (0.96 - 2.11) Low 1.14 (0.95 - 1.35) 0.84 (0.54 - 1.28) Between 50 and 150 m from secondary highways and major roads (IV) High 1.02 (0.85 - 1.21) 1.06 (0.77 - 1.47) Low 1.05 (0.89 - 1.24) 1.03 (0.73 - 1.45) University Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 1.10 (0.91 - 1.34) 1.17 (0.82 - 1.67) Low 1.26 (0.79 - 2.02) 1.00 (0.32 - 3.15) Within 50 m from expressways and primary highways (I)  High 1.29 (0.90 - 1.86) 1.81 (0.92 - 3.55) Low 1.39 (1.03 - 1.87) 0.75 (0.33 - 1.71) Between 50 and 150 m from expressways and primary highways (II) High 0.91 (0.68 - 1.22) 0.98 (0.53 - 1.82) Low 1.02 (0.82 - 1.26) 1.23 (0.80 - 1.90) Within 50 m from secondary highways and major roads (III) High 1.13 (0.91 - 1.40) 1.47 (0.97 - 2.23) Low 1.01 (0.84 - 1.20) 1.18 (0.81 - 1.72) Between 50 and 150 m from secondary highways and major roads (IV) High 0.93 (0.78 - 1.10) 0.80 (0.55 - 1.17) Low 1.17 (0.98 - 1.39) 1.15 (0.79 - 1.69) Unemployment Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 1.05 (0.88 - 1.25) 1.50 (1.06 - 2.12) Low 1.21 (0.80 - 1.84) 0.98 (0.36 - 2.64) Within 50 m from expressways and primary highways (I)  High 1.36 (0.78 - 2.35) 0.50 (0.07 - 3.59) Low 0.85 (0.63 - 1.14) 1.05 (0.59 - 1.84) Between 50 and 150 m from expressways and primary highways (II) High 1.24 (0.89 - 1.73) 1.34 (0.68 - 2.63) Low 1.09 (0.87 - 1.36) 1.51 (1.01 - 2.26) Within 50 m from secondary highways and major roads (III) High 1.05 (0.83 - 1.32) 1.13 (0.70 - 1.80) Low 0.95 (0.80 - 1.13) 0.70 (0.47 - 1.03) Between 50 and 150 m from secondary highways and major roads (IV) High 1.09 (0.92 - 1.30) 0.80 (0.54 - 1.19) Low 0.97 (0.81 - 1.16) 1.43 (1.02 - 2.01) Family income Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 1.16 (0.96 - 1.41) 1.06 (0.70 - 1.60)         106 Table 36. Comparison between HR of different road proximity categories for low and high levels of Neighborhood -SES variables, when considering ACS and CHF health outcomes (cont.) Health Outcomes Neighborhood- level SES Pollutant Level ACS CHF Low 1.38 (0.96 - 1.99) 1.12 (0.49 - 2.52) Within 50 m from expressways and primary highways (I)  High 1.20 (0.66 - 2.18) 0.98 (0.24 - 3.97) Low 0.83 (0.61 - 1.13) 1.02 (0.58 - 1.81) Between 50 and 150 m from expressways and primary highways (II) High 1.05 (0.75 - 1.45) 1.19 (0.62 - 2.25) Low 1.13 (0.91 - 1.42) 1.50 (1.00 - 2.24) Within 50 m from secondary highways and major roads (III) High 1.06 (0.83 - 1.34) 1.28 (0.83 - 1.99) Low 0.92 (0.77 - 1.09) 0.67 (0.45 - 1.00) Between 50 and 150 m from secondary highways and major roads (IV) High 1.07 (0.90 - 1.27) 1.04 (0.73 - 1.47) Low 1.02 (0.86 - 1.23) 1.37 (0.98 - 1.92) Personal income Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 1.10 (0.91 - 1.34) 1.21 (0.82 - 1.77) Low 0.65 (0.21 - 2.01) 0 (0 - 0) Within 50 m from expressways and primary highways (I)  High 1.28 (0.90 - 1.83) 1.31 (0.65 - 2.67) Low 0.92 (0.59 - 1.46) 0.47 (0.12 - 1.90) Between 50 and 150 m from expressways and primary highways (II) High 0.85 (0.65 - 1.12) 1.08 (0.66 - 1.78) Low 0.96 (0.73 - 1.25) 1.17 (0.68 - 2.02) Within 50 m from secondary highways and major roads (III) High 1.02 (0.83 - 1.25) 1.46 (1.01 - 2.11) Low 1.07 (0.89 - 1.29) 0.84 (0.54 - 1.31) Between 50 and 150 m from secondary highways and major roads (IV) High 0.93 (0.79 - 1.10) 0.68 (0.47 - 0.99) Low 0.99 (0.79 - 1.25) 0.93 (0.56 - 1.56) Governmental transfers Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 1.00 (0.85 - 1.18) 1.44 (1.06 - 1.96) Low 1.48 (1.00 - 2.19) 1.10 (0.41 - 2.97) Within 50 m from expressways and primary highways (I)  High 1.38 (0.90 - 2.11) 2.12 (0.99 - 4.54) Low 1.38 (1.01 - 1.90) 0.56 (0.21 - 1.53) Between 50 and 150 m from expressways and primary highways (II) High 1.05 (0.79 - 1.40) 0.95 (0.50 - 1.80) Low 0.92 (0.73 - 1.15) 1.08 (0.67 - 1.73) Within 50 m from secondary highways and major roads (III) High 0.99 (0.79 - 1.26) 1.66 (1.10 - 2.51) Low 1.05 (0.87 - 1.27) 0.91 (0.59 - 1.43) Between 50 and 150 m from secondary highways and major roads (IV) High 0.98 (0.82 - 1.17) 1.02 (0.71 - 1.47) Low 1.11 (0.93 - 1.33) 0.96 (0.64 - 1.45) Low income Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 1.04 (0.86 - 1.25) 1.69 (1.19 - 2.41) Low 1.41 (0.97 - 2.04) 1.26 (0.56 - 2.86) Within 50 m from expressways and primary highways (I)  High 1.21 (0.75 - 1.95) 0.77 (0.19 - 3.11) Low 0.92 (0.70 - 1.22) 1.23 (0.75 - 2.04) Between 50 and 150 m from expressways and primary highways (II) High 1.53 (1.06 - 2.19) 0.78 (0.25 - 2.47) Low 1.18 (0.96 - 1.46) 1.52 (1.03 - 2.24) Within 50 m from secondary highways and major roads (III) High 0.88 (0.68 - 1.13) 1.09 (0.63 - 1.86) Low 1.07 (0.91 - 1.27) 1.10 (0.79 - 1.53) Between 50 and 150 m from secondary highways and major roads (IV) High 1.09 (0.90 - 1.32) 1.02 (0.64 - 1.63) Low 1.10 (0.93 - 1.31) 1.57 (1.14 - 2.18) Home ownership Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 1.06 (0.87 - 1.30) 0.99 (0.61 - 1.59) Low 1.56 (0.86 - 2.84) 1.32 (0.33 - 5.34) Within 50 m from expressways and primary highways (I)  High 1.40 (0.98 - 2.01) 1.50 (0.70 - 3.20) Low 1.22 (0.87 - 1.72) 0.68 (0.25 - 1.84) Between 50 and 150 m from expressways and primary highways (II) High 1.08 (0.82 - 1.41) 0.92 (0.50 - 1.71) Low 0.91 (0.71 - 1.17) 1.13 (0.67 - 1.92) Within 50 m from secondary highways and major roads (III) High 0.99 (0.80 - 1.24) 1.42 (0.95 - 2.12) Low 1.01 (0.84 - 1.22) 0.88 (0.55 - 1.39) Between 50 and 150 m from secondary highways and major roads (IV) High 0.98 (0.82 - 1.16) 1.01 (0.71 - 1.43) Low 1.08 (0.88 - 1.31) 1.04 (0.65 - 1.64) Neighborhood stress Within 150 m from expressways and primary highways or within 50 m from secondary highways and major roads (V) High 1.05 (0.89 - 1.25) 1.44 (1.03 - 2.02)       107 Appendix I Table 37. Distribution of health outcomes by different covariate categories (DA-SES) for black carbon traffic pollution ACS CCS CHF Variables Category Censored Event Censored Event Censored Event Male 158,125 2,443 158,125 2,443 158,125 2,443 Sex   Female 183,324 1,125 183,324 1,125 183,324 1,125 Born before 1925 28,511 862 28,511 862 28,511 862 Born btw 1925 - 1934 54,737 1,002 54,737 1,002 54,737 1,002 Born btw 1935 - 1944 94,475 940 94,475 940 94,475 940 Age     Born after 1944 163,726 764 163,726 764 163,726 764 1ST QUARTILE- LOW 68,122 888 68,089 921 68,821 189 2ND QUARTILE 68,648 823 68,578 893 69,286 185 3RD QUARTILE 67,854 703 67,776 781 68,402 155 4TH QUARTILE 68,387 613 68,321 679 68,861 139 Chinese minority      5TH QUARTILE - HIGH 68,438 541 68,398 581 68,855 124 1ST QUARTILE- LOW 68,315 723 68,276 762 68,866 172 2ND QUARTILE 68,203 757 68,171 789 68,784 176 3RD QUARTILE 68,368 761 68,301 828 68,950 179 4TH QUARTILE 68,178 688 68,095 771 68,733 133 Income      5TH QUARTILE - HIGH 68,385 639 68,319 705 68,892 132 1ST QUARTILE- LOW 68,577 880 68,531 926 69,267 190 2ND QUARTILE 68,095 719 68,015 799 68,658 156 3RD QUARTILE 67,794 757 67,769 782 68,392 159 4TH QUARTILE 68,374 656 68,351 679 68,880 150 University      5TH QUARTILE - HIGH 68,609 556 68,496 669 69,028 137 1ST QUARTILE- LOW 69,292 767 69,222 837 69,907 152 2ND QUARTILE 68,737 742 68,640 839 69,327 152 3RD QUARTILE 65,494 716 65,437 773 66,073 137 4TH QUARTILE 69,490 709 69,485 714 70,013 186 Transportation      5TH QUARTILE - HIGH 68,436 634 68,378 692 68,905 165 1ST QUARTILE- LOW 68,204 813 68,129 888 68,840 177 2ND QUARTILE 67,815 733 67,771 777 68,382 166 3RD QUARTILE 70,916 732 70,855 793 71,482 166 4TH QUARTILE 66,015 676 65,989 702 66,555 136 Coefficient of variation 5TH QUARTILE - HIGH 68,499 614 68,418 695 68,966 147 1ST QUARTILE- LOW 68,268 764 68,248 784 68,838 194 2ND QUARTILE 67,800 698 67,756 742 68,340 158 3RD QUARTILE 69,308 688 69,263 733 69,827 169 4TH QUARTILE 67,191 671 67,102 760 67,719 143 Percent of owned dwellings 5TH QUARTILE - HIGH 68,882 747 68,793 836 69,501 128 1ST QUARTILE- LOW 68,191 822 68,175 838 68,823 190 2ND QUARTILE 68,236 753 68,192 797 68,822 167 3RD QUARTILE 68,290 696 68,242 744 68,826 160 4TH QUARTILE 68,384 651 68,258 777 68,899 136 Average 2000 family income ($) 5TH QUARTILE - HIGH 68,348 646 68,295 699 68,855 139 1ST QUARTILE- LOW 68,239 834 68,219 854 68,853 220 2ND QUARTILE 68,109 752 68,071 790 68,705 156 3RD QUARTILE 68,043 680 68,000 723 68,578 145 4TH QUARTILE 68,707 698 68,654 751 69,273 132 Employment rate (%) 5TH QUARTILE - HIGH 68,351 604 68,218 737 68,816 139 1ST QUARTILE- LOW 68,247 768 68,222 793 68,842 173 2ND QUARTILE 67,409 715 67,343 781 67,964 160 3RD QUARTILE 68,000 693 67,935 758 68,518 175 4TH QUARTILE 69,583 714 69,533 764 70,154 143 % people in labor working in management 5TH QUARTILE - HIGH 68,210 678 68,129 759 68,747 141 1ST QUARTILE- LOW 68,074 764 68,048 790 68,693 145 2ND QUARTILE 68,109 698 68,008 799 68,654 153 3RD QUARTILE 68,792 677 68,723 746 69,303 166 4TH QUARTILE 67,990 653 67,891 752 68,482 161 Incidence of low income in 2000 % 5TH QUARTILE - HIGH 68,484 776 68,492 768 69,093 167   108 Table 38. Distribution of health outcomes by different covariate categories (DA-SES) for NO traffic pollution ACS CCS CHF Variables Category Censored Event Censored Event Censored Event Male 158,677 2,455 158,677 2,455 158,677 2,455 Sex   Female 183,974 1,126 183,974 1,126 183,974 1,126 Born before 1925 28,593 863 28,593 863 28,593 863 Born btw 1925 - 1934 54,911 1,008 54,911 1,008 54,911 1,008 Born btw 1935 - 1944 94,843 942 94,843 942 94,843 942 Age     Born after 1944 164,304 768 164,304 768 164,304 768 1ST QUARTILE- LOW 68,548 891 68,508 931 69,248 191 2ND QUARTILE 67,908 819 67,837 890 68,546 181 3RD QUARTILE 68,745 709 68,668 786 69,297 157 4TH QUARTILE 68,837 621 68,773 685 69,317 141 Chinese minority      5TH QUARTILE - HIGH 68,613 541 68,572 582 69,030 124 1ST QUARTILE- LOW 68,527 726 68,483 770 69,079 174 2ND QUARTILE 68,444 758 68,411 791 69,026 176 3RD QUARTILE 68,548 765 68,482 831 69,134 179 4TH QUARTILE 68,545 690 68,461 774 69,102 133 Income      5TH QUARTILE - HIGH 68,587 642 68,521 708 69,097 132 1ST QUARTILE- LOW 68,028 871 67,971 928 68,709 190 2ND QUARTILE 69,391 737 69,317 811 69,970 158 3RD QUARTILE 67,963 757 67,936 784 68,561 159 4TH QUARTILE 68,450 657 68,427 680 68,957 150 University      5TH QUARTILE - HIGH 68,819 559 68,707 671 69,241 137 1ST QUARTILE- LOW 69,866 773 69,789 850 70,485 154 2ND QUARTILE 68,852 742 68,755 839 69,442 152 3RD QUARTILE 65,725 721 65,672 774 66,309 137 4TH QUARTILE 69,703 709 69,694 718 70,226 186 Transportation      5TH QUARTILE - HIGH 68,505 636 68,448 693 68,976 165 1ST QUARTILE- LOW 68,429 813 68,353 889 69,065 177 2ND QUARTILE 68,310 739 68,259 790 68,881 168 3RD QUARTILE 71,066 737 71,008 795 71,637 166 4TH QUARTILE 66,123 677 66,098 702 66,664 136 Coefficient of variation 5TH QUARTILE - HIGH 68,723 615 68,640 698 69,191 147 1ST QUARTILE- LOW 68,457 766 68,438 785 69,029 194 2ND QUARTILE 68,314 705 68,261 758 68,859 160 3RD QUARTILE 68,265 670 68,216 719 68,769 166 4TH QUARTILE 68,499 693 68,417 775 69,046 146 Percent of owned dwellings 5TH QUARTILE - HIGH 69,116 747 69,026 837 69,735 128 1ST QUARTILE- LOW 68,412 824 68,390 846 69,044 192 2ND QUARTILE 68,452 755 68,408 799 69,040 167 3RD QUARTILE 68,649 698 68,598 749 69,187 160 4TH QUARTILE 68,573 657 68,450 780 69,094 136 Average 2000 family income ($) 5TH QUARTILE - HIGH 68,565 647 68,512 700 69,073 139 1ST QUARTILE- LOW 68,500 834 68,478 856 69,114 220 2ND QUARTILE 68,108 753 68,072 789 68,705 156 3RD QUARTILE 68,869 687 68,827 729 69,409 147 4TH QUARTILE 68,470 696 68,408 758 69,034 132 Employment rate (%) 5TH QUARTILE - HIGH 68,704 611 68,573 742 69,176 139 1ST QUARTILE- LOW 68,643 773 68,611 805 69,241 175 2ND QUARTILE 67,683 717 67,617 783 68,240 160 3RD QUARTILE 68,095 694 68,028 761 68,614 175 4TH QUARTILE 69,887 718 69,840 765 70,462 143 % people in labor working in management 5TH QUARTILE - HIGH 68,343 679 68,262 760 68,881 141 1ST QUARTILE- LOW 68,400 768 68,376 792 69,023 145 2ND QUARTILE 68,662 704 68,553 813 69,211 155 3RD QUARTILE 68,637 676 68,567 746 69,148 165 4TH QUARTILE 68,390 656 68,291 755 68,884 162 Incidence of low income in 2000 % 5TH QUARTILE - HIGH 68,562 777 68,571 768 69,172 167     109 Table 39. Distribution of health outcomes by different covariate categories (DA-SES) for NO2 traffic pollution ACS CCS CHF Variables Category Censored Event Censored Event Censored Event Male 158,664 2,454 158,664 2,454 158,664 2,454 Sex   Female 183,966 1,126 183,966 1,126 183,966 1,126 Born before 1925 28,592 862 28,592 862 28,592 862 Born btw 1925 - 1934 54,909 1,008 54,909 1,008 54,909 1,008 Born btw 1935 - 1944 94,836 942 94,836 942 94,836 942 Age     Born after 1944 164,293 768 164,293 768 164,293 768 1ST QUARTILE- LOW 68,545 890 68,504 931 69,244 191 2ND QUARTILE 67,904 819 67,833 890 68,542 181 3RD QUARTILE 68,731 709 68,654 786 69,283 157 4TH QUARTILE 68,837 621 68,773 685 69,317 141 Chinese minority      5TH QUARTILE - HIGH 68,613 541 68,572 582 69,030 124 1ST QUARTILE- LOW 68,526 726 68,482 770 69,078 174 2ND QUARTILE 68,443 758 68,410 791 69,025 176 3RD QUARTILE 68,482 762 68,414 830 69,065 179 4TH QUARTILE 68,604 692 68,521 775 69,163 133 Income      5TH QUARTILE - HIGH 68,575 642 68,509 708 69,085 132 1ST QUARTILE- LOW 68,026 871 67,969 928 68,707 190 2ND QUARTILE 69,388 737 69,314 811 69,967 158 3RD QUARTILE 67,962 757 67,935 784 68,560 159 4TH QUARTILE 68,445 656 68,421 680 68,951 150 University      5TH QUARTILE - HIGH 68,809 559 68,697 671 69,231 137 1ST QUARTILE- LOW 69,853 773 69,776 850 70,472 154 2ND QUARTILE 68,849 742 68,752 839 69,439 152 3RD QUARTILE 65,721 721 65,668 774 66,305 137 4TH QUARTILE 69,703 709 69,694 718 70,226 186 Transportation      5TH QUARTILE - HIGH 68,504 635 68,446 693 68,974 165 1ST QUARTILE- LOW 68,428 813 68,352 889 69,064 177 2ND QUARTILE 68,309 738 68,257 790 68,879 168 3RD QUARTILE 71,053 737 70,995 795 71,624 166 4TH QUARTILE 66,121 677 66,096 702 66,662 136 Coefficient of variation 5TH QUARTILE - HIGH 68,719 615 68,636 698 69,187 147 1ST QUARTILE- LOW 68,455 765 68,435 785 69,026 194 2ND QUARTILE 68,312 705 68,259 758 68,857 160 3RD QUARTILE 68,262 670 68,213 719 68,766 166 4TH QUARTILE 68,490 693 68,408 775 69,037 146 Percent of owned dwellings 5TH QUARTILE - HIGH 69,111 747 69,021 837 69,730 128 1ST QUARTILE- LOW 68,410 824 68,388 846 69,042 192 2ND QUARTILE 68,452 754 68,407 799 69,039 167 3RD QUARTILE 68,545 698 68,494 749 69,083 160 4TH QUARTILE 68,670 657 68,547 780 69,191 136 Average 2000 family income ($) 5TH QUARTILE - HIGH 68,553 647 68,500 700 69,061 139 1ST QUARTILE- LOW 68,500 833 68,477 856 69,113 220 2ND QUARTILE 68,106 753 68,070 789 68,703 156 3RD QUARTILE 68,867 687 68,825 729 69,407 147 4TH QUARTILE 68,453 696 68,391 758 69,017 132 Employment rate (%) 5TH QUARTILE - HIGH 68,704 611 68,573 742 69,176 139 1ST QUARTILE- LOW 68,642 773 68,610 805 69,240 175 2ND QUARTILE 67,681 717 67,615 783 68,238 160 3RD QUARTILE 68,094 694 68,027 761 68,613 175 4TH QUARTILE 69,876 717 69,828 765 70,450 143 % people in labor working in management 5TH QUARTILE - HIGH 68,337 679 68,256 760 68,875 141 1ST QUARTILE- LOW 68,386 768 68,362 792 69,009 145 2ND QUARTILE 68,660 704 68,551 813 69,209 155 3RD QUARTILE 68,636 676 68,566 746 69,147 165 4TH QUARTILE 68,387 656 68,288 755 68,881 162 Incidence of low income in 2000 % 5TH QUARTILE - HIGH 68,561 776 68,569 768 69,170 167     110 Table 40. Distribution of health outcomes by different covariate categories (DA-SES) for PM2.5 traffic pollution ACS CCS CHF Variables Category Censored Event Censored Event Censored Event Male 149,974 2,320 149,974 2,320 149,974 2,320 Sex   Female 174,916 1,080 174,916 1,080 174,916 1,080 Born before 1925 27,467 824 27,467 824 27,467 824 Born btw 1925 - 1934 52,185 958 52,185 958 52,185 958 Born btw 1935 - 1944 89,562 899 89,562 899 89,562 899 Age     Born after 1944 155,676 719 155,676 719 155,676 719 1ST QUARTILE- LOW 64,612 863 64,613 862 65,294 181 2ND QUARTILE 65,111 771 65,034 848 65,703 179 3RD QUARTILE 64,808 673 64,753 728 65,334 147 4TH QUARTILE 65,209 574 65,140 643 65,650 133 Chinese minority      5TH QUARTILE - HIGH 65,150 519 65,110 559 65,553 116 1ST QUARTILE- LOW 64,937 674 64,906 705 65,447 164 2ND QUARTILE 64,941 722 64,919 744 65,493 170 3RD QUARTILE 64,947 733 64,896 784 65,509 171 4TH QUARTILE 65,022 657 64,949 730 65,548 131 Income      5TH QUARTILE - HIGH 65,043 614 64,980 677 65,537 120 1ST QUARTILE- LOW 64,781 829 64,752 858 65,429 181 2ND QUARTILE 65,105 697 65,051 751 65,648 154 3RD QUARTILE 64,739 718 64,710 747 65,307 150 4TH QUARTILE 65,016 628 64,997 647 65,500 144 University      5TH QUARTILE - HIGH 65,249 528 65,140 637 65,650 127 1ST QUARTILE- LOW 64,367 722 64,327 762 64,951 138 2ND QUARTILE 66,213 700 66,116 797 66,768 145 3RD QUARTILE 63,831 715 63,793 753 64,408 138 4TH QUARTILE 65,583 661 65,565 679 66,062 182 Transportation      5TH QUARTILE - HIGH 64,896 602 64,849 649 65,345 153 1ST QUARTILE- LOW 63,263 756 63,211 808 63,854 165 2ND QUARTILE 64,424 691 64,389 726 64,954 161 3RD QUARTILE 67,983 707 67,930 760 68,532 158 4TH QUARTILE 63,298 652 63,273 677 63,820 130 Coefficient of variation 5TH QUARTILE - HIGH 65,922 594 65,847 669 66,374 142 1ST QUARTILE- LOW 64,885 737 64,875 747 65,444 178 2ND QUARTILE 64,956 658 64,922 692 65,449 165 3RD QUARTILE 64,729 637 64,682 684 65,208 158 4TH QUARTILE 65,466 659 65,397 728 65,989 136 Percent of owned dwellings 5TH QUARTILE - HIGH 64,854 709 64,774 789 65,444 119 1ST QUARTILE- LOW 64,889 769 64,889 769 65,474 184 2ND QUARTILE 64,927 724 64,889 762 65,492 159 3RD QUARTILE 64,963 661 64,918 706 65,474 150 4TH QUARTILE 65,075 625 64,970 730 65,568 132 Average 2000 family income ($) 5TH QUARTILE - HIGH 65,036 621 64,984 673 65,526 131 1ST QUARTILE- LOW 64,748 792 64,740 800 65,327 213 2ND QUARTILE 65,078 716 65,053 741 65,643 151 3RD QUARTILE 65,121 653 65,085 689 65,633 141 4TH QUARTILE 65,022 659 64,969 712 65,559 122 Employment rate (%) 5TH QUARTILE - HIGH 64,921 580 64,803 698 65,372 129 1ST QUARTILE- LOW 65,254 729 65,245 738 65,817 166 2ND QUARTILE 63,916 679 63,865 730 64,441 154 3RD QUARTILE 65,136 675 65,075 736 65,643 168 4TH QUARTILE 65,131 669 65,097 703 65,668 132 % people in labor working in management 5TH QUARTILE - HIGH 65,453 648 65,368 733 65,965 136 1ST QUARTILE- LOW 65,434 743 65,421 756 66,044 133 2ND QUARTILE 64,291 649 64,206 734 64,786 154 3RD QUARTILE 65,568 658 65,509 717 66,074 152 4TH QUARTILE 64,541 615 64,443 713 65,003 153 Incidence of low income in 2000 % 5TH QUARTILE - HIGH 65,056 735 65,071 720 65,627 164     111 Table 41. Distribution of health outcomes by different covariate categories (DA-SES) for road proximity ACS CCS CHF Variables Category Censored Event Censored Event Censored Event Male 158,808 2,460 158,808 2,460 158,808 2,460 Sex   Female 184,140 1,128 184,140 1,128 184,140 1,128 Born before 1925 28,614 863 28,614 863 28,614 863 Born btw 1925 - 1934 54,939 1,010 54,939 1,010 54,939 1,010 Born btw 1935 - 1944 94,937 946 94,937 946 94,937 946 Age     Born after 1944 164,458 769 164,458 769 164,458 769 1ST QUARTILE- LOW 68,625 893 68,586 932 69,327 191 2ND QUARTILE 67,978 822 67,908 892 68,619 181 3RD QUARTILE 68,812 709 68,735 786 69,364 157 4TH QUARTILE 68,886 623 68,823 686 69,368 141 Chinese minority      5TH QUARTILE - HIGH 68,647 541 68,606 582 69,064 124 1ST QUARTILE- LOW 68,591 727 68,547 771 69,144 174 2ND QUARTILE 68,505 760 68,473 792 69,089 176 3RD QUARTILE 68,599 767 68,534 832 69,187 179 4TH QUARTILE 68,611 691 68,527 775 69,169 133 Income      5TH QUARTILE - HIGH 68,642 643 68,577 708 69,153 132 1ST QUARTILE- LOW 68,097 873 68,041 929 68,780 190 2ND QUARTILE 69,456 739 69,383 812 70,037 158 3RD QUARTILE 68,007 759 67,980 786 68,607 159 4TH QUARTILE 68,506 658 68,484 680 69,014 150 University      5TH QUARTILE - HIGH 68,882 559 68,770 671 69,304 137 1ST QUARTILE- LOW 69,916 774 69,839 851 70,536 154 2ND QUARTILE 68,906 742 68,809 839 69,496 152 3RD QUARTILE 65,781 724 65,730 775 66,368 137 4TH QUARTILE 69,757 710 69,749 718 70,281 186 Transportation      5TH QUARTILE - HIGH 68,588 638 68,531 695 69,061 165 1ST QUARTILE- LOW 68,505 814 68,429 890 69,142 177 2ND QUARTILE 68,368 740 68,317 791 68,940 168 3RD QUARTILE 71,121 740 71,064 797 71,695 166 4TH QUARTILE 66,172 678 66,148 702 66,714 136 Coefficient of variation 5TH QUARTILE - HIGH 68,782 616 68,700 698 69,251 147 1ST QUARTILE- LOW 68,541 767 68,522 786 69,114 194 2ND QUARTILE 68,377 707 68,325 759 68,924 160 3RD QUARTILE 68,312 670 68,263 719 68,816 166 4TH QUARTILE 68,552 696 68,472 776 69,102 146 Percent of owned dwellings 5TH QUARTILE - HIGH 69,166 748 69,076 838 69,786 128 1ST QUARTILE- LOW 68,494 824 68,472 846 69,126 192 2ND QUARTILE 68,506 757 68,462 801 69,096 167 3RD QUARTILE 68,596 702 68,547 751 69,138 160 4TH QUARTILE 68,732 657 68,609 780 69,253 136 Average 2000 family income ($) 5TH QUARTILE - HIGH 68,620 648 68,568 700 69,129 139 1ST QUARTILE- LOW 68,382 835 68,361 856 68,997 220 2ND QUARTILE 68,350 754 68,313 791 68,948 156 3RD QUARTILE 68,922 688 68,881 729 69,463 147 4TH QUARTILE 68,522 700 68,462 760 69,090 132 Employment rate (%) 5TH QUARTILE - HIGH 68,772 611 68,641 742 69,244 139 1ST QUARTILE- LOW 68,709 775 68,678 806 69,309 175 2ND QUARTILE 67,744 719 67,679 784 68,303 160 3RD QUARTILE 68,161 694 68,094 761 68,680 175 4TH QUARTILE 69,933 721 69,887 767 70,511 143 % people in labor working in management 5TH QUARTILE - HIGH 68,401 679 68,320 760 68,939 141 1ST QUARTILE- LOW 68,448 770 68,426 792 69,073 145 2ND QUARTILE 68,721 707 68,613 815 69,273 155 3RD QUARTILE 68,698 677 68,628 747 69,210 165 4TH QUARTILE 68,442 656 68,343 755 68,936 162 Incidence of low income in 2000 % 5TH QUARTILE - HIGH 68,639 778 68,648 769 69,250 167     112 Table 42. Distribution of health outcomes by different covariate categories (Neighborhood-SES) for black carbon traffic pollution ACS CCS CHF Variables Category Censored Event Censored Event Censored Event Male 158,128 2,440 157,683 2,885 160,183 385 Sex   Female 183,326 1,123 183,479 970 184,042 407 Born before 1925 28,513 860 28,740 633 28,977 396 Born btw 1925 - 1934 54,740 999 54,635 1,104 55,500 239 Born btw 1935 - 1944 94,475 940 94,199 1,216 95,313 102 Age     Born after 1944 163,726 764 163,588 902 164,435 55 1ST QUARTILE- LOW 65,030 847 64,974 903 65,712 165 2ND QUARTILE 70,988 752 70,930 810 71,554 186 3RD QUARTILE 67,971 682 67,904 749 68,492 161 4TH QUARTILE 69,249 654 69,144 759 69,769 134 % of total population whose home language is neither English nor French (OTHLANG) 5TH QUARTILE - HIGH 68,216 628 68,210 634 68,698 146 1ST QUARTILE- LOW 64,698 790 64,641 847 65,335 153 2ND QUARTILE 70,147 802 70,087 862 70,758 191 3RD QUARTILE 68,304 687 68,229 762 68,821 170 4TH QUARTILE 74,924 712 74,809 827 75,489 147 % of total population without knowledge of English or French (LINGISOL) 5TH QUARTILE - HIGH 63,381 572 63,396 557 63,822 131 1ST QUARTILE- LOW 68,075 894 67,981 988 68,780 189 2ND QUARTILE 69,090 713 69,051 752 69,650 153 3RD QUARTILE 68,164 702 68,125 741 68,723 143 4TH QUARTILE 66,934 654 66,919 669 67,448 140 % of total population (>=20 years of age) with any university degree (UNIVERSITY) 5TH QUARTILE - HIGH 69,191 600 69,086 705 69,624 167 1ST QUARTILE- LOW 67,041 779 66,973 847 67,666 154 2ND QUARTILE 67,954 687 67,873 768 68,489 152 3RD QUARTILE 71,847 717 71,797 767 72,400 164 4TH QUARTILE 66,343 671 66,243 771 66,844 170 Seasonally adjusted unemployment rate among persons aged 25 years and over (UNEMPLOYMENT) 5TH QUARTILE - HIGH 68,269 709 68,276 702 68,826 152 1ST QUARTILE- LOW 68,325 687 68,348 664 68,854 158 2ND QUARTILE 68,542 717 68,457 802 69,099 160 3RD QUARTILE 67,798 785 67,677 906 68,394 189 4TH QUARTILE 68,871 680 68,803 748 69,416 135 Median annual family income ($) (FAM_INCOME) 5TH QUARTILE - HIGH 67,918 694 67,877 735 68,462 150 1ST QUARTILE- LOW 68,023 695 68,021 697 68,553 165 2ND QUARTILE 68,665 703 68,598 770 69,226 142 3RD QUARTILE 68,089 810 68,015 884 68,718 181 4TH QUARTILE 68,816 709 68,726 799 69,373 152 Average annual employment income ($) (INCOME) 5TH QUARTILE - HIGH 67,861 646 67,802 705 68,355 152 1ST QUARTILE- LOW 69,119 635 69,053 701 69,633 121 2ND QUARTILE 68,691 684 68,622 753 69,226 149 3RD QUARTILE 68,302 736 68,218 820 68,860 178 4TH QUARTILE 67,326 736 67,256 806 67,900 162 % of aggregate neighbourhood income from any government transfer (TRANSFERS) 5TH QUARTILE - HIGH 68,016 772 68,013 775 68,606 182 1ST QUARTILE- LOW 67,214 785 67,118 881 67,852 147 2ND QUARTILE 69,975 768 69,937 806 70,562 181 3RD QUARTILE 68,533 710 68,472 771 69,079 164 4TH QUARTILE 67,105 676 67,031 750 67,622 159 % of persons in households below the low-income cut-off (LICO) (LOW_INCOME) 5TH QUARTILE - HIGH 68,627 624 68,604 647 69,110 141 1ST QUARTILE- LOW 68,879 661 68,813 727 69,378 162 2ND QUARTILE 67,354 627 67,333 648 67,823 158 3RD QUARTILE 67,983 739 67,906 816 68,573 149 4TH QUARTILE 69,355 808 69,314 849 69,966 197 % of occupied dwellings that are owner-occupied (OWNED_HOMES) 5TH QUARTILE - HIGH 67,883 728 67,796 815 68,485 126 1ST QUARTILE- LOW 68,559 735 68,508 786 69,166 128 2ND QUARTILE 69,954 788 69,856 886 70,568 174 3RD QUARTILE 65,880 664 65,819 725 66,376 168 4TH QUARTILE 69,007 673 68,976 704 69,518 162 % of families spending 30% or more of income on shelter costs (STRESS) 5TH QUARTILE - HIGH 68,054 703 68,003 754 68,597 160     113 Table 43. Distribution of health outcomes by different covariate categories (Neighbourhood-SES) for NO traffic pollution ACS CCS CHF Variables Category Censored Event Censored Event Censored Event Male 158,680 2,452 158,232 2,900 160,746 386 Sex   Female 183,976 1,124 184,126 974 184,692 408 Born before 1925 28,595 861 28,820 636 29,058 398 Born btw 1925 - 1934 54,914 1,005 54,807 1,112 55,680 239 Born btw 1935 - 1944 94,843 942 94,565 1,220 95,683 102 Age     Born after 1944 164,304 768 164,166 906 165,017 55 1ST QUARTILE- LOW 65,576 852 65,512 916 66,261 167 2ND QUARTILE 71,054 752 70,996 810 71,620 186 3RD QUARTILE 68,295 686 68,229 752 68,820 161 4TH QUARTILE 68,798 645 68,690 753 69,311 132 % of total population whose home language is neither English nor French (OTHLANG) 5TH QUARTILE - HIGH 68,933 641 68,931 643 69,426 148 1ST QUARTILE- LOW 65,230 795 65,166 859 65,870 155 2ND QUARTILE 70,225 802 70,164 863 70,836 191 3RD QUARTILE 68,757 691 68,682 766 69,278 170 4TH QUARTILE 64,644 628 64,542 730 65,147 125 % of total population without knowledge of English or French (LINGISOL) 5TH QUARTILE - HIGH 73,800 660 73,804 656 74,307 153 1ST QUARTILE- LOW 68,622 899 68,521 1,000 69,330 191 2ND QUARTILE 69,133 713 69,093 753 69,693 153 3RD QUARTILE 68,232 702 68,193 741 68,791 143 4TH QUARTILE 67,280 659 67,266 673 67,799 140 % of total population (>=20 years of age) with any university degree (UNIVERSITY) 5TH QUARTILE - HIGH 69,389 603 69,285 707 69,825 167 1ST QUARTILE- LOW 67,174 782 67,106 850 67,802 154 2ND QUARTILE 68,366 691 68,286 771 68,905 152 3RD QUARTILE 71,929 718 71,878 769 72,483 164 4TH QUARTILE 66,742 675 66,636 781 67,245 172 Seasonally adjusted unemployment rate among persons aged 25 years and over (UNEMPLOYMENT) 5TH QUARTILE - HIGH 68,445 710 68,452 703 69,003 152 1ST QUARTILE- LOW 68,723 690 68,740 673 69,253 160 2ND QUARTILE 68,594 718 68,509 803 69,152 160 3RD QUARTILE 68,161 788 68,039 910 68,760 189 4TH QUARTILE 68,159 670 68,085 744 68,695 134 Median annual family income ($) (FAM_INCOME) 5TH QUARTILE - HIGH 69,019 710 68,985 744 69,578 151 1ST QUARTILE- LOW 68,409 698 68,401 706 68,940 167 2ND QUARTILE 68,751 704 68,684 771 69,313 142 3RD QUARTILE 68,280 812 68,204 888 68,911 181 4TH QUARTILE 69,291 713 69,201 803 69,852 152 Average annual employment income ($) (INCOME) 5TH QUARTILE - HIGH 67,925 649 67,868 706 68,422 152 1ST QUARTILE- LOW 69,324 637 69,258 703 69,840 121 2ND QUARTILE 69,032 689 68,965 756 69,572 149 3RD QUARTILE 68,391 736 68,306 821 68,949 178 4TH QUARTILE 67,478 739 67,407 810 68,055 162 % of aggregate neighbourhood income from any government transfer (TRANSFERS) 5TH QUARTILE - HIGH 68,431 775 68,422 784 69,022 184 1ST QUARTILE- LOW 67,306 785 67,210 881 67,944 147 2ND QUARTILE 70,755 778 70,711 822 71,350 183 3RD QUARTILE 68,621 711 68,560 772 69,168 164 4TH QUARTILE 67,144 677 67,070 751 67,662 159 % of persons in households below the low-income cut-off (LICO) (LOW_INCOME) 5TH QUARTILE - HIGH 68,830 625 68,807 648 69,314 141 1ST QUARTILE- LOW 69,067 662 69,001 728 69,567 162 2ND QUARTILE 67,392 628 67,371 649 67,862 158 3RD QUARTILE 69,915 761 69,829 847 70,524 152 4TH QUARTILE 68,324 797 68,286 835 68,925 196 % of occupied dwellings that are owner-occupied (OWNED_HOMES) 5TH QUARTILE - HIGH 67,958 728 67,871 815 68,560 126 1ST QUARTILE- LOW 68,993 741 68,942 792 69,606 128 2ND QUARTILE 70,418 792 70,315 895 71,034 176 3RD QUARTILE 65,930 665 65,868 727 66,427 168 4TH QUARTILE 69,048 674 69,017 705 69,560 162 % of families spending 30% or more of income on shelter costs (STRESS) 5TH QUARTILE - HIGH 68,267 704 68,216 755 68,811 160     114 Table 44. Distribution of health outcomes by different covariate categories (Neighborhood-SES) for NO2 traffic pollution ACS CCS CHF Variables Category Censored Event Censored Event Censored Event Male 158,667 2,451 158,218 2,900 160,732 386 Sex   Female 183,968 1,124 184,118 974 184,684 408 Born before 1925 28,594 860 28,818 636 29,056 398 Born btw 1925 - 1934 54,912 1,005 54,805 1,112 55,678 239 Born btw 1935 - 1944 94,836 942 94,558 1,220 95,676 102 Age     Born after 1944 164,293 768 164,155 906 165,006 55 1ST QUARTILE- LOW 65,563 852 65,499 916 66,248 167 2ND QUARTILE 71,047 751 70,988 810 71,612 186 3RD QUARTILE 68,295 686 68,229 752 68,820 161 4TH QUARTILE 69,446 657 69,342 761 69,969 134 % of total population whose home language is neither English nor French (OTHLANG) 5TH QUARTILE - HIGH 68,284 629 68,278 635 68,767 146 1ST QUARTILE- LOW 65,217 795 65,153 859 65,857 155 2ND QUARTILE 70,218 801 70,156 863 70,828 191 3RD QUARTILE 68,757 691 68,682 766 69,278 170 4TH QUARTILE 64,644 628 64,542 730 65,147 125 % of total population without knowledge of English or French (LINGISOL) 5TH QUARTILE - HIGH 73,799 660 73,803 656 74,306 153 1ST QUARTILE- LOW 68,619 899 68,518 1,000 69,327 191 2ND QUARTILE 69,128 713 69,088 753 69,688 153 3RD QUARTILE 68,232 702 68,193 741 68,791 143 4TH QUARTILE 67,279 659 67,265 673 67,798 140 % of total population (>=20 years of age) with any university degree (UNIVERSITY) 5TH QUARTILE - HIGH 69,377 602 69,272 707 69,812 167 1ST QUARTILE- LOW 67,167 782 67,099 850 67,795 154 2ND QUARTILE 68,355 690 68,274 771 68,893 152 3RD QUARTILE 71,927 718 71,876 769 72,481 164 4TH QUARTILE 66,742 675 66,636 781 67,245 172 Seasonally adjusted unemployment rate among persons aged 25 years and over (UNEMPLOYMENT) 5TH QUARTILE - HIGH 68,444 710 68,451 703 69,002 152 1ST QUARTILE- LOW 68,723 690 68,740 673 69,253 160 2ND QUARTILE 68,593 718 68,508 803 69,151 160 3RD QUARTILE 68,159 788 68,037 910 68,758 189 4TH QUARTILE 68,153 670 68,079 744 68,689 134 Median annual family income ($) (FAM_INCOME) 5TH QUARTILE - HIGH 69,007 709 68,972 744 69,565 151 1ST QUARTILE- LOW 68,409 698 68,401 706 68,940 167 2ND QUARTILE 68,750 704 68,683 771 69,312 142 3RD QUARTILE 68,277 812 68,201 888 68,908 181 4TH QUARTILE 69,286 713 69,196 803 69,847 152 Average annual employment income ($) (INCOME) 5TH QUARTILE - HIGH 67,913 648 67,855 706 68,409 152 1ST QUARTILE- LOW 69,307 637 69,241 703 69,823 121 2ND QUARTILE 69,031 688 68,963 756 69,570 149 3RD QUARTILE 68,391 736 68,306 821 68,949 178 4TH QUARTILE 67,475 739 67,404 810 68,052 162 % of aggregate neighbourhood income from any government transfer (TRANSFERS) 5TH QUARTILE - HIGH 68,431 775 68,422 784 69,022 184 1ST QUARTILE- LOW 67,289 785 67,193 881 67,927 147 2ND QUARTILE 70,754 778 70,710 822 71,349 183 3RD QUARTILE 68,619 710 68,557 772 69,165 164 4TH QUARTILE 67,143 677 67,069 751 67,661 159 % of persons in households below the low-income cut-off (LICO) (LOW_INCOME) 5TH QUARTILE - HIGH 68,830 625 68,807 648 69,314 141 1ST QUARTILE- LOW 69,067 662 69,001 728 69,567 162 2ND QUARTILE 67,391 627 67,369 649 67,860 158 3RD QUARTILE 69,915 761 69,829 847 70,524 152 4TH QUARTILE 68,310 797 68,272 835 68,911 196 % of occupied dwellings that are owner-occupied (OWNED_HOMES) 5TH QUARTILE - HIGH 67,952 728 67,865 815 68,554 126 1ST QUARTILE- LOW 68,981 741 68,930 792 69,594 128 2ND QUARTILE 70,409 792 70,306 895 71,025 176 3RD QUARTILE 65,930 665 65,868 727 66,427 168 4TH QUARTILE 69,048 673 69,016 705 69,559 162 % of families spending 30% or more of income on shelter costs (STRESS) 5TH QUARTILE - HIGH 68,267 704 68,216 755 68,811 160     115 Table 45. Distribution of health outcomes by different covariate categories (Neighborhood-SES) for PM2.5 traffic pollution ACS CCS CHF Variables Category Censored Event Censored Event Censored Event Male 149,977 2,317 149,575 2,719 151,924 370 Sex   Female 174,918 1,078 175,075 921 175,610 386 Born before 1925 27,469 822 27,687 604 27,912 379 Born btw 1925 - 1934 52,188 955 52,103 1,040 52,916 227 Born btw 1935 - 1944 89,562 899 89,305 1,156 90,362 99 Age     Born after 1944 155,676 719 155,555 840 156,344 51 1ST QUARTILE- LOW 68,524 844 68,476 892 69,197 171 2ND QUARTILE 58,702 669 58,678 693 59,203 168 3RD QUARTILE 68,025 666 67,926 765 68,543 148 4TH QUARTILE 63,882 619 63,811 690 64,372 129 % of total population whose home language is neither English nor French (OTHLANG) 5TH QUARTILE - HIGH 65,762 597 65,759 600 66,219 140 1ST QUARTILE- LOW 63,624 789 63,578 835 64,257 156 2ND QUARTILE 66,862 748 66,791 819 67,425 185 3RD QUARTILE 64,984 641 64,916 709 65,472 153 4TH QUARTILE 66,719 651 66,641 729 67,239 131 % of total population without knowledge of English or French (LINGISOL) 5TH QUARTILE - HIGH 62,706 566 62,724 548 63,141 131 1ST QUARTILE- LOW 64,974 837 64,905 906 65,632 179 2ND QUARTILE 67,917 697 67,883 731 68,459 155 3RD QUARTILE 61,167 647 61,139 675 61,687 127 4TH QUARTILE 65,510 647 65,497 660 66,019 138 % of total population (>=20 years of age) with any university degree (UNIVERSITY) 5TH QUARTILE - HIGH 65,327 567 65,226 668 65,737 157 1ST QUARTILE- LOW 65,525 776 65,452 849 66,150 151 2ND QUARTILE 69,088 684 69,030 742 69,626 146 3RD QUARTILE 59,574 603 59,538 639 60,032 145 4TH QUARTILE 65,305 661 65,207 759 65,797 169 Seasonally adjusted unemployment rate among persons aged 25 years and over (UNEMPLOYMENT) 5TH QUARTILE - HIGH 65,403 671 65,423 651 65,929 145 1ST QUARTILE- LOW 65,369 651 65,407 613 65,869 151 2ND QUARTILE 64,604 667 64,522 749 65,115 156 3RD QUARTILE 65,005 753 64,898 860 65,578 180 4TH QUARTILE 65,769 652 65,704 717 66,292 129 Median annual family income ($) (FAM_INCOME) 5TH QUARTILE - HIGH 64,148 672 64,119 701 64,680 140 1ST QUARTILE- LOW 64,356 647 64,368 635 64,844 159 2ND QUARTILE 66,252 678 66,196 734 66,797 133 3RD QUARTILE 64,531 769 64,468 832 65,127 173 4TH QUARTILE 65,961 687 65,876 772 66,498 150 Average annual employment income ($) (INCOME) 5TH QUARTILE - HIGH 63,795 614 63,742 667 64,268 141 1ST QUARTILE- LOW 64,883 604 64,820 667 65,376 111 2ND QUARTILE 65,644 661 65,585 720 66,162 143 3RD QUARTILE 65,374 706 65,300 780 65,908 172 4TH QUARTILE 63,876 696 63,819 753 64,416 156 % of aggregate neighbourhood income from any government transfer (TRANSFERS) 5TH QUARTILE - HIGH 65,118 728 65,126 720 65,672 174 1ST QUARTILE- LOW 65,056 797 64,996 857 65,712 141 2ND QUARTILE 68,018 716 67,965 769 68,549 185 3RD QUARTILE 63,329 670 63,267 732 63,852 147 4TH QUARTILE 64,160 629 64,110 679 64,640 149 % of persons in households below the low-income cut-off (LICO) (LOW_INCOME) 5TH QUARTILE - HIGH 64,332 583 64,312 603 64,781 134 1ST QUARTILE- LOW 64,918 618 64,869 667 65,384 152 2ND QUARTILE 64,902 608 64,885 625 65,354 156 3RD QUARTILE 65,153 713 65,088 778 65,728 138 4TH QUARTILE 64,843 720 64,802 761 65,379 184 % of occupied dwellings that are owner-occupied (OWNED_HOMES) 5TH QUARTILE - HIGH 65,079 736 65,006 809 65,689 126 1ST QUARTILE- LOW 65,453 709 65,403 759 66,044 118 2ND QUARTILE 63,936 728 63,860 804 64,502 162 3RD QUARTILE 64,775 652 64,718 709 65,258 169 4TH QUARTILE 65,721 639 65,698 662 66,205 155 % of families spending 30% or more of income on shelter costs (STRESS) 5TH QUARTILE - HIGH 65,010 667 64,971 706 65,525 152     116 Table 46. Distribution of health outcomes by different covariate categories (Neighborhood-SES) for different road proximity categories ACS CCS CHF Variables Category Censored Event Censored Event Censored Event Male 158,811 2,457 158,365 2,903 160,882 386 Sex   Female 184,142 1,126 184,293 975 184,860 408 Born before 1925 28,616 861 28,841 636 29,079 398 Born btw 1925 - 1934 54,942 1,007 54,835 1,114 55,710 239 Born btw 1935 - 1944 94,937 946 94,661 1,222 95,781 102 Age     Born after 1944 164,458 769 164,321 906 165,172 55 1ST QUARTILE- LOW 65,642 853 65,578 917 66,328 167 2ND QUARTILE 71,122 755 71,065 812 71,691 186 3RD QUARTILE 68,364 687 68,299 752 68,890 161 4TH QUARTILE 68,850 646 68,743 753 69,364 132 % of total population whose home language is neither English nor French (OTHLANG) 5TH QUARTILE - HIGH 68,975 642 68,973 644 69,469 148 1ST QUARTILE- LOW 65,296 796 65,232 860 65,937 155 2ND QUARTILE 70,302 804 70,242 864 70,915 191 3RD QUARTILE 68,818 692 68,743 767 69,340 170 4TH QUARTILE 64,689 630 64,589 730 65,194 125 % of total population without knowledge of English or French (LINGISOL) 5TH QUARTILE - HIGH 73,848 661 73,852 657 74,356 153 1ST QUARTILE- LOW 68,704 902 68,603 1,003 69,415 191 2ND QUARTILE 69,192 715 69,153 754 69,754 153 3RD QUARTILE 68,280 703 68,242 741 68,840 143 4TH QUARTILE 67,321 660 67,308 673 67,841 140 % of total population (>=20 years of age) with any university degree (UNIVERSITY) 5TH QUARTILE - HIGH 69,456 603 69,352 707 69,892 167 1ST QUARTILE- LOW 67,234 783 67,166 851 67,863 154 2ND QUARTILE 68,419 691 68,339 771 68,958 152 3RD QUARTILE 71,976 721 71,927 770 72,533 164 4TH QUARTILE 66,803 676 66,698 781 67,307 172 Seasonally adjusted unemployment rate among persons aged 25 years and over (UNEMPLOYMENT) 5TH QUARTILE - HIGH 68,521 712 68,528 705 69,081 152 1ST QUARTILE- LOW 68,796 691 68,813 674 69,327 160 2ND QUARTILE 68,644 720 68,560 804 69,204 160 3RD QUARTILE 68,220 791 68,100 911 68,822 189 4TH QUARTILE 68,220 670 68,146 744 68,756 134 Median annual family income ($) (FAM_INCOME) 5TH QUARTILE - HIGH 69,073 711 69,039 745 69,633 151 1ST QUARTILE- LOW 68,480 700 68,472 708 69,013 167 2ND QUARTILE 68,803 706 68,738 771 69,367 142 3RD QUARTILE 68,339 814 68,264 889 68,972 181 4TH QUARTILE 69,343 714 69,253 804 69,905 152 Average annual employment income ($) (INCOME) 5TH QUARTILE - HIGH 67,988 649 67,931 706 68,485 152 1ST QUARTILE- LOW 69,390 637 69,324 703 69,906 121 2ND QUARTILE 69,079 690 69,012 757 69,620 149 3RD QUARTILE 68,443 739 68,360 822 69,004 178 4TH QUARTILE 67,536 740 67,466 810 68,114 162 % of aggregate neighbourhood income from any government transfer (TRANSFERS) 5TH QUARTILE - HIGH 68,505 777 68,496 786 69,098 184 1ST QUARTILE- LOW 67,362 786 67,266 882 68,001 147 2ND QUARTILE 70,818 778 70,774 822 71,413 183 3RD QUARTILE 68,675 714 68,616 773 69,225 164 4TH QUARTILE 67,199 679 67,126 752 67,719 159 % of persons in households below the low-income cut-off (LICO) (LOW_INCOME) 5TH QUARTILE - HIGH 68,899 626 68,876 649 69,384 141 1ST QUARTILE- LOW 69,151 663 69,085 729 69,652 162 2ND QUARTILE 67,425 630 67,405 650 67,897 158 3RD QUARTILE 69,986 763 69,901 848 70,597 152 4TH QUARTILE 68,375 798 68,338 835 68,977 196 % of occupied dwellings that are owner-occupied (OWNED_HOMES) 5TH QUARTILE - HIGH 68,016 729 67,929 816 68,619 126 1ST QUARTILE- LOW 69,041 742 68,990 793 69,655 128 2ND QUARTILE 70,480 793 70,378 895 71,097 176 3RD QUARTILE 65,984 666 65,923 727 66,482 168 4TH QUARTILE 69,091 676 69,061 706 69,605 162 % of families spending 30% or more of income on shelter costs (STRESS) 5TH QUARTILE - HIGH 68,357 706 68,306 757 68,903 160     117 Table 47. Distribution of subjects by the two levels of socioeconomic indicators at DA-level of aggregation and the four quartiles of traffic pollutants Pollutant Quartile DA-level SES Pollutant SES Level 1st Q 2nd Q 3rd Q 4th Q Low 24,800 20,312 14,507 9,820 NO High 7,076 14,302 20,995 26,781 Low 30,512 19,613 9,553 9,757 NO2 High 4,892 12,349 29,569 22,344 Low 18,705 19,827 13,979 16,499 Black Carbon High 16,990 16,934 16,916 18,139 Low 19,060 18,026 16,309 12,049 Chinese population PM2.5 High 8,185 14,341 20,311 22,831 Low 22,929 20,852 14,713 10,405 NO High 14,376 15,429 18,889 20,684 Low 26,491 18,997 10,676 12,733 NO2 High 11,941 16,196 21,696 19,535 Low 15,016 19,890 16,677 17,874 Black Carbon High 19,198 14,683 17,975 17,308 Low 13,315 17,557 19,698 15,017 University PM2.5 High 18,137 11,973 16,270 19,379 Low 12,313 17,350 18,680 20,991 NO High 22,399 16,259 15,620 15,037 Low 11,657 14,403 20,722 22,551 NO2 High 23,483 17,538 13,179 15,115 Low 14,096 16,605 18,001 20,371 Black Carbon High 19,088 18,980 15,683 15,203 Low 12,119 16,590 17,804 19,017 Employment PM2.5 High 19,588 16,174 15,068 14,652 Low 11,894 15,052 19,128 23,162 NO High 20,487 18,140 17,551 13,034 Low 10,932 12,717 19,363 26,222 NO2 High 20,909 21,722 16,661 9,908 Low 10,360 14,568 20,397 23,688 Black Carbon High 22,419 18,995 13,703 13,877 Low 8,638 15,859 20,142 21,003 Family income PM2.5 High 26,168 14,126 14,399 10,940 Low 11,003 15,470 19,396 23,384 NO High 20,244 16,891 17,454 14,640 Low 10,039 13,127 21,436 24,650 NO2 High 19,869 20,497 16,876 11,975 Low 11,941 16,415 19,023 21,659 Black Carbon High 21,452 18,122 14,443 15,007 Low 7,750 15,720 20,782 21,344 Personal income PM2.5 High 25,307 14,207 13,952 12,168        118 Table 47. Distribution of subjects by the two levels of socioeconomic indicators at DA-level of aggregation and the four quartiles of traffic pollutants (cont.) Pollutant Quartile DA-level SES Pollutant SES Level 1st Q 2nd Q 3rd Q 4th Q Low 26,667 18,258 13,595 12,119 NO High 5,181 13,264 20,570 30,126 Low 33,665 18,996 10,167 7,798 NO2 High 2,995 8,829 21,926 35,389 Low 23,305 21,009 11,602 14,142 Black Carbon High 7,945 14,945 23,137 23,043 Low 23,341 15,508 14,169 12,037 Transportation PM2.5 High 8,160 15,239 18,413 23,683 Low 25,434 18,210 14,694 10,830 NO High 9,932 15,040 18,836 25,531 Low 29,769 21,173 10,356 7,856 NO2 High 7,244 12,080 21,733 28,280 Low 23,528 20,135 11,579 13,596 Black Carbon High 10,905 14,927 21,435 21,993 Low 25,575 17,572 13,741 9,265 Low income PM2.5 High 8,767 15,348 17,617 24,055 Low 7,776 12,830 19,075 29,542 NO High 24,139 18,834 14,690 12,200 Low 6,027 9,886 19,751 33,556 NO2 High 27,787 21,778 12,505 7,788 Low 8,358 13,236 22,913 24,525 Black Carbon High 23,992 20,654 12,563 12,420 Low 8,969 14,563 17,549 24,535 Home ownership PM2.5 High 24,481 16,556 13,489 11,009 Low 17,514 16,734 15,276 19,718 NO High 16,777 18,227 17,674 16,660 Low 17,874 15,279 14,098 21,990 NO2 High 15,447 19,423 20,989 13,475 Low 15,629 15,965 17,466 19,957 Black Carbon High 19,295 18,963 15,043 15,811 Low 14,038 15,852 14,895 19,219 Income variation PM2.5 High 18,950 16,427 16,215 14,906 Low 15,335 19,099 17,613 17,369 NO High 19,508 16,314 16,830 16,370 Low 15,288 16,706 17,666 19,755 NO2 High 19,271 19,824 16,782 13,139 Low 13,873 18,152 18,642 18,346 Black Carbon High 21,383 18,244 13,932 15,329 Low 12,086 17,821 18,818 17,239 Management PM2.5 High 23,036 14,411 14,027 14,608       119 Table 48. Distribution of subjects by the two levels of socioeconomic indicators at neighborhood level of aggregation and the four quartiles of traffic pollutants Pollutant Quartile Neighborhood- level SES Pollutant SES Level 1st Q 2nd Q 3rd Q 4th Q Low 23,159 15,579 16,466 11,224 NO High 5,792 14,553 21,247 27,982 Low 29,388 18,894 8,826 9,307 NO2 High 4,945 9,697 27,986 26,285 Low 21,033 17,972 11,106 15,766 Black Carbon High 16,856 14,723 16,581 20,684 Low 20,050 17,921 14,360 17,010 Other language PM2.5 High 7,733 14,227 20,811 23,579 Low 22,496 14,807 16,287 12,435 NO High 5,031 14,287 24,499 30,643 Low 27,538 16,933 10,562 10,979 NO2 High 4,571 9,947 30,852 29,089 Low 20,965 15,807 12,709 16,007 Black Carbon High 13,187 16,229 15,840 18,697 Low 20,905 16,441 12,517 14,518 Linguistic isolation PM2.5 High 7,842 14,209 20,093 21,118 Low 26,931 23,150 12,104 7,336 NO High 14,117 14,553 19,563 21,759 Low 34,032 20,006 6,222 9,258 NO2 High 10,576 15,229 21,982 22,192 Low 17,255 21,473 13,926 16,315 Black Carbon High 15,892 14,263 19,522 20,114 Low 14,398 18,701 18,642 14,034 University PM2.5 High 17,634 11,513 17,700 19,026 Low 23,153 17,056 16,592 11,155 NO High 12,706 18,298 18,852 19,299 Low 28,027 20,770 10,483 8,669 NO2 High 11,665 13,282 19,229 24,978 Low 21,234 17,754 13,809 15,023 Black Carbon High 11,182 18,302 20,018 19,476 Low 26,784 16,412 13,945 9,134 Unemployment PM2.5 High 9,853 18,085 21,123 17,002 Low 9,326 14,131 22,078 23,878 NO High 22,764 18,468 17,432 11,065 Low 9,315 11,100 22,848 26,150 NO2 High 22,078 23,428 15,441 8,769 Low 9,056 17,412 20,834 21,710 Black Carbon High 24,397 17,779 12,238 14,198 Low 6,607 15,588 22,584 21,228 Family income PM2.5 High 28,234 13,570 13,792 9,200         120 Table 48. Distribution of subjects by the two levels of socioeconomic indicators at neighborhood level of aggregation and the four quartiles of traffic pollutants (cont.) Pollutant Quartile Neighborhood- level SES Pollutant SES Level 1st Q 2nd Q 3rd Q 4th Q Low 23,159 15,579 16,466 11,224 NO High 5,792 14,553 21,247 27,982 Low 29,388 18,894 8,826 9,307 NO2 High 4,945 9,697 27,986 26,285 Low 21,033 17,972 11,106 15,766 Black Carbon High 16,856 14,723 16,581 20,684 Low 20,050 17,921 14,360 17,010 Personal income PM2.5 High 7,733 14,227 20,811 23,579 Low 22,496 14,807 16,287 12,435 NO High 5,031 14,287 24,499 30,643 Low 27,538 16,933 10,562 10,979 NO2 High 4,571 9,947 30,852 29,089 Low 20,965 15,807 12,709 16,007 Black Carbon High 13,187 16,229 15,840 18,697 Low 20,905 16,441 12,517 14,518 Governmental transfers PM2.5 High 7,842 14,209 20,093 21,118 Low 26,931 23,150 12,104 7,336 NO High 14,117 14,553 19,563 21,759 Low 34,032 20,006 6,222 9,258 NO2 High 10,576 15,229 21,982 22,192 Low 17,255 21,473 13,926 16,315 Black Carbon High 15,892 14,263 19,522 20,114 Low 14,398 18,701 18,642 14,034 Low income PM2.5 High 17,634 11,513 17,700 19,026 Low 23,153 17,056 16,592 11,155 NO High 12,706 18,298 18,852 19,299 Low 28,027 20,770 10,483 8,669 NO2 High 11,665 13,282 19,229 24,978 Low 21,234 17,754 13,809 15,023 Black Carbon High 11,182 18,302 20,018 19,476 Low 26,784 16,412 13,945 9,134 Home ownership PM2.5 High 9,853 18,085 21,123 17,002 Low 9,326 14,131 22,078 23,878 NO High 22,764 18,468 17,432 11,065 Low 9,315 11,100 22,848 26,150 NO2 High 22,078 23,428 15,441 8,769 Low 9,056 17,412 20,834 21,710 Black Carbon High 24,397 17,779 12,238 14,198 Low 6,607 15,588 22,584 21,228 Neighborhood stress PM2.5 High 28,234 13,570 13,792 9,200         121 Table 49. Distribution of subjects by the two levels of socioeconomic indicators at DA-level of aggregation and the five road proximity categories Road proximity Road proximity DA-level SES Pollutant SES Level 0 1 DA-level SES Pollutant SES Level 0 1 Low 67,845 1,673 Low 69,592 1,098 Road I High 68,416 772 Road I High 67,193 2,033 Low 66,787 2,731 Low 68,158 2,532 Road II High 64,990 4,198 Road II High 63,537 5,689 Low 61,764 7,754 Low 63,568 7,122 Road III High 62,126 7,062 Road III High 60,157 9,069 Low 56,558 12,960 Low 58,937 11,753 Road IV High 52,072 17,116 Road IV High 47,673 21,553 Low 57,717 11,801 Low 60,203 10,487 Chinese population Road V High 57,360 11,828 Transportation Road V High 52,771 16,455 Low 66,941 2,029 Low 68,259 959 Road I High 68,715 726 Road I High 67,377 2,040 Low 64,779 4,191 Low 66,791 2,427 Road II High 66,034 3,407 Road II High 64,272 5,145 Low 61,397 7,573 Low 62,615 6,603 Road III High 62,576 6,865 Road III High 61,097 8,320 Low 56,095 12,875 Low 56,569 12,649 Road IV High 50,563 18,878 Road IV High 51,256 18,161 Low 55,560 13,410 Low 59,523 9,695 University Road V High 58,654 10,787 Low income Road V High 54,228 15,189 Low 67,524 1,693 Low 67,172 2,136 Road I High 68,518 865 Road I High 68,939 975 Low 64,921 4,296 Low 63,715 5,593 Road II High 66,476 2,907 Road II High 67,425 2,489 Low 60,513 8,704 Low 59,824 9,484 Road III High 62,792 6,591 Road III High 63,717 6,197 Low 52,238 16,979 Low 48,301 21,007 Road IV High 55,210 14,173 Road IV High 58,395 11,519 Low 54,714 14,503 Low 52,474 16,834 Employment Road V High 59,271 10,112 Home ownership Road V High 60,346 9,568 Low 67,000 2,318 Low 67,635 1,684 Road I High 68,639 629 Road I High 68,741 657 Low 63,670 5,648 Low 65,132 4,187 Road II High 66,653 2,615 Road II High 66,457 2,941 Low 60,833 8,485 Low 61,189 8,130 Road III High 63,009 6,259 Road III High 61,907 7,491 Low 52,228 17,090 Low 55,814 13,505 Road IV High 54,210 15,058 Road IV High 53,287 16,111 Low 53,146 16,172 Low 55,603 13,716 Family income Road V High 59,932 9,336 Income variation Road V High 58,564 10,834 Low 67,260 2,058 Low 67,634 1,850 Road I High 68,502 783 Road I High 68,118 962 Low 64,126 5,192 Low 64,993 4,491 Road II High 66,387 2,898 Road II High 65,493 3,587 Low 61,321 7,997 Low 61,823 7,661 Road III High 63,032 6,253 Road III High 62,586 6,494 Low 52,434 16,884 Low 53,330 16,154 Road IV High 53,259 16,026 Road IV High 54,067 15,013 Low 54,333 14,985 Low 55,757 13,727 Personal income Road V High 59,537 9,748 Management Road V High 58,294 10,786      122 Table 50. Distribution of subjects by the two levels of socioeconomic indicators at neighborhood-level of aggregation and the five road proximity categories Road proximity Road proximity DA-level SES Pollutant SES Level 0 1 DA-level SES Polluta nt SES Level 0 1 Low 64,958 1,537 Low 67,290 1,890 Road I High 68,412 1,205 Road I High 67,732 905 Low 63,816 2,679 Low 64,438 4,742 Road II High 65,521 4,096 Road II High 65,189 3,448 Low 59,011 7,484 Low 61,949 7,231 Road III High 62,199 7,418 Road III High 61,533 7,104 Low 53,753 12,742 Low 52,777 16,403 Road IV High 52,139 17,478 Road IV High 50,636 18,001 Low 55,089 11,406 Low 55,512 13,668 Other language Road V High 57,065 12,552 Personal income Road V High 57,491 11,146 Low 64,652 1,440 Low 69,455 572 Road I High 73,369 1,140 Road I High 67,337 1,945 Low 63,586 2,506 Low 67,791 2,236 Road II High 70,196 4,313 Road II High 64,100 5,182 Low 58,876 7,216 Low 63,963 6,064 Road III High 66,278 8,231 Road III High 60,626 8,656 Low 52,164 13,928 Low 54,741 15,286 Road IV High 54,920 19,589 Road IV High 52,274 17,008 Low 55,223 10,869 Low 61,284 8,743 Linguistic isolation Road V High 61,016 13,493 Governmental transfers Road V High 53,906 15,376 Low 67,778 1,828 Low 66,804 1,344 Road I High 69,055 1,004 Road I High 67,962 1,563 Low 66,705 2,901 Low 65,830 2,318 Road II High 65,927 4,132 Road II High 64,534 4,991 Low 61,644 7,962 Low 60,949 7,199 Road III High 62,323 7,736 Road III High 60,951 8,574 Low 58,222 11,384 Low 56,402 11,746 Road IV High 49,437 20,622 Road IV High 50,479 19,046 Low 57,220 12,386 Low 57,555 10,593 University Road V High 57,508 12,551 Low income Road V High 54,777 14,748 Low 66,946 1,071 Low 67,920 1,894 Road I High 67,238 1,995 Road I High 67,520 1,225 Low 65,426 2,591 Low 64,251 5,563 Road II High 64,617 4,616 Road II High 66,824 1,921 Low 60,610 7,407 Low 61,169 8,645 Road III High 61,352 7,881 Road III High 62,286 6,459 Low 55,273 12,744 Low 50,309 19,505 Road IV High 51,863 17,370 Road IV High 57,595 11,150 Low 57,114 10,903 Low 54,038 15,776 Unemployment Road V High 55,012 14,221 Home ownership Road V High 59,217 9,528 Low 67,726 1,761 Low 69,185 598 Road I High 68,930 854 Road I High 67,157 1,906 Low 64,253 5,234 Low 67,466 2,317 Road II High 67,033 2,751 Road II High 64,102 4,961 Low 61,477 8,010 Low 63,156 6,627 Road III High 63,241 6,543 Road III High 60,476 8,587 Low 52,189 17,298 Low 57,089 12,694 Road IV High 54,948 14,836 Road IV High 51,853 17,210 Low 54,746 14,741 Low 60,337 9,446 Family income Road V High 59,760 10,024 Neighborhood stress Road V High 53,918 15,145   123 Appendix II: Dissemination area level covariates in conjunction with traffic related pollutants ACS health outcomes Chinese - NO 0.80 0.90 1.00 1.10 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Chinese - NO2 0.80 0.90 1.00 1.10 1 2 3 4 Pollution Quartiles H R s Low High  Chinese - B.C. 0.80 0.90 1.00 1.10 1.20 1 2 3 4 Pollution Quartiles H R s Low High Chinese - PM 2.5 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  University - NO 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High University - NO 2 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  University - BC 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High University -  PM 2.5 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Em ployment - NO 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High Employment - NO 2 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Employment - BC 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High Employment - PM 2.5 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High     124 Personal Income - NO 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High Personal Income - NO 2 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Personal Income - B.C. 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High Personal Income - PM 2.5 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Transportation - NO 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High Transportation - NO 2 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Transportation - B.C. 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High Transportation - PM2.5 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Low Income - NO 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High Low  Income - NO 2 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Low Income - B.C. 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High Low  Income - PM 2.5 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High        125 Hom e ow nership - NO 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High Hom e owne rship - NO 2 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Hom e ow nership - B.C. 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High Home ow nership - PM 2.5 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Income Variation - NO 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High Income Variation - NO 2 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Income Variation - B.C. 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High Income Variation - PM 2.5 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Managem ent - NO 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High Management - NO 2 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Management - B.C. 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High Managem ent - PM 2.5 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High         126 Family Income - NO 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High Family Income - NO 2 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Family Income - B.C. 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High Family Income -  PM2.5 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High   CCS health outcomes Chinese - NO 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Chinese - NO2 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Chinese - B.C. 0.80 0.90 1.00 1.10 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Chinese - PM2.5 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  University - NO 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  University - NO2 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  University - BC 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  University - PM2.5 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High     127 Em ployment - NO 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Employment - NO2 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Employment - BC 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Employment - PM2.5 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Family Income - NO 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Family Income - NO2 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Family Income - B.C. 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Family Income - PM2.5 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Personal Income - NO 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Personal Income - NO2 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Personal Income - B.C. 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Personal Income - PM2.5 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High         128 Transportation - NO 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Transportation - NO2 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Transportation - B.C. 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Transportation - PM2.5 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Low Income - NO 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Low Income - NO2 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Low Income - B.C. 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Low  Incom e - PM2.5 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Hom e ow nership - NO 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Home ow nership - NO2 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Home ow nership - B.C. 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Hom e ow nership - PM2.5 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High        129 Income Variation - NO 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Income Variation - NO2 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Income Variation - B.C. 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Income Variation - PM2.5 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Managem ent - NO 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Management - NO2 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Management - B.C. 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Management - PM2.5 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High   CHF health outcomes  Chinese - NO 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Chinese - NO2 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Chinese - B.C. 0.80 1.00 1.20 1.40 1.60 1.80 1 2 3 4 Pollution Quartiles H R s Low High  Chinese - PM2.5 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High       130  University - NO 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  University - NO2 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  University - BC 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00 1 2 3 4 Pollution Quartiles H R s Low High  University - PM2.5 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Em ployment - NO 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Employment - NO2 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Employment - BC 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Employment - PM2.5 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Family Income - NO 0.40 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Family Income - NO2 0.40 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Family Income - B.C. 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00 1 2 3 4 Pollution Quartiles H R s Low High  Family Income - PM2.5 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High        131 Personal Income - NO 0.40 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Personal Incom e - NO2 0.40 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Personal Incom e - B.C. 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00 1 2 3 4 Pollution Quartiles H R s Low High  Personal Income - PM2.5 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Transportation - NO 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Transportation - NO2 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Transportation - B.C. 0.60 0.80 1.00 1.20 1.40 1.60 1.80 1 2 3 4 Pollution Quartiles H R s Low High  Transportation - PM2.5 0.60 0.80 1.00 1.20 1.40 1.60 1 2 3 4 Pollution Quartiles H R s Low High  Low  Income - NO 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Low Income - NO2 0.40 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Low Income - B.C. 0.60 1.00 1.40 1.80 2.20 2.60 3.00 1 2 3 4 Pollution Quartiles H R s Low High  Low Income - PM2.5 0.60 0.80 1.00 1.20 1.40 1.60 1 2 3 4 Pollution Quartiles H R s Low High          132 Hom e ow nership - NO 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Home ow nership - NO2 0.40 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Home ow nership - B.C. 0.80 1.00 1.20 1.40 1.60 1 2 3 4 Pollution Quartiles H R s Low High  Hom e ow nership - PM2.5 0.40 0.80 1.20 1.60 1 2 3 4 Pollution Quartiles H R s Low High  Income Variation - NO 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Income Variation - NO2 0.40 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Income Variation - B.C. 0.60 1.00 1.40 1.80 1 2 3 4 Pollution Quartiles H R s Low High  Income Variation - PM2.5 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Managem ent - NO 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Management - NO2 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Management - B.C. 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Management - PM2.5 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High          133 Appendix III: Neighborhood area level covariates in conjunction with traffic related pollutants ACS health outcomes Other language - NO 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Other language - NO2 0.50 0.70 0.90 1.10 1 2 3 4 Pollution Quartiles H R s Low High  Other language - B.C. 0.40 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Other language - PM2.5 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Linguistic Isolation - NO 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Linguistic Isolation - NO2 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Linguistic Isolation - BC 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Linguistic Isolation - PM2.5 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  University - NO 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  University - NO2 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  University - BC 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  University - PM2.5 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High     134 Unemployment - NO 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Unemployment - NO2 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Unemployment - B.C. 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Unemployment - PM2.5 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Family Income - NO 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Family Income - NO2 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Family Income - B.C. 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Family Income - PM2.5 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Personal Income - NO 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Personal Income - NO2 0.40 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Personal Income - B.C. 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Personal Income - PM2.5 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High        135 Governmental Transfers - NO 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Governmental Transfers - NO2 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Governmental Transfers  - B.C. 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Governm ental Transfers - PM2.5 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Low  Income - NO 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Low  Income - NO2 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Low  Income - B.C. 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Low  Income - PM2.5 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Hom e ownership - NO 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Home ow nership - NO2 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Hom e ow nership - B.C. 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Hom e ow nership - PM2.5 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High         136 Neighborhood Stress - NO 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Neighborhood Stress - NO2 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Neighborhood Stress - B.C. 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Neighborhood Stress - PM2.5 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High   CCS health outcomes  Other language - NO 0.40 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Other language - NO2 0.50 0.70 0.90 1.10 1 2 3 4 Pollution Quartiles H R s Low High  Other language - B.C. 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Other language - PM2.5 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Linguistic Isolation - NO 0.40 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Linguistic Isolation - NO2 0.40 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Linguistic Isolation - BC 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Linguistic Isolation - PM2.5 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High      137 University - NO 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  University - NO2 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  University - BC 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  University - PM2.5 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Unemploym ent - NO 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Unemployment - NO2 0.40 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Unemployment - B.C. 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Unemployment - PM2.5 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Family Income - NO 0.40 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Family Income - NO2 0.40 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Family Income - B.C. 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Family Incom e - PM2.5 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High          138 Personal Income - NO 0.40 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Personal Income - NO2 0.40 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Personal Income - B.C. 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Personal Income - PM2.5 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Governmental Transfers - NO 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Governmental Transfers - NO2 0.40 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Governmental Transfers  - B.C. 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Governm ental Transfers - PM2.5 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Low  Income - NO 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Low  Income - NO2 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Low  Income - B.C. 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Low  Income - PM2.5 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High         139 Hom e ownership - NO 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Home ow nership - NO2 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Hom e ow nership - B.C. 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Hom e ow nership - PM2.5 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Neighborhood Stress - NO 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Neighborhood Stress - NO2 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Neighborhood Stress - B.C. 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Neighborhood Stress - PM2.5 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High   CHF health outcomes  Other language - NO 0.40 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Other language - NO2 0.50 0.70 0.90 1.10 1 2 3 4 Pollution Quartiles H R s Low High  Other language - B.C. 0.80 1.00 1.20 1.40 1.60 1 2 3 4 Pollution Quartiles H R s Low High  Other language - PM2.5 0.60 1.00 1.40 1.80 2.20 1 2 3 4 Pollution Quartiles H R s Low High     140 Linguistic Isolation - NO 0.40 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Linguistic Isolation - NO2 0.40 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Linguistic Isolation - BC 0.60 1.00 1.40 1.80 2.20 1 2 3 4 Pollution Quartiles H R s Low High  Linguistic Isolation - PM2.5 0.60 1.00 1.40 1.80 2.20 1 2 3 4 Pollution Quartiles H R s Low High  University - NO 0.40 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  University - NO2 0.40 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  University - BC 0.60 1.00 1.40 1.80 2.20 1 2 3 4 Pollution Quartiles H R s Low High  University - PM2.5 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Unemployment - NO 0.40 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Unemployment - NO2 0.40 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Unemployment - B.C. 0.60 1.00 1.40 1.80 2.20 1 2 3 4 Pollution Quartiles H R s Low High  Unemployment - PM2.5 0.60 1.00 1.40 1.80 1 2 3 4 Pollution Quartiles H R s Low High        141 Family Income - NO 0.40 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Family Income - NO2 0.40 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Family Income - B.C. 0.60 0.80 1.00 1.20 1.40 1.60 1 2 3 4 Pollution Quartiles H R s Low High  Family Income - PM2.5 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Personal Income - NO 0.40 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Personal Income - NO2 0.40 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Personal Income - B.C. 0.60 0.80 1.00 1.20 1.40 1.60 1.80 1 2 3 4 Pollution Quartiles H R s Low High  Personal Income - PM2.5 0.80 1.00 1.20 1.40 1.60 1.80 2.00 1 2 3 4 Pollution Quartiles H R s Low High  Governmental Transfers - NO 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Governmental Transfers - NO2 0.40 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Governmental Transfers  - B.C. 0.80 1.00 1.20 1.40 1.60 1.80 2.00 1 2 3 4 Pollution Quartiles H R s Low High  Governm ental Transfers - PM2.5 0.60 0.80 1.00 1.20 1.40 1.60 1.80 1 2 3 4 Pollution Quartiles H R s Low High         142 Low  Income - NO 0.60 0.80 1.00 1.20 1.40 1.60 1 2 3 4 Pollution Quartiles H R s Low High  Low  Income - NO2 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Low  Income - B.C. 0.80 1.20 1.60 2.00 1 2 3 4 Pollution Quartiles H R s Low High  Low  Income - PM2.5 0.80 1.00 1.20 1.40 1.60 1 2 3 4 Pollution Quartiles H R s Low High  Hom e ownership - NO 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Home ow nership - NO2 0.60 0.80 1.00 1.20 1.40 1 2 3 4 Pollution Quartiles H R s Low High  Hom e ow nership - B.C. 0.60 1.20 1.80 2.40 3.00 3.60 1 2 3 4 Pollution Quartiles H R s Low High  Hom e ow nership - PM2.5 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00 1 2 3 4 Pollution Quartiles H R s Low High  Neighborhood Stress - NO 0.40 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Neighborhood Stress - NO2 0.60 0.80 1.00 1.20 1 2 3 4 Pollution Quartiles H R s Low High  Neighborhood Stress - B.C. 0.60 1.00 1.40 1.80 2.20 2.60 1 2 3 4 Pollution Quartiles H R s Low High  Neighborhood Stress - PM2.5 0.80 1.00 1.20 1.40 1.60 1.80 1 2 3 4 Pollution Quartiles H R s Low High          143 Appendix IV: Dissemination area level covariates in conjunction with road proximity ACS health outcomes  ACS HRs and road proximity - by Chinese population variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 1.60 I V IV Low High  ACS HRs and road proximity - by University variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 I V IV Low High   ACS HRs and road proximity - by Employment variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 I V IV Low High  ACS HRs and road proximity - by Family Income variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 1.60 I V IV Low High   ACS HRs and road proximity - by Personal Income variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 1.60 I V IV Low High  ACS HRs and road proximity - by Transportation variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 1.60 I V IV Low High             144 ACS HRs and road proximity - by Low Income variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 I V IV Low High  ACS HRs and road proximity - by Home Ownership variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00 I V IV Low High  ACS HRs and road proximity - by Income Variation variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 I V IV Low High  ACS HRs and road proximity - by Management variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 I V IV Low High   CCS health outcomes CCS HRs and road proximity - by Chinese population variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 1.60 I V IV Low High  CCS HRs and road proximity - by University variable, low/high categories 0.80 1.00 1.20 1.40 I V IV Low High  CCS HRs and road proximity - by Employment variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 I V IV Low High  CCS HRs and road proximity - by Family Income variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 I V IV Low High       145  CCS HRs and road proximity - by Personal Income variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 I V IV Low High  CCS HRs and road proximity - by Transportation variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 I V IV Low High  CCS HRs and road proximity - by Low Income variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 I V IV Low High  CCS HRs and road proximity - by Home Ownership variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 1.60 I V IV Low High  CCS HRs and road proximity - by Income Variation variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 I V IV Low High  CCS HRs and road proximity - by Management variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 1.60 I V IV Low High   CHF health outcomes CHF HRs and road proximity - by Chinese population variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00 I V IV Low High  CHF HRs and road proximity - by University variable, low/high categories 0.40 0.50 0.60 0.70 0.80 0.90 1.00 1.10 1.20 I V IV Low High      146  CHF HRs and road proximity - by Employment variable, low/high categories 0.00 0.50 1.00 1.50 2.00 I V IV Low High  CHF HRs and road proximity - by Family Income variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 1.60 I V IV Low High  CHF HRs and road proximity - by Personal Income variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 I V IV Low High  CHF HRs and road proximity - by Transportation variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 I V IV Low High  CHF HRs and road proximity - by Low Income variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00 I V IV Low High  CHF HRs and road proximity - by Home Ownership variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 1.60 I V IV Low High  CHF HRs and road proximity - by Income Variation variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 1.60 I V IV Low High  CHF HRs and road proximity - by Management variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 1.60 I V IV Low High        147 Appendix V: Neighborhood area level covariates in conjunction with road proximity ACS health outcomes ACS HRs and road proximity - by Other Language variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 I V IV Low High  ACS HRs and road proximity - by Linguistic Isolation variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 I V IV Low High  ACS HRs and road proximity - by University variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 I V IV Low High  ACS HRs and road proximity - by Unemployment variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 I V IV Low High  ACS HRs and road proximity - by Family Income variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 1.60 I V IV Low High  ACS HRs and road proximity - by Personal Income variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 1.60 I V IV Low High  ACS HRs and road proximity - by Governmental Transfers variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 I V IV Low High  ACS HRs and road proximity - by Low Income variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 1.60 I V IV Low High      148 ACS HRs and road proximity - by Home Ownership variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 1.60 I V IV Low High  ACS HRs and road proximity - by Neighborhood Stress variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 I V IV Low High   CCS health outcomes  CCS HRs and road proximity - by Other Language variable, low/high categories 0.80 0.90 1.00 1.10 1.20 I V IV Low High  CCS HRs and road proximity - by Linguistic Isolation variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 I V IV Low High  CCS HRs and road proximity - by University variable, low/high categories 0.80 1.00 1.20 1.40 I V IV Low High  CCS HRs and road proximity - by Unemployment variable, low/high categories 0.60 0.80 1.00 1.20 1.40 I V IV Low High  CCS HRs and road proximity - by Family Income variable, low/high categories 0.60 0.80 1.00 1.20 1.40 I V IV Low High CCS HRs and road proximity - by Personal Income variable, low/high categories 0.60 0.80 1.00 1.20 1.40 I V IV Low High      149 CCS HRs and road proximity - by Governmental Transfers variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 I V IV Low High  CCS HRs and road proximity - by Low Income variable, low/high categories 0.80 1.00 1.20 1.40 I V IV Low High  CCS HRs and road proximity - by Home Ownership variable, low/high categories 0.80 1.00 1.20 1.40 1.60 I V IV Low High  CCS HRs and road proximity - by Neighborhood Stress variable, low/high categories 0.80 1.00 1.20 1.40 I V IV Low High   CHF health outcomes CHF HRs and road proximity - by Other Language variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 I V IV Low High  CHF HRs and road proximity - by Linguistic Isolation variable, low/high categories 0.40 0.90 1.40 1.90 2.40 I V IV Low High  CHF HRs and road proximity - by University variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 I V IV Low High  CHF HRs and road proximity - by Unemployment variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00 I V IV Low High        150 CHF HRs and road proximity - by Family Income variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 1.60 I V IV Low High  CHF HRs and road proximity - by Personal Income variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 1.60 I V IV Low High  CHF HRs and road proximity - by Governmental Transfers variable, low/high categories 0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 I V IV Low High  CHF HRs and road proximity - by Low Income variable, low/high categories 0.40 0.90 1.40 1.90 2.40 I V IV Low High  CHF HRs and road proximity - by Home Ownership variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 I V IV Low High  CHF HRs and road proximity - by Neighborhood Stress variable, low/high categories 0.40 0.60 0.80 1.00 1.20 1.40 1.60 I V IV Low High  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0071316/manifest

Comment

Related Items